Su Lin Lim Thesis

Malnutrition in hospitalised patients and clinical outcomes: A missed opportunity?
MALNUTRITION IN HOSPITALISED PATIENTS AND
CLINICAL OUTCOMES:
A MISSED OPPORTUNITY?
SU LIN LIM
Bachelor of Science in Dietetics (Honours)
Thesis by Publication
submitted for the
Doctor of Philosophy
in the
School of Exercise and Nutritional Sciences
Queensland University of Technology
Australia
January 2014
ABSTRACT
Introduction
Studies have shown that malnutrition is prevalent in hospitals and ranges from
13-78%. Poor nutrition leads to a range of poor clinical and functional outcomes.
Although previous studies have shown a prospective association between
malnutrition and clinical outcomes, the confounding effect of disease and its
complexity using diagnosis-related groups (DRG) has never been taken into
consideration. It is widely agreed that disease and malnutrition are closely linked
and that disease may cause secondary malnutrition and vice-versa. However it is
often argued that length of stay (LOS), mortality and hospitalisation costs are
primarily determined by the patients medical condition rather than malnutrition.
In order for malnutrition to be properly addressed, there must be a

comprehensive start-to-end system that provides continuity of care from
admission to post-discharge. This should include screening hospitalised patients
to identify those who are malnourished or at risk of malnutrition, referral of
appropriate patients for nutrition assessment, inpatient nutrition intervention and
post-discharge follow-up.
The first critical step in managing malnutrition is the identification of malnourished

patients. Despite the prevalence and consequences of malnutrition these
patients are often not identified, with up to 70% not receiving any nutrition
intervention. All patients admitted to hospital should be systematically screened,
using a nutrition screening tool that is simple, quick, reliable, valid and cost
effective. Although there are many nutrition screening tools, none have yet been
developed and validated in Singapore, since most studies have been within the
Caucasian population. With its multi-ethnic population, the applicability of existing
nutrition screening tools to Singaporean patients is uncertain, especially the
1
cutoffs used to identify risk of malnutrition. A screening tool specific for the
Singapore population is needed, and once developed should be validated.
For a nutrition screening tool to be effective, it must be completed fully and

accurately. Studies have reported screening incompletion and error rates of 28-
97%. High levels of missing data were found in commonly used nutrition
screening tools. Failure to achieve accurate and complete nutrition screening will
affect the final score allocated to a patient, which may result in a malnourished or
at risk patient not being referred for nutritional intervention.
To complete the process of nutrition screening, any patient identified at screening

to be malnourished or at risk of malnutrition should be referred to receive a full
nutrition assessment and intervention. This should be followed by a
comprehensive management of malnourished patients, including monitoring
and/or intervention. However, these patients often become lost to follow-up after
discharge. There is limited evidence on methods of follow-up post-discharge from
hospital to effectively treat malnutrition.
Aims
The aims of the research programme were to:

i) determine the prevalence of malnutrition on admission to a tertiary
hospital in Singapore and its impact on cost of hospitalisation, length of
stay, readmission and 3-year mortality.
ii) develop and validate a new nutrition screening tool for use in the
Singaporean adult population admitted to an acute hospital.
iii) confirm the reliability and validity of the new nutrition screening tool
administered by nurses in a new cohort of patients.
iv) investigate the compliance rate of nurses in conducting nutrition screening
and referring at risk patients to the dietitians; and determine the effect of
2
quality improvement initiatives in improving the overall performance of

nutrition screening.
v) explore and determine the effectiveness of the current system and an
alternative model on dietetics follow-up rate of malnourished hospital
patients post-discharge.
Methodology
The study programme was conducted in four phases:

i) a cross-sectional 3-year prospective study on 818 newly admitted
patients to determine the prevalence and outcomes of malnutrition,
adjusted for gender, age and ethnicity and matched for Diagnosis-Related
Groups (DRG). The group of patients were also screened using five
parameters that contribute to the risk of malnutrition, resulting in
development and validation of a new nutrition screening tool called 3-
Minute Nutrition Screening (3-MinNS)
ii) a cross-sectional prospective study on 121 patients to confirm the validity
and determine the reliability of 3-MinNS when used by nursing staff, the
intended users of the tool
iii) a 6-year audit and quality improvement study on 4467 patients to improve
nurses compliance with 3-MinNS
iv) an audit on dietetic follow-up of 261 malnourished patients discharged
from the hospital in which the results were used to develop a novel model
of care called Ambulatory Nutrition Support (ANS). The effectiveness of
ANS was evaluated via a prospective interventional cohort study on 163
malnourished patients discharged from the hospital.
Results
Using Subjective Global Assessment (SGA), the prevalence of malnutrition was

29%. Malnourished patients had longer hospital stays (6.9 7.3 days vs. 4.6
3
5.6 days, p < 0.001) and were more likely to be readmitted within 15 days
(adjusted relative risk = 1.9, 95%CI 1.13.2, p = 0.025). Within a DRG, the mean
difference between actual cost of hospitalisation and the average cost for
malnourished patients was three times higher than well-nourished patients (p =
0.014). Mortality was higher in malnourished patients than well-nourished
patients at 1 year (34% vs. 4.1 %), 2 years (42.6% vs. 6.7%) and 3 years (48.5%
vs. 9.9%); p < 0.001. Overall, malnutrition was a significant predictor of mortality
(adjusted hazard ratio = 4.4, 95% CI 3.3-6.0, p < 0.001).
In the development of the new nutrition screening tool (3-MinNS), a combination

of the parameters of weight loss, intake and muscle wastage yielded the largest
area under the curve (AUC) when compared to SGA. The best cutoff point for 3-
MinNS to identify malnourished patients was three (sensitivity 86%, specificity
83%). The subsequent study using nurses to conduct 3-MinNS confirmed the
validity (sensitivity 89%, specificity 88%) and reliability (agreement =78.3%, k=
0.58, p<0.001) of the tool.
After the hospital-wide implementation of nutrition screening, the error rates were
33% and 31% in 2008 and 2009 respectively, with 5% and 8% blank or missing
forms. Of all patients scored to be at risk of malnutrition, 10% were not referred
to a dietitian. With the implementation of a series of quality improvement
initiatives, error rates reduced to 25%, 15%, 7% and 5% in 2010, 2011, 2012 and
2013 respectively; with a reduction in blank or missing forms to 1% for four
consecutive years. Failure to refer appropriate patients to Dietetics decreased to
7% (2010), 4% (2011) and 3% (2012 and 2013).
In the fourth phase, among the malnourished patients seen by dietitians in the
inpatient wards, only 15% of patients returned for follow-up with a dietitian within
four months post-discharge. After implementation of ANS in 2010, the follow-up
rate was 100%. Mean weight improved from 44.0 8.5kg to 46.3 9.6kg
(p<0.001). The Euro Quality of Life - 5 Domain Visual Analogue Scale improved
4
from 61.2 19.8 to 71.6 17.4 and handgrip strength from 15.1 7.1 kg force to
17.5 8.5 kg force; p<0.001. Seventy-four percent of patients improved in SGA
score.
Conclusion
This research programme is amongst the first to examine the impact of

malnutrition on length of hospital stay, readmission, hospitalisation cost and
mortality in a large sample representative of patients admitted to a major
Singaporean tertiary hospital. It has provided clear evidence that the adverse
outcomes of malnutrition are not just a consequence of the disease process, and
lead to substantial increases in length of hospital stay, readmission rate, mortality
and hospitalisation cost when compared with well-nourished patients of similar
diagnoses and complexities. The research programme led to the development
and validation of a new nutrition screening tool (3-MinNS) and confirms that the
3-MinNS is a valid and reliable nutrition screening tool to be used in Singaporean
acute hospitals. Quality improvement initiatives proved successful in improving
the compliance of nurses to 3-MinNS and ensuring referral of malnourished or at
risk patients to dietitians. Finally, this research programme has provided an
evidence-based and effective method for following up malnourished patients
post-discharge, which resulted in improved nutritional status of these patients.
In conclusion, this research programme has successfully delivered a

comprehensive model for managing hospital malnutrition, from screening on
admission and referral for assessment, to intervention and post-discharge follow-
up.
KEYWORDS
Malnutrition, Prevalence, Nutrition Screening, Subjective Global Assessment,

Nutrition Intervention, Hospitalisation Outcomes
5
TABLE OF CONTENTS
ABSTRACT. 1
KEYWORDS 5
TABLE OF CONTENTS 6
LIST OF TABLES.. 10
LIST OF FIGURES.... 11
ABBREVIATIONS..... 12
PUBLICATIONS RELATED TO THE RESEARCH.... 14
CONFERENCE ABSTRACTS, POSTERS AND ORAL
PRESENTATIONS 15
AWARDS ... 17
MEDIA INTEREST RELATED TO THE PHD RESEARCH... 18
SPEAKER AT VARIOUS PLATFORMS RELATED TO TOPIC OF
THESIS . 19
ORIGINALITY OF WORK.... 21
ACKNOWLEDGEMENTS. 22
BACKGROUND.. 23
MODE OF THESIS PRESENTATION.... 27
CHAPTER 1: LITERATURE REVIEW 30
1.1 Malnutrition... 30
1.1.1 Definitions of Malnutrition.. 30
1.1.2 Prevalence of Malnutrition in the Acute Setting .... 34
1.1.3 Risk factors for Malnutrition . 40
1.1.4 Consequences, Clinical Outcomes and Cost of Malnutrition 42
1.1.5 Potential Confounders for Malnutrition Outcomes ... 56
6
1.2 Nutrition Screening. 60

1.2.1 Practical Considerations in Nutrition Screening............... 66
1.2.2 Validity of Nutrition Screening Tools 74
1.2.3 Reliability of Nutrition Screening Tools 78
1.2.4 Reference Standard to Validate a Nutrition Screening Tool 87
1.2.5 Compliance with Nutrition Screening.. 103
1.3 Interventions for Patients with Malnutrition. 109

1.3.1 Nutrition Support for Malnourished Patients . 111
1.3.2 Post-discharged Follow-up of Malnourished Patients.. 113
1.3.3 Strategies to Improve Post-discharged Follow- up 116
1.4 Overall Summary and Gaps in Current Research. 119
1.5 Conceptual Framework of Research Programme. 121
1.6 Research Questions and Objectives... 124
1.7 Aims of Research Programme. 126
CHAPTER 2: LINKING THE RESEARCH QUESTIONS .. 127
2.1 Linking the Research Questions and Objectives .. 127
2.2 How the Research Projects are Related 127
2.3 Overview of Research Methodology ... 130
CHAPTER 3: PREVALENCE OF MALNUTRITION AND ITS IMPACT ON
CLINICAL OUTCOMES AND COST . 135
3.1 Introduction .. 135
3.2 Publication: Malnutrition and its impact on cost of hospitalisation,
length of stay, readmission and 3-year mortality ... 136
3.3 Conclusion. 136
CHAPTER 4: DEVELOPMENT AND VALIDATION OF 3-MINUTE
NUTRITION SCREENING TOOL (3-MinNS) 145
4.1 Introduction... 145
4.2 Publication: Development and validation of 3-Minutes Nutrition
Screening (3-MinNS) Tool for acute hospital patients in Singapore 146
4.3 Conclusion. 147
7
CHAPTER 5: VALIDITY AND RELIABILITY OF 3-MINUTE NUTRITION 156

SCREENING ADMINISTERED BY NURSES...
5.1 Introduction. 156
5.2 Publication: Validity and reliability of nutrition screening administered

by nurses. 157
5.3 Conclusion.. 157
CHAPTER 6: IMPROVING THE PERFORMANCE OF NUTRITION

SCREENING... 165
6.1 Introduction. 165
6.2 Publication: Improving the performance of nutrition screening through
continuous quality improvement initiatives 166
6.3 Conclusion.. 166
CHAPTER 7: FOLLOW-UP FOR MALNOURISHED POST-

DISCHARGED HOSPITAL PATIENTS.. 191
7.1 Introduction 191
7.2 Publication: A pre-post evaluation of an ambulatory nutrition support

service for malnourished patients post hospital discharge: a pilot
study... 192
7.3 Conclusion. 192
CHAPTER 8: DISCUSSION AND RECOMMENDATIONS...... 200
8.1 Overview of Research Questions and Key Findings 200

8.1.1 Prevalence of malnutrition and outcomes.. 201
8.1.2 Identifying patients at nutrition risk.. 204
8.1.3 Compliance and referral process of nutrition screening... 209
8.1.4 Follow-up of malnourished patients post discharge.. 211
8.2 Recent Developments .. 215
8.3 Recommendations. 219
8.3.1 Clinical practice.. 219
8.3.2 Education. 220
8.3.3 Research.. 221
8.4 Conclusion and Contribution to Knowledge 223
8
REFERENCES 225
APPENDICES 248
Appendix A: 3-Minute Nutrition Screening.. 248

Appendix B: Subjective Global Assessment 249
Appendix C: 7-Point Subjective Global Assessment.. 250
Appendix D: Euro Quality of Life5 Domain and Visual Analogue 251
Scale
Appendix E: Paper on 7-point Subjective Global Assessment 253
Appendix F: Standard questions and advice given by the dietetics
assistant via telephone follow-up as part of Ambulatory
Nutrition Support Service . 276
Appendix G: Research Grant.. 277
Appendix H: Domain Specific Review Board Approval, Singapore... 279
Appendix I: Human Research Ethics Committee Approval, Queensland
University of Technology... 284
Appendix J: Statement of Contribution of Co-Authors... 288
9
LIST OF TABLES
Table 1.0 The amount a patient needs to pay for Dietetics services
with and without government subvention in National
University Hospital, Singapore.. 25
Table 1.1 NHMRC levels of evidence for prognostic, diagnostic and
intervention research. 29
Table 1.2 Summary of studies on prevalence of malnutrition in acute
hospitalised patients (multidisciplinary).. 37
Table 1.3 Impact of malnutrition on mortality, length of hospital stay
(LOS), readmission and cost, using different nutritional
assessment/ screening tools and evidence of them being
controlled for confounders 50
Table 1.4 Components, strengths and limitations of nutrition screening
tools for hospitalised patients.. 62
Table 1.5 Nutrition screening tools developed for targeted groups of
people.. 72
Table 1.6 Reliability and validity of nutrition screening tools in
hospitalised adult patients and method of validation,
including blinding of assessors 81
Table 1.7 Comparison of nutrition assessment methods to validate a
nutrition screening tool.. 89
Table 1.8 Studies on reliability and validity of Subjective Global
Assessment (SGA) and other modified versions of SGA 98
Table 1.9 Non-compliance and error rates with various nutrition
screening tools in hospitalised adult patients 107
Table 1.10 Studies on follow-up attendance rate of patients seen by
dietitians.. 115
Table 2.1 Research questions, objectives and relevant publications.. 129
Table 2.2 Summary of study design, sample size and sampling
strategy 130
10
LIST OF FIGURES
Figure 1.1 Vicious cycle of the development and progression of

disease-related malnutrition.. 31
Figure 1.2 Etiology-based malnutrition definitions 32
Figure 1.3 Prevalence of malnutrition in adult hospitalised patients
(multidisciplinary) differentiated by assessment tools... 35
Figure 1.4 Risk factors for malnutrition.. 41
Figure 1.5 Nutrition Care Process for malnutrition 110
Figure 1.6 Conceptual Framework for the identification and
management of malnutrition in hospitalised patients... 122
Figure 2.1 Schematic diagram on the methodology of research
programme. 133
11
ABBREVIATIONS
3-MinNS 3-Minute Nutrition Screening

ADA American Dietetic Association
(Academy of Nutrition and Dietetics)
ANS Ambulatory Nutrition Support
ASPEN American Society for Parenteral and Enteral Nutrition
BCM Body Cell Mass
BIA Bioelectrical Impedance Analysis
BMI Body Mass Index
BOD Burden of Disease
CAMA Corrected arm muscle area
CCI Charlson Comorbidity Index
CIRS Cumulative Illness Rating Scale
CPSS Computerised Patient Support System
DEXA Dual Energy X-Ray Absorptiometry
DMS Dialysis Malnutrition Score
DRG Diagnosis-related groups
DSRB Domain Specific Review Board
E7-SGA Expanded 7-point Subjective Global Assessment
EQ-5D VAS Euro Quality of Life 5 Domain Visual Analogue Scale
ESPEN The European Society for Clinical Nutrition and Metabolism
ICED Index of Coexisting Disease
IF Impact factor
LOS Length of hospital stay
MAC Mid arm circumference
MAMC Mid arm muscle circumference
MDS Malnutrition Dialysis Score
MNA Mini Nutritional Assessment
MNA-SF Mini Nutritional Assessment Short Form
MST Malnutrition Screening Tool
MUST Malnutrition Universal Screening Tool
12
NPV Negative Predictive Value

NRIa Nutrition Risk Index
NRIb Nutritional Risk Index
NRS Nutrition Risk Score
NRS 2002 Nutritional Risk Screening 2002
NS Non significant
NST Nutrition Screening Tool
NHG National Healthcare Group
NHMRC National Health and Medical Research Council
NUH National University Hospital
NUS National University of Singapore
NUHS National University Health System
PDCA Plan Do Check Action
PG-SGA Patient-Generated Subjective Global Assessment
PPV Positive Predictive Value
QI Quality Initiative
QOL Quality of Life
RCT Randomised Controlled Trial
ROC Receiver Operator Characteristic
RQ Research Question
SGA Subjective Global Assessment
SNAQ Short Nutritional Assessment Questionnaire
SAP System Application Product
TST Triceps Skinfold Thickness
URS Undernutrition Risk Score
VSM Value Stream Mapping
13
PUBLICATIONS RELATED TO THE RESEARCH
1. Lim SL, Ong KCB, Chan YH, Loke WC, Ferguson M, Daniels L.
Malnutrition and its impact on cost of hospitalisation, length of stay,
readmission and 3-year mortality. Clinical Nutrition 2012; 31(3):345-350.
(IF = 3.603) (CHAPTER 3)
2. Lim SL, Daniels L. Reply - Malnutrition and its impact on cost of

hospitalisation, length of stay, readmission and 3-year mortality. Clinical
Nutrition 2013; 32(3):489-490. (IF = 3.603) (CHAPTER 3)
3. Lim SL, Tong CY, Ang E, Lee EJC, Loke WC, Chen Y, Ferguson M,
Daniels L. Development and validation of 3-Minutes Nutrition Screening
(3-MinNS) Tool for acute hospital patients in Singapore. Asia Pacific
Journal of Clinical Nutrition 2009; 18(3):395-403. (IF = 1.438). (CHAPTER
4)
4. Lim SL, Ang E, Foo YL, Ng LY, Tong CY, Ferguson M, Daniels L. Validity
and reliability of nutrition screening administered by nurses. Nutrition in
Clinical Practice 2013; 28:730-736. (IF = 1.594). (CHAPTER 5)
5. Lim SL, Ng SC, Lye J, Loke WC, Ferguson M, Daniels L. Improving the
performance of nutrition screening through continuous quality
improvement initiatives. Joint Commission Journal of Quality and Patient
Safety 2014. (In press) (CHAPTER 6)
6. Lim SL, Lin XH, Chan YH, Ferguson M, Daniels L. A pre-post evaluation
of an ambulatory nutrition support service for malnourished patients post
hospital discharge: a pilot study. Annals Academy of Medicine Singapore
2013. 42:507-513. (IF = 1.362). (CHAPTER 7)
7. Lim SL, Lye J, Liang S, Miller M, Chong YS. Development and Validation
of an Expedited 10g Protein Counter (EP-10) for Dietary Protein Intake
14
Quantification. Journal of Renal Nutrition 2012; 22(6):558-566. (IF =

1.480)
8. Lim SL. Using Expedited 10g Protein Counter (EP-10) for Meal Planning.
2012; Journal of Renal Nutrition 2012; 22(6):e55. (IF = 1.480)
9. Lim SL, Lye J. Nutritional Intervention Incorporating Expedited 10g

Protein Counter (EP-10) to Improve the Albumin and Transferrin of
Chronic Hemodialysis Patients. International Scholarly Research Network
- Nutrition 2012. Article ID 396570, 5 pages. Doi: 10.5402/2013/396570
10. Ang Emily, Lim HL, Mordiffi SZ, Chan MF, Lim SL. Nutritional risk of
cancer patients receiving chemotherapy in the ambulatory care setting: A
prospective study. Singapore Nursing Journal 2012; 39(4):2-11.
CONFERENCE ABSTRACTS, POSTERS AND ORAL PRESENTATIONS
1. Lim SL, Tong CY, Ang NK, Loke WC, Chen Y, Lee EJC, Ferguson M,
Daniels L. A Simplified Nutrition Screening Tool (3-MinNS) has Good
Correlation with Subjective Global Assessment (SGA). 15th International
Congress of Dietetics, Seoul, Korea, Sept 2008.
2. Lim SL, Tong CY, Seet RCS, Loke WC, Ferguson M, Daniels L.
Validation of Mid Arm Muscle Circumference and Triceps Skinfold
Thickness with Subjective Global Assessment. In: Oh VMS, Yap HK,
editors. Proceedings of the 8th Annual Scientific Congress National
Healthcare Group; 16-17 Oct 2009: Singapore: Annals of the Academy of
Medicine; 2009; 38:S8.
3. Lim SL, Lye J, Liang S, Miller M, Chong YS. Development and Validation
of an Expedited 10g Protein Counter (EP-10) for Dietary Protein Intake
Quantification. In: Tan EK, editor. Proceedings of the Singapore Health
15
and Biomedical Congress; 11-12 Nov 2011: Singapore: Annals of the

Academy of Medicine; 2011. 40: S41.
4. Lim SL, Ang E, Foo YL, Ng LY, Tong CY, Ferguson M, Daniels L. Validity
and reliability of 3-Minute Nutrition Screening performed by nurses on
Oncology and Surgical patients. 14th Congress of Parenteral and Enteral
Nutrition Society of Asia, Taipei, Taiwan, 14-16 October 2011.
readmission and 3-year mortality. In: Tan EK, editor. Proceedings of the
2nd Singapore Health and Biomedical Congress; 11-12 Nov 2011:
Singapore: Annals of the Academy of Medicine; 2011;40:S10.
6. Lim SL, Ang E, Ng SC, Tong CY, Ferguson M, Daniels L. Reducing

incompletion and error rates in nutrition screening. Proceedings of the 3rd
Singapore Health and Biomedical Congress; 28-29 Sept 2012: Singapore:
Annals of the Academy of Medicine; 2012; 41:S194.
of a post-discharge ambulatory nutrition support service for malnourished
patients: a pilot study. Proceedings of the Singapore Health and
Biomedical Congress; 28-29 Sept 2012: Singapore: Annals of the
Academy of Medicine; 2012; 41:S67.
of a post-discharge ambulatory nutrition support service for malnourished
patients: a pilot study. 21st International HPH Conference, Gothenburg,
Sweden, 22-24 May 2013.
16
AWARDS
Year Awards
2013 Awarded NUHS Outstanding Quality Improvement Award for Multi-year
Project to Improve the Performance of Nutrition Screening Hospital-wide
(the only project accorded outstanding status out of 29 projects
evaluated)
2013 Awarded Singapore Allied Health Award 1st Prize Winner for Best
Oral Presentation in Singapore Health and Biomedical Congress for
research on 7-point Subjective Global Assessment is more time sensitive
than conventional Subjective Global Assessment in detecting nutritional
changes.
2011 Awarded Singapore Allied Health Award 1st Prize Winner for Best
Oral Presentation in Singapore Health and Biomedical Congress for
research on Malnutrition and its impact on cost of hospitalisation, length
of stay, readmission and 3-year mortality.
2011 Awarded National Healthcare Quality Improvement Commendation

Prize for project titled Reducing the turnaround time from nutrition
screening referral of nutritionally at risk patients till being seen by
dietitians.
2011 Awarded Model Allied Health Professional Award 2011 (for Research
Contribution)
2011 Awarded Outstanding Project Award for project titled Dietetics Referral
by Nurses Using Nutrition screening Score upon Admission.
2011 Gold Medalist for Singapore Public Service 21 Excellence Award for
best ideas contributed in the area of public service (nutrition screening and
management)
2010 Awarded a grant (S$153,000) from the Healthcare Quality Improvement

and Innovation Funds for proposal on Development of novel
approaches to improve the nutritional status of malnourished patients
discharged from hospital and evaluation of its effectiveness.
2009 2nd Runner-up for Best Oral Presentation for research on Validation of
Mid Arm Muscle Circumference and Triceps Skinfold Thickness with
Subjective Global Assessment
2009 Merit Award for NUHS Way Quality Improvement Project titled Improving
the Nutritional Status of Patients Discharged to Nursing Home on Tube
Feeding
17
Media Interest Related to the PhD Research
Newspaper interview: More Malnutrition cases detected with NUH

Screening Tool, TODAY, 11 January 2013.
http://www.nuh.com.sg/wbn/slot/u3007/Patients%20and%20Visitors/Newsro
om/Media%20Articles/2013/JAN/today_11Jan_more%20malnutrition%20ca
ses%20detected%20with%20NUH%20screening%20tool.pdf
Newspaper interview: NUH sees Success in Tackling Patient with

Malnutrition Strait Times.com, 11 January 2013.
http://www.straitstimes.com/breaking-news/singapore/story/nuh-sees-
success-tackling-patient-malnutrition-20130111
Newspaper interview: Malnutrition Problems Berita Harian (a Singapore

National newspaper in Malay language), 27 February 2013.
http://www.nuh.com.sg/wbn/slot/u3007/Patients%20and%20Visitors/Newsro
om/Media%20Articles/2013/FEB/270213%20BH%20Pg%2012.pdf
Health Magazine interview: Malnutrition and Ageing EzyHealth, April

Issue, 2013
National University Hospital Publication: Beating Malnutrition, LifeLine,

2012, Issue 1, Page 6.
http://www.nuh.com.sg/wbn/slot/u3007/lifeline/Lifeline_1_2012.pdf
Newspaper interview: Eating Well in Your Autumn Years, TODAY, 20

February 2010
http://www.nuh.com.sg/wbn/slot/u2995/mediaarticlesfeb10/TDY_20-
21Feb10_Eating_well_in_your_autumn_years.pdf
Agency of Integrated Care Publication: 3-Minute Nutrition Screening

Mosaic, October 2011, Issue 7, Page 9
http://www.aic.sg/newsletter/archives/mosAIC/Oct2011/index.html#/10/
18
SPEAKER AT VARIOUS PLATFORMS RELATED TO TOPIC OF THESIS
DATE TOPICS AUDIENCE

New Nurses in National
9/6/09 Nutrition Screening
University Hospital (NUH)
14/7/09 Nutrition Screening New Nurses in NUH
Malnutrition cutoffs and validation of upper-arm
27/10/09 NUH Dietitians and students
anthropometrics for Singapore patients
15/12/09 Nutrition Screening Roadshows NUH Nurses
17/12/09 Nutrition Screening Roadshows NUH Nurses
Medical Students Year 2,
8/3/10 Nutritional Needs of the Elderly National University of
Singapore
Total Nutrition Therapy
Course for Doctors,
8 & 9/10/10 Nutrition Screening and Assessment
Nurses, Dietitians in
Singapore
Dietitians & Nurses from
22/12/10 3-Minute Nutrition Screening University Hospital, Kuala
Lumpur, Malaysia
Nutrition Screening
12/1/11 New Nurses in NUH
Development and Validation of Expedited 10 g
25/1/11 Dietitians, NUH
Protein Exchange (EP-10)
Dietitians & Nurses from
25/3/11 3-Minute Nutrition Screening
Jurong General Hospital
Course for Community
8/4/11 Nutrition Screening and Assessment
Hospitals and Nursing
Homes
4th year Medical Students
Nutrition assessment and Clinical Outcomes of
28/6/11 from National University of
Malnutrition
Singapore
14/4/11 Subjective Global Assessment Dietitians in NUH
13/4/11 3-Minute Nutrition Screening New Nurses in NUH
Nurses from Sree
3/8/11 3-Minute Nutrition Screening
Narayanan Nursing Home
19
Nursing Directors from all

15/8/11 Importance of Nutrition Screening
nursing homes in Singapore
Malnutrition
Singapore
Malnutrition
Singapore
nd
Malnutrition and its impact on cost of 2 Singapore Health and
11/11/11 hospitalisation, length of stay, readmission and 3- Biomedical Congress
year mortality.
Advanced Diploma in
21/11/11 Nutritional Management of Renal Patient
Nursing (Nephrology)
Expedited 10g Protein Counter (EP-10) for
Dietitians in Malaysia
6/1/12 Quantification and Recommendation of Dietary
(Central Region)
Protein Intake
Advanced Diploma in
25/1/12 Nutritional Management of Orthopaedic Patient
Nursing (Orthopaedics)
Gastroenterology Society of
Importance of Hospital Malnutrition, Nutrition
25/2/12 Singapore Nutrition
Screening and Assessment
Support Course 2012
Using a novel Expedited 10 g Protein Counter Malaysian Dietetics
17/7/12 (EP-10) for assessment and meal planning of Congress (Dietitians in
dialysis patients Malaysia)
24/8/12 Screening and Nutrition Assessment Course for Clinicians and
Dietitians in Singapore

Malnutrition and Nutrition Management in
10/10/12 Surgical Doctors in NUH
Surgical patients
Nutritional Requirements and Challenges for the
5/11/12 Doctors, Dietitians, Nurses
Elderly in Singapore
23/8/13 Screening and Nutrition assessment Course for Clinicians and
Dietitians in Singapore
20
ORIGINALITY OF WORK
The work in this thesis is my original work not submitted elsewhere for a degree
at any other institution of higher education. To the best of my knowledge, this
thesis contains no material previously published or written by another person
except where due reference is made.
Signed: Date: 17 January 2014
21
ACKNOWLEDGEMENTS
There are many people to whom I am indebted and the journey to complete my
PhD would not be possible without them.
First and foremost, I would like to thank my supervisors Professor Lynne Daniels
and Dr. Maree Ferguson for their continual support and encouragement.
Professor Lynne Daniels encouraged me to pursue a Masters and later nudged
me to articulate to PhD. I would not have dreamed of doing Masters 6 years ago,
what more a PhD. Lynne, thanks for your continual support, guidance, direction
and your belief in me.
I would like to thank both my supervisors for their expertise, valuable input and
advice on my thesis and publications. If Prof. Daniels provided her wealth of
wisdom and the eagles eye view, Dr. Ferguson complemented with her detailed
and meticulous input. I must say that I have the best pair of supervisors I could
ever wish for. I greatly admire their intellect, rigor and dedication.
I would also like to thank my husband Dr. Loke Wai Chiong, whose tremendous
encouragement, love, understanding and support keeps me going when the
journey gets tough. Many times when I had to burn the midnight oil to write my
manuscript for publication or thesis, he would keep me company. My two young
boys were incredibly understanding and somehow understood that their Mama
could not be as available to them during this time.
My gratitude goes also to National University Hospitals Director of Clinical

Support Services Mr. Dennie Hsu, my boss who supported my scholarship
application; and encouraged and supported me in my journey.
Finally, I would like to acknowledge National Healthcare Group and subsequently

National University Hospital for providing me with a scholarship for my PhD.
22
BACKGROUND
This section outlines the background to the research setting, student and the
Singapore healthcare financing system.
Setting
National University Hospital (NUH), a 987-bed acute tertiary hospital, is one of

the leading hospitals in Singapore with a comprehensive range of medical and
surgical specialties. It has close links with the National University of Singapore
(NUS), which is situated adjacent to the hospital. All study participants recruited
under this research programme are patients of National University Hospital.
Students Background
The student, Ms. Su Lin Lim is presently the Senior Assistant Director and Chief
Dietitian of the Dietetics Department at the National University Hospital,
Singapore. She developed a nutrition screening tool in 2002 for National
University Hospital, Singapore. As principal investigator, she was subsequently
awarded a S$45,000 grant from Singapores National Healthcare Group to
validate the tool on newly admitted hospitalised patients. After embarking on her
PhD, the tool was refined, with validation of the final version of the tool called 3-
Minute Nutrition Screening (3-MinNS) published in 2009 (Lim et al., 2009),
(Appendix A). The tool is currently being used hospital-wide in National
University Hospital, in other hospitals and nursing homes in Singapore and in
Malaysia. She has since taught over 7000 Singaporean and Malaysian nurses
and dietitians to use the 3-MinNS.
Another focus of her PhD research programme was to examine ways to improve
the outcomes of malnourished hospitalised patients discharged from the hospital.
23
The proposal was awarded funding of $153,000 from the Healthcare Quality
Initiative and Innovative Fund (HQI2F).
Su Lin has received numerous awards for her research and quality improvement
initiatives. She was awarded the National University Health Services Model Allied
Health Award in 2011 for her research contributions. At the national level, she
won the National Healthcare Group (NHG) Best Oral Presentation in 2006, NHG
Young Investigators Bronze Award in 2007, Singapore Allied Health Award for
best oral presentation for 3 consecutive years from 2009 to 2011. She was
awarded the Singapore National Day Efficiency Medal in year 2007 and was a
Gold Medalist for Singapore Public Service 21 Excellence Award in 2011.
Singapore Healthcare Financing System
Singapore emphasises personal responsibility in healthcare financing. In general,

acute health care services in Singapore are financed through the individual,
government and/or means testing.
a) Individual
Out-of-pocket payment at the point of consumption Patient have to pay out

of pocket for almost all medical and other outpatient services including
Dietetics services. Typically, a patient will have to pay S$55 for the 1st
Dietetics Consult and S$30 for a Follow-up Dietetics Consult.
Medisave this is a medical savings scheme with emphasis on personal
responsibility. For an employed person, 20% of his/her pay and 14.5% of the
employers contribution will go towards Central Provident Fund (CPF) every
month. The Central Provident Fund is a social security savings plan
that provides for Singaporeans in old age. Out of the monthly 34.5%
contribution, 6.5% will go into Medisave, which can be used to pay for
24
inpatient charges when a patient is admitted to hospital. Medisave cannot be

used for outpatient payment.
Medishield this is health insurance for catastrophic illness with premiums
payable from Medisave. A person can also subscribe to other insurance
schemes from private insurance companies. A person who has Medishield
can claim insurance for his/her hospitalisation expenses.
b) The Government
The Singapore government provides subventions to subsidise the healthcare

cost of patients in the public sector health services. These subventions are
demonstrated in Table 1.0 for Dietetics services. An estimate of 80% of
patients seen by a dietitian pay the subsidised rate.
Table 1.0: The amount a patient needs to pay for Dietetics services with and
without government subvention in National University Hospital, Singapore
1st Dietetics Follow-up Dietetics
Patient Group
Intervention Intervention
Private patient (no

S$ 55 S$ 30
government subvention)
Subsidised patient (with
S$ 22 S$ 12
government subvention)
Medifund or social assistance is available for patients who cannot afford to

pay the subsidised rate, following financial assessment by a medical social
worker. Patients who qualify for Medifund will have between 50%-100% of
their already subsidised medical bills paid by the government for both
inpatient and outpatient services. About 5% of Dietetics outpatients are
funded by Medifund.
25
c) Means Testing
The Singapore Government recognises that intermediate and long-term

healthcare can be expensive.
To ensure government subsidies for healthcare services are distributed
appropriately and equitably, an income assessment framework Means
Testing was introduced in the year 2000.
Subsidies are available only to individuals who are Singapore citizens or
permanent residents and are applicable only to government-funded
institutions and services.
Means Testing takes into consideration gross income of the patient, spouse
and immediate family members, number of family members and ownership of
major assets such as private property.
In summary, healthcare expenses in Singapore are partially borne by the user,

with subsidies available to the majority of patients in a government restructured
hospital. Compulsory medical savings play an important role, and insurance such
as Medishield helps to cover for catastrophic illnesses and large hospitalisation
bills. Medifund provides a second-layer safety net for those who are financially
challenged. However financial support for outpatient care is limited.
26
MODE OF THESIS PRESENTATION
The mode of thesis presentation will be by publication. The thesis contains 5

research papers which have been published in peer-reviewed indexed journals.
All the papers are formatted according to the requirements of the individual
journal. Careful planning and execution of the research programme has been
taken to ensure smooth flow of the papers from one to another as the chapters of
the thesis unfold, which are then linked back to the research questions. In order
to minimise repetition and rework, each chapter other than chapters 1 (literature
review), 2 (linking the research questions) and 8 (discussion and
recommendations), will contain a brief introduction of the respective research
paper, the paper itself and a summary of the chapter. The introduction and
summary in the relevant chapters will serve to connect the research questions
and papers to each other.
27
LITERATURE SEARCH STRATEGY
The literature search was undertaken as a selective narrative review to address a

range of questions and identify gaps in the current literature, which would lead to
formulation of research questions for this thesis.
Literature search was conducted for English only published or unpublished

scientific papers from 1980 to 2012 using keyword search terms of malnutrition,
prevalence, outcomes, nutrition screening, nutrition assessment, nutrition
intervention, nutrition support, multidisciplinary, hospital and inpatients. Additional
filters were used to narrow the searches to adult age groups. PubMed, Web of
Science and Medline databases were used. Searches were supplemented by
reviews, textbooks and by reviewing the references in the studies found. The
initial search was conducted from 1980 to 2008 to develop the conceptual
framework and research questions. Subsequent searches were conducted using
the same search terms and strategies from 2008 to 2013 to include new
references. To maintain focus and smooth flow of the thesis, article exclusion
criteria were applied on papers which focused on non-hospitalised population,
elderly patients and discipline/disease-specific studies. Studies which were
published subsequent to my research and publications are discussed and
outlined in the discussion chapter.
28
LEVELS OF EVIDENCE
The levels of evidence were determined according to the NHMRC (National

Health and Medical Research Council, 2009) criteria for prognostic, diagnostic
and intervention studies (Table 1.1). The main purpose of this was to evaluate
the quality of studies in the context of discussing specific review questions and
identifying gaps in the current evidence.
Table 1.1: NHMRC levels of evidence for prognostic, diagnostic and intervention
research (National Health and Medical Research Council, 2009)
Level Prognosis Diagnostic accuracy Intervention
I A systematic review of A systematic review of level A systematic review of level II
level II studies II studies studies
II A prospective cohort A study of test accuracy with: an A randomised controlled trial
study independent, blinded comparison
with a valid reference standard,
among consecutive persons with a
defined clinical presentation6
III-1 All or none A study of test accuracy with: an A pseudorandomised controlled
independent, blinded comparison trial
with a valid reference standard, (i.e. alternate allocation or some
among non-consecutive persons other method)
with a defined clinical presentation6
III-2 Analysis of prognostic A comparison with reference A comparative study with
factors amongst persons standard that does not meet the concurrent controls:
in a single arm of a criteria required for Non-randomised, experimental
randomised controlled trial Level II and III-1 evidence trial9
Cohort study
Case-control study
Interrupted time series with a
control group
III-3 A retrospective cohort Diagnostic case-control study A comparative study without
study concurrent controls:
Historical control study
Two or more single arm study
Interrupted time series without
a parallel control group
IV Case series, or cohort Study of diagnostic yield (no Case series with either post-test
study of persons at reference standard) or pre-test/post-test outcomes
different stages of disease
29
Chapter 1: LITERATURE REVIEW
Introduction
This chapter includes a literature review on the prevalence of malnutrition, its

risks and consequences, and current practices in nutrition screening for
hospitalised adult patients. This is followed by a review of interventions and
follow-up for patients with malnutrition. The review will identify significant gaps in
the evidence for local malnutrition data and its impact on outcomes, nutrition
screening and interventions, which form the basis of the research questions in
this thesis.
1.1 MALNUTRITION
1.1.1 Definition of Malnutrition
There are several definitions of malnutrition and there is as yet no consensus on

a standard one. Malnutrition can also include overnutrition or obesity (Soeters et
al., 2008), however in this thesis the focus will be on the undernutrition aspect of
malnutrition.
Soeters et al. defined malnutrition as a subacute or chronic state of nutrition in

which a combination of varying degrees of over- or under-nutrition and
inflammatory activity has led to a change in body composition and diminished
function (Soeters et al., 2008). In a review paper by Norman et al. (2008), the
definition of malnutrition focused on imbalance between intake (energy, protein
and micronutrients) and requirements as the main cause of malnutrition (Norman
et al., 2008b). However, the authors also described the inflammatory response
as a contributing factor to malnutrition (Norman et al., 2008b). They explained
that malnutrition together with stress-related catabolism caused by inflammation
30
increases the risk for infections, organ dysfunction and impaired healing. These,
as with all other severe acute illnesses, can be a trigger for the inflammatory
response that consequently results in starvation and catabolism, which further
aggravates malnutrition (see Figure 1.1).
Figure 1.1: Vicious cycle of the development and progression of disease-

related malnutrition (Norman et al., 2008b)
Reprinted with permission from Elsevier
The European Society for Clinical Nutrition and Metabolism (ESPEN) consensus
report defined malnutrition as a state resulting from lack of uptake or intake of
nutrition leading to altered body composition (decreased fat free mass and body
cell mass) and diminished function (Lochs et al., 2006). A similar definition was
proposed by Hoffer and Jeejeebhoy whereby malnutrition was defined as a state
of nutrient insufficiency, as a result of either inadequate nutrient intake or inability
to absorb or use ingested nutrients (Kinosian & Jeejeebhoy, 1995; Jeejeebhoy,
2000; Hoffer, 2001).
An International Committee was constituted to develop a consensus approach to

defining malnutrition syndromes for adults in the clinical setting (Jensen et al.,
31
2009; Jensen et al., 2010; White et al., 2012). Consensus was achieved through
a series of meetings held at the American Society for Parenteral and Enteral
Nutrition (ASPEN) and ESPEN Congresses. It was agreed that an etiology-based
approach, which incorporates the current understanding of the role of
inflammatory response in the development of malnutrition, would be most
appropriate (Jensen et al., 2009; Jensen et al., 2010). The Committee proposed
the following nomenclature for nutrition diagnosis in adults in the clinical practice
setting: 1) "starvation-related malnutrition", when there is chronic starvation
without inflammation, 2) "chronic disease-related malnutrition", when
inflammation is chronic and of mild to moderate degree, and 3) "acute disease or
injury-related malnutrition", when inflammation is acute and of severe degree
(Jensen et al., 2010; White et al., 2012) (Figure 1.2).
Figure 1.2: Etiology-based malnutrition definitions. Originated from Jensen

GL and adapted by White JV (Jensen et al., 2009; White et al., 2012)
Reprinted with permission from Elsevier
32
It is likely that screening identifies malnutrition risk with a variety of etiologies

such as nutrient deficiencies, sarcopenia, frailty, cachexia of ageing and cancer
cachexia which will have an impact on treatment and outcomes (Jeejeebhoy,
2012; Vandewoude et al., 2012). Therefore, for patients who have been identified
as at risk via screening, it is important to follow through with a thorough
nutritional and clinical assessment so that the etiology of malnutrition can be
established and a more targeted approach to treatment carried out (Hamerman,
2002; Burton & Sumukadas, 2010; Morley et al., 2010; Theou et al., 2011; Sayer
et al., 2013).
33
1.1.2 Prevalence of Malnutrition in the Acute Setting
Many previous studies have shown that malnutrition is prevalent in hospitals

(McWhirter & Pennington, 1994; Edington et al., 2000; Waitzberg et al., 2001;
Thomas et al., 2002; Gout et al., 2009; Beghetto et al., 2010). Figure 1.3
compares the prevalence of malnutrition in adult hospitalised patients
(multidisciplinary) using different tools. Prevalence rates of malnutrition between
2% and 69% were observed depending on the tools used and the countries
where the studies were conducted (Kelly et al., 2000; Middleton et al., 2001;
Waitzberg et al., 2001; Thomas et al., 2002; Correia & Campos, 2003; Kyle et al.,
2003; Wyszynski et al., 2003; Pirlich et al., 2006; Singh et al., 2006; Beghetto et
al., 2010; Velasco et al., 2011). Table 1.2 provides more details of the studies
with regard to sample size, cutoffs for malnutrition, time of assessment and the
age of the subjects. Different methodologies used in studies make comparison
and determination of the true prevalence of malnutrition difficult (Corish &
Kennedy, 2000; Covinsky et al., 2002; Pirlich et al., 2003; Kubrak & Jensen,
2007; Gomes Beghetto et al., 2011). For example, in Pirlichs (2003) study, even
when prevalence was studied on the same cohort of patients, results varied
depending on the tool used; 27% using Subjective Global Assessment (SGA),
17% using Arm Muscle Area (AMA) and only 4% using body mass index (BMI).
When the same tool for example SGA, was used in large multicentre studies,
there had been variability in the prevalence amongst studies in different countries
from about 27% in German hospitals (Pirlich et al., 2003) to 50% in Latin
American hospitals (Correia & Campos, 2003). This shows that the prevalence of
malnutrition may vary according to the nutrition assessment tool used as well as
patient characteristics such as disease state, age, ethnic mix, culture, country
and institutional setting. In addition, various studies have used different cutoffs
for malnutrition, which affects the rate of malnutrition. For example, three studies
used different BMI cutoffs for malnutrition: <18.5 kg/m2, <19 kg/m2 and < 20
kg/m2 (Kelly et al., 2000; Thomas et al., 2002; Kyle et al., 2003).
34
Although many studies have been conducted in various countries and contexts,
the use of different tools and cutoff points in these studies limits their
comparability and hence prevents us from truly understanding the prevalence
and scale of the problem globally and in Singapore. Despite the variability in the
tools used, SGA remains the most commonly used nutrition assessment tool to
determine the prevalence of malnutrition in many countries.
At the time prior to this research programme, the true prevalence of malnutrition
in a Singapore acute hospital had not yet been established. A study done in Tan
Tock Seng Hospital, an acute hospital in Singapore, found the prevalence to be
14.7% using Subjective Global Assessment (SGA) (Raja et al., 2004). However,
the figure could be higher as the author performed SGA only on patients who
were screened to be at risk of malnutrition with the Malnutrition Screening Tool,
and thus might have missed those who were not detected by the screening tool
(Raja et al., 2004).
Summary of Issues on Prevalence of Malnutrition
In conclusion, there have been studies to measure and confirm the prevalence of
hospital malnutrition in many countries. Evidence of this being a problem is
important to create the appropriate level of awareness that can lead to an action
plan. However there has not been any study on the true prevalence of
malnutrition in Singapore acute hospitals. As Singapore is a developed country,
many healthcare professionals are not convinced that malnutrition exists.
Establishing the prevalence of malnutrition in hospitalised patients in Singapore
is the first-step in understanding whether this is indeed an issue worth
addressing.
35
Figure 1.3: Prevalence of malnutrition in adult hospitalised patients (multidisciplinary) differentiated by nutrition
assessment tools*
Singh 2006 (Canada)
80 Braunchweig 2000 (United States)

Correia 2003 (Latin America)
Wyzynski 2003 (Argentina)
70
Prevalence of malnutrition (%)
Waitzberg 2001 (Brazil)

Lazarus 2005 (Australia)
60 Middleton 2001 (Australia)
Banks 2010 (Australia)
Gout 2009 (Australia)
50 Velasco 2011 (Spain)
Pirlich 2006 (Germany)
40 O'Keefe 1986 (South Africa)
Barreto Penie 2005 (Cuba)
30 Beghetto 2010 (Brazil)

Devoto 2006 (Italy)
Thomas 2002 (United States)
20 Kyle 2003 (Switzerland)
Kyle 2003 (Germany)
10 Kelly 2000 (United Kingdom)
Dzieniszewski 2005 (Poland)
McWhirter 1994 (United Kingdom)
0 Edington 2000 (United Kingdom)
Meijers 2009 (Netherlands)
al
A
ds
A
M
M
M
SG
FF
ic
ho
B
M
et
e
ch
m
io
d
ne
B
bi
om
C
SGA = Subjective Global Assessment, MNA = Mini Nutritional Assessment, BMI = Body Mass Index, AMA = Arm muscle area, FFM = Fat free mass,
Biochemical = Albumin, prealbumin or total lymphocyte count, Combined methods = Combination of two or more nutritional indicators. *See Table 1.2 for details of
study
36
Table 1.2: Summary of studies on prevalence of malnutrition in adult hospitalised patients (multidisciplinary), sorted
by nutrition assessment tools
Prevalence of
Author, Assessment Tool used and Cutoff/ Malnutrition (%)
N Subjects (Country) Age range (years) Time of Assessment
Year Indicator for Malnutrition
(Singh et al., Inpatients from a general medical ward > 20 yrs (upper range not
69 SGA: Ratings B & C 69 During hospital stay
2006) (Canada) specified)
(Braunschweig > 18 yrs (upper range not Admission: 54 Within 72 hours of admission and
404 Inpatients staying > 7 days (United States) SGA: Ratings B & C
et al., 2000) specified) Discharge: 59 during discharge
(Correia & Inpatients from hospitals in 13 Latin > 18 yrs (upper range not
Campos, 2003) America countries (Latin America) specified)
(Wyszynski et
Inpatients from 38 general hospitals > 18 yrs (upper range not
al., 2003) (Argentina) specified)
(Waitzberg et
Inpatients from 25 general hospitals
4000 18-90 SGA: Ratings B & C 48 During hospital stay
al., 2001) (Brazil)
(Lazarus &
Inpatients in an acute private hospital in > 18 yrs (upper range not
324 SGA: Ratings B & C 42.3 During hospital stay
Hamlyn, 2005) Sydney (Australia) specified)
(Middleton et
Inpatients in two Sydney teaching
819 18-99 SGA: Ratings B & C 36 During hospital stay
al., 2001) hospitals (Australia)
(Banks et al.,
Inpatients from 22 acute care facilities in > 18 yrs (upper range not
2007) Queensland (Australia) specified)
(Velasco et al., Inpatients from internal medicine and

> 18 yrs (upper range not SGA: Ratings B & C 35.3 (SGA)
400 surgery departments in three university Within 36 hours of admission
2011) specified) MNA Score < 17 58.5 (MNA)
hospitals (Spain)
(Barreto Penie,
Inpatients from 12 Cuban hospitals > 19 yrs (upper range not 41
1905 SGA: Ratings B & C During hospital stay
2005) (Cuba) specified)
Year 02 Year 06
Yr 02
185
(Beghetto et al., Inpatients from a university hospital Adult (age range not SGA: Ratings B & C 39.3 40.2
On admission
2010) (Brazil) specified) BMI <18.5 kg/m2 1.6 5.7
Yr 06
Albumin <3.5 g/dL 21.3 33.5
1503
Total Lymphocyte Count < 1.5x103/m3 63.9 42.2
SGA = Subjective Global Assessment, PG-SGA = Patient-Generated Subjective Global Assessment, BMI = Body Mass Index, MNA = Mini Nutritional Assessment
37
by nutrition assessment tools (continued)
Author, Assessment Tool used and Cutoff/ Indicator for Prevalence of
N Subjects (Country) Age range (years) Malnutrition (%) Time of Assessment
Year Malnutrition
(Gout et al.,
Inpatients from a public tertiary hospital Within 2 weeks of
275 Not specified SGA: Ratings B & C 23
2009) (Australia) admission
(Pirlich et al., SGA: Ratings B & C 27.4 (SGA)
Inpatients from 7 teaching hospitals and Same day as hospital
1886 18-100 BMI <18.5 kg/m2 4.1 (BMI)
2006) 6 community hospitals (Germany) admission
Arm muscle area < 10th percentile norm value 17.1 (AMA)
SGA: Ratings B & C 53
(Devoto et al.,
Prealbumin 0.17g/L 60 On the 3rd day after
108 Patients admitted to a hospital (Italy) 28-99
2006) PINI Score 1 64 admission
RBP 0.03g/L 59
(Thomas et al., Patients admitted to an Acute > 18 yrs (upper range Within 48 hours of
64 PG-SGA: Ratings B & C 53
2007) Assessment Unit (Australia) not specified) admission
BMI <19 kg/m2 18
(Thomas et al.,
489 Sub-acute care hospital (United States) 23-102 Albumin <35 g/L 53 During hospital stay
2002)
MNA Score < 17 30
Geneva Berlin
Patients admitted to 2 hospitals in 17.3 8.5
(Kyle et al., BMI <20 kg/m2 14.9 11.2 Within 24 hours of
1760 Geneva and Berlin Not specified
2003) Albumin <35 g/L 31 17 admission
(Switzerland & Germany)
Fat free mass < 10th percentile of healthy Swiss subjects
(Kelly et al., Patients admitted to a tertiary hospital BMI <18.5 kg/m2 Within 24 hours of
219 18-94 9
2000) (United Kingdom) admission
BMI <20 kg/m2 10
(Dzieniszewski Patients admitted to 4 teaching On the 1st day of
3310 16-100 Albumin <3.5 g/dL 21
et al., 2005) hospitals (Poland) admission
Total Lymphocyte Count < 1.5x103/m3 21
Body weight below 20% of ideal weight 20
(O'Keefe et al., Patients admitted medical and surgical Adult (age range not Fat stores less than 60% of standard 30
700 Not specified
1986) wards (South Africa) specified) Arm muscle circumference and 15
area less than 80% of ideal
SGA = Subjective Global Assessment, PG-SGA = Patient-Generated Subjective Global Assessment, BMI = Body Mass Index, AMA = Arm Muscle Area, MNA = Mini Nutritional Assessment,
TST = Triceps Skinfold Thickness, MAMC = Mid arm Muscle Circumference, PINI = Prognostic Inflammatory and Nutritional Index, RBP = Retinol binding protein
38
by nutrition assessment tools (continued)
Author, Assessment Tool used and Cutoff/ Indicator for Prevalence of
N Subjects (Country) Age range (years) Malnutrition (%) Time of Assessment
Year Malnutrition
(McWhirter & Combined method
Patients admitted to a tertiary hospital
Pennington, 500 Not specified BMI < 20 kg/m2 with either 40 Not specified
(United Kingdom)
1994) TST or MAMC below the 15th percentile
Combined method
(Edington et al., Patients admitted to 4 hospitals in > 18 yrs (upper range Within 48 hours of
850 BMI < 20 kg/m2 with either 20
2000) England (United Kingdom) not specified) admission
TST or MAMC below the 15th percentile
Combined method
1) BMI 18.5 (age 1865 y) or 20 (age > 65 y); or
Patients admitted to 57 hospitals (year 2) Unintentional weight loss ( 6 kg in the last 6 months or
(Meijers et al., 8220 to > 18 yrs (upper range Year 2004: 27
2004) and 49 hospitals (year 2007) 3 kg in the last month); or During hospital stay
2009) 11036 not specified) Year 2007: 22
(Netherlands) 3) No nutritional intake for 3 days or reduced intake for >10
days combined with a BMI of 18.520 (age 1865 y) or 20
23 (age >65 y)
TST = Triceps Skinfold Thickness, MAMC = Mid arm Muscle Circumference, BMI = Body Mass Index
39
1.1.3 Risk Factors for Malnutrition
The risk factors for malnutrition include pathological factors, socioeconomic

status, and poor awareness of importance of nutrition or adverse hospitalisation
outcomes (Norman et al., 2008b). Medical conditions can contribute to
malnutrition, mediated by inflammatory processes, increased requirements,
appetite suppression, poor absorption of nutrients or mechanical obstructions
(Campbell, 1999; Norman et al., 2008b; Soeters et al., 2008; Jensen et al., 2010).
Socioeconomic factors such as poor income, lack of family support and isolation
are also risk factors for malnutrition (Pirlich et al., 2005). Lack of awareness on
the part of patients as well as healthcare workers is common, including poor
recognition of malnutrition and monitoring of nutritional status. This can lead to
inaction to prevent or treat early signs of malnutrition (Pennington & McWhirter,
1997). For patients who are frequently admitted to hospital, this situation is further
aggravated by hospital routines which often require a nil by mouth order, or
adverse conditions arising from hospitalisation such as hospital-acquired infection,
poor appetite, side effects of treatment, depression, missed meals due to
procedures, lack of food choices and inadequate feeding assistance (Butterworth,
1994; Incalzi et al., 1996; Sullivan et al., 1999; Weekes et al., 2009). In a review
paper, Kubrak and Jensen grouped factors contributing to malnutrition into two
main categories: personal and organisational (Kubrak & Jensen, 2007). Personal
factors included disease, age, response to treatment, physical, psychological,
social and financial status (Kubrak & Jensen, 2007). Organisational factors
associated with malnutrition included lack of nutrition screening and
documentation, inadequate training of staff, confusion regarding nutritional
responsibility, increased nursing and dietetics workload and lack of adequate
nutritional intake in the hospital (Kubrak & Jensen, 2007). Kubrak and Jensens
broad categorisation into personal and organisational factors is useful, and can in
fact be expanded and sub-categorised further. For example, personal factors can
be further divided into effects of ageing, health (or disease), and social factors.
These are summarised and presented in Figure 1.4.
40
Figure 1.4: Risk factors for malnutrition

HOSPITAL
Nil by mouth, multiple tests
Uninteresting hospital food
Lack of access to food
Repeated admissions to hospital
Restrictive diet
Lack of nutrition screening
Non-compliance to screening
(a) Organisational Inadequate training of staff
High workload
Inadequate monitoring
Lack of follow-up post-discharge
Failure to recognise malnutrition
Risk
factors
for EFFECTS OF AGING
Decreased smell / sensation
malnutrition Early satiety
Disinterest in food
HEALTH
(b) Personal Cognitive Impairment
Depression
Infection, pressure ulcers
Catabolic state secondary to disease
Nausea and vomiting
Malabsorption/
constipation/diarrhoea
Side effects of treatments i.e.
radiation, chemotherapy
Side effects of medications
SOCIAL
Poor caregiver competency/resources
Cultural/religious factors
Lack of family support
Poor eating habits
Low income
Adapted and expanded from Kubrak & Jensen, 2007 Poor understanding of nutrition
by Lim Su Lin, 2012 Low literacy level
Poor self-help skills
41
1.1.4 Consequences, Clinical Outcomes and Cost of Malnutrition
Malnutrition leads to a range of poor clinical outcomes (Braunschweig et al., 2000;

Middleton et al., 2001; Correia & Waitzberg, 2003; Norman et al., 2008b; Charlton
et al., 2012). Table 1.3 presents the impact of malnutrition on mortality, length of
hospital stay, readmission and cost, using different nutrition assessment tools.
Most clinical outcome studies are on the Caucasian population and none have
been done on the Asian population (Table 1.3). Non-Caucasian populations may
show different results due to ethnic differences in genetics, lifestyle, behaviours,
exposures to risk factors, perceptions and values. It is important to know the
impact of malnutrition on the clinical outcomes of patients in different healthcare
settings and populations. Consequences of malnutrition on a wide range of
outcomes are discussed further in the following sections.
Length of Hospital Stay
Malnourished patients stay in hospitals 1.5 to 1.7 times longer than well-nourished
patients (Middleton et al., 2001; Correia & Waitzberg, 2003; Planas et al., 2004;
Kagansky et al., 2005). Possible reasons for malnourished patients longer
hospital stay are closely associated with functional status (Cereda et al., 2008a),
cognitive function (Lang et al., 2006), socioeconomic factors (Aliabadi et al., 2008)
and complications such as infection (Correia & Campos, 2003) and pressure
ulcers (Banks et al., 2010a; Banks et al., 2010b). McWhirter & Pennington
showed that 80% of malnourished patients who did not receive any nutritional
intervention experienced further deterioration in nutritional status in the seven
days following admission (McWhirter & Pennington, 1994).
Readmission to Hospital
Malnourished patients are twice as likely to be readmitted to hospital compared to

well-nourished patients (Planas et al., 2004). In Planas (2004) study, when
42
patients were classified using SGA, there were more total and non-elective
readmissions over the next 6 months in patients with malnutrition (30%) than in
patients without malnutrition (15%). When anthropometric measurements (BMI
and upper arm anthropometry) were used, there were more total readmissions in
the malnourished group, although no significant differences were observed with
the non-elective readmission rate.
Cost
The longer length of hospital stay for malnourished patients is associated with
greater healthcare costs (Braunschweig et al., 2000; Correia & Waitzberg, 2003;
Planas et al., 2004). Malnutrition also results in increased use of hospital
resources (Correia & Waitzberg, 2003), secondary to higher rates of re-
admissions (Planas et al., 2004), increased infection (Braunschweig et al., 2000;
Rypkema et al., 2004), pressure ulcer and poor wound healing (Banks et al.,
2010a; Banks et al., 2010b; Iizaka et al., 2010). As shown by Correia & Waitzberg
(2003), a malnourished patient costs on average 1.7 times more than a well-
nourished patient (US$228 per day vs. US$138 per day) (Correia & Waitzberg,
2003). This represented an increased cost of 60% for malnutrition. When the cost
of medications and tests were added using respiratory patients for comparison,
the costs of malnourished patients increased up to 310% (Correia & Waitzberg,
2003).
Braunschweig et al. (2000) studied 414 inpatients who stayed in hospital for more
than 7 days in an acute care university hospital in the United States. The study
showed that patients who declined nutritionally, regardless of nutritional status on
admission, had significantly higher hospital charges, which was 1.6 times higher
than patients who did not have any decline in nutritional status (Braunschweig et
al., 2000). Banks et al. (2010) conducted random re-samples of 1000 subjects
extrapolated from statistical models to predict the number of pressure ulcer cases,
associated bed days lost and the economic losses in public hospitals in Australia.
43
The model predicted a mean 16,060 bed days lost caused by pressure ulcer
attributable to malnutrition, which equates to AU$12,968,668 in the two-year study
in public hospitals in Queensland, Australia (Banks et al., 2010b). In a recent
study by Freijer et al. (2013), the additional costs of managing adult patients with
disease-related malnutrition were 1.9 billion in 2011, which equates to 2.1% of
national health expenditure in Netherlands (Freijer et al., 2013).
There are potentially enormous cost savings in addressing hospital malnutrition.

One hospital developed a comprehensive nutrition intervention program that
resulted in savings of $2.4 million over a 2-year period due to decreased LOS
(Brugler et al., 1999). The programme consisted of implementation of a nutrition
screening and malnutrition clinical pathway, organised into four stages which
outlined the progression and timing of care, identification of the patient at high risk
of malnutrition, nutrition care decisions, treatment process during hospitalisation
and discharge planning. The savings were estimated at $1,000 for each patient at
high risk of malnutrition (Brugler et al., 1999).
The cost benefit of extending nutrition services to the home is potentially through
shorter length of hospital stay or reduced hospital readmission. Most studies on
the economic benefit of home nutrition services focus on home parenteral nutrition
(HPN) (Baxter et al., 2005; Marshall et al., 2005). A retrospective cohort study of
29 HPN patients reported monthly savings of US$4,860 per patient in comparison
to the patient receiving parenteral nutrition in the hospital (Marshall et al., 2005).
In another retrospective controlled paired study on 56 digestive surgery patients,
the benefits of nutrition therapy (enteral and parenteral) were compared between
a home-hospital model and a traditional hospital model. The patients from the
home-hospital model achieved the same nutritional benefits as those in the
control group, but with expenses 3 times lower (Brazil Reals (R)$3237 vs.
R$8647; p <0.05). The home-hospital model resulted in economic benefit to the
institution, mainly from the avoidance of hospitalisation and the optimisation in the
usage of hospital beds (Baxter et al., 2005). In preventing unnecessary
44
hospitalisation, beyond savings to the hospital, the patient is also spared the
inconvenience, emotional stress, and time and financial costs of hospitalisation.
Functional Status
Functional status represents patients ability to perform activities of daily living

such as bathing, changing, self-feeding, shopping and cooking. Functional status
is usually determined either by interviewing patients using a questionnaire such as
the Barthel Index (Mahoney & Barthel, 1965) or using equipment such as the
handgrip dynamometer (Hunt et al., 1985b). The Barthel Index has been widely
used by physiotherapists and occupational therapists to measure disability and to
evaluate changes in activities of daily living (ADL) function over time especially
during post-stroke therapy (Law & Letts, 1989). The Barthel Index evaluates 10
items related to functional status i.e. bowels, bladder, grooming, toilet use,
feeding, transfer, mobility, dressing, stairs and bathing (Mahoney & Barthel, 1965;
Collin et al., 1988). The total score ranges from 0-20, with lower scores indicating
increased disability (Mahoney & Barthel, 1965).
A simpler method to measure functional status is the hand grip strength which has
gained popularity as a simple, non-invasive measure of muscle strength of upper
extremities and can easily be used in the clinical setting (Norman et al., 2011).
Impaired muscle strength is known to occur in disease-related malnutrition.
Prolonged decreases in nutritional intake lead to loss of body protein from muscle
mass, which in turn results in decreased muscle strength, loss of physical
functionality and poor recovery from illness (Norman et al., 2006). Among all the
methods for measuring functional status, handgrip strength has been the most
commonly used and has been shown to have an inverse association with
malnutrition, and therefore is often used as a good proxy for measuring functional
status (Turnbull & Sinclair, 2002; Thomas et al., 2005; Matos et al., 2007; Norman
et al., 2011). Malnourished hospitalised patients had 25.8% lower hand grip
strength values than well-nourished patients (Norman et al., 2011). In a recent
45
study on 217 patients, it was found that handgrip strength was significantly
correlated with nutritional status using PG-SGA (r = 0.292, p < 0.01). In the same
study, a change in handgrip measurement was an independent predictor of a
change in PG-SGA score (p = 0.04) (Flood et al., 2013). Hence, it is not surprising
that handgrip strength is commonly used as one of the outcome measures for
nutrition intervention in malnourished patients (Paton et al., 2004; Ha et al., 2010;
Norman et al., 2011). For example, Paton et al (2004) found a significant increase
in hand grip strength in malnourished tuberculosis patients after six weeks of
intervention with nutritional supplements when compared to a control group (n=
36, 2.79 3.11 versus. -0.65 4.88, p=0.016) (Paton et al., 2004). A study by Ha
et al. (2010) also found a significant increase in handgrip strength (n = 124,
p=0.002) in stroke patients provided with individualised nutrition support for 3
months compared with a control group (Ha et al., 2010).
Although studies have linked poor handgrip to malnutrition and there is a

suggestion to use it to detect malnutrition or risk of malnutrition (Flood et al.,
2013), handgrip strength cannot be used as a sole marker of malnutrition.
Malnutrition can cause poor handgrip strength but poor handgrip strength may not
necessarily indicate malnutrition, as in the case of stroke or trauma patients. A
recent study found low positive predictive value and specificity when handgrip
strength is used to diagnose malnutrition (positive predictive value = 13%)
(Haverkort et al., 2012). Another limitation of using handgrip strength is that there
are patients who will not be able to grip the dynamometer, for example the
critically ill, stroke patients and patients with severe arthritis of their hands. If
handgrip strength is used to monitor malnourished patients, it should be carried
out together with another established nutrition assessment method e.g. SGA.
Quality of Life
Malnutrition leads to decline in functional status which in turn leads to decreased

quality of life (Norman et al., 2006). There are not many studies examining the
46
direct impact of malnutrition on quality of life (QOL). The few studies showing a
relationship between malnutrition and QOL are disease specific (Norman et al.,
2006) or limited to geriatric populations (Neumann et al., 2005; Rasheed &
Woods, 2013b). In a study on 200 patients with benign gastrointestinal disease,
quality of life was measured using the Medical Outcomes Study 36-item Short-
Form General Health Survey (SF 36) (Ware & Sherbourne, 1992), and found to
be reduced in patients who were classified as malnourished using SGA (Norman
et al., 2006). In this study, malnourished patients suffered significantly impaired
QOL in both the mental and physical domains.
In another study on 133 older adults in rehabilitation, malnourished subjects or

those at risk of malnutrition had poorer QOL than those well-nourished or not at
risk of malnutrition (17 5 versus 19 5, p = 0.008) (Neumann et al., 2005). A
systematic review by Rasheed et al. showed that elderly with malnutrition were
almost three times more likely to experience poor QOL than those who were well
nourished (Odds ratio: 2.85; 95% CI: 2.20-3.70, p<0.001) (Rasheed & Woods,
2013b). As malnutrition leads to poor QOL, improving the nutritional status of
malnourished patients should result in better QOL (Ravasco et al., 2005a;
Ravasco et al., 2005b; Norman et al., 2008a; Rufenacht et al., 2010; Rasheed &
Woods, 2013b).
Pressure Ulcer, Infection and Other Complications
Malnutrition increases the risk of developing infection and pressure ulcers

(Davalos et al., 1996; Banks et al., 2010a). Malnourished patients had at least
twice the odds ratio of having a pressure ulcer compared to well-nourished
patients in public healthcare facilities in Queensland, Australia (Banks et al.,
2010a). This multicenter, cross-sectional audit of 2208 acute and 839 aged care
subjects found that subjects with malnutrition had adjusted odds ratios of 2.6 of
having a pressure ulcer in acute care facilities and 2.0 for residential aged care
facilities. There was also an increased odds ratio of having a pressure ulcer, and
47
having a more severe pressure ulcer (a higher number and/or higher stage
pressure ulcer) with increased severity of malnutrition (Banks et al., 2010a). In
another study of 104 patients admitted for acute stroke, malnourished patients
were more susceptible to urinary tract or respiratory infections (50% versus 24%,
p=.0017) and bedsores (17% versus 4%, p=0.054) than well-nourished patients
(Davalos et al., 1996).
A complication is defined as a condition (disease or accident) which occurs during

hospitalisation in addition to an existing illness (Naber et al., 1997). Pressure
ulcers and infections can be considered complications of hospitalisation if the
patient was not originally admitted with these conditions (iatrogenic conditions).
Complications can be grouped under infectious (such as pneumonia, septicaemia,
wound infection, cystitis) or non-infectious complications (such as intestinal
bleeding, kidney failure, dehydration) (Naber et al., 1997). Two studies have
reported higher complications among patients who were malnourished as
measured by SGA at admission compared to well-nourished patients (Naber et
al., 1997; Braunschweig et al., 2000). Naber et al found a 3-fold increased risk of
infection and all complications in malnourished patients when compared to well-
nourished patients (Naber et al., 1997). Similarly, Braunschweig et al found a 1.4
and 1.8 times increased risk of complications in moderately and severely
malnourished newly admitted patients, respectively (Braunschweig et al., 2000).
In addition, patients who had a decline in nutritional status whilst hospitalised
experienced more complications than those who did not (Braunschweig et al.,
2000). The author suggested that clinicians should focus on reducing declines in
patients' nutritional status as a priority regardless of patients' nutritional status on
admission (Braunschweig et al., 2004).
48
Mortality
Besides increased risk of morbidity and cost, hospitalised patients with poor
nutritional status face an increased risk of mortality (Middleton et al., 2001;
Correia & Waitzberg, 2003; Kagansky et al., 2005). Studies have shown that
patients with malnutrition have a 1.6-1.9 relative risk of death when compared to
well-nourished inpatients (Middleton et al., 2001; Correia & Waitzberg, 2003;
Kagansky et al., 2005). There has only been one study to prospectively
investigate mortality outcomes of malnutrition using data from national death
registers (Middleton et al., 2001). This Australian study revealed that mortality at
12 months was 29.7% in malnourished subjects compared with 10.1% in well-
nourished subjects (p<0.0005) (Middleton et al., 2001). The mortality data was
obtained either from hospital records or from the New South Wales Registry of
Births Deaths & Marriages by requesting a copy of death certificates for all
subjects involved in the study whose survival status was unknown (Middleton et
al., 2001). Obtaining mortality data from the national death registry is an accurate
way to ensure information is captured on subjects who have passed away after
being discharged from the hospital.
So far, all the studies which have found a relationship between malnutrition and
increased mortality in adult patients did not control for illness and other
confounders except for Correia & Waitzbergs study which took into account the
presence of infection, cancer and age (Correia & Waitzberg, 2003). This will be
discussed in the next section.
49
Table 1.3: Impact of malnutrition on mortality, length of hospital stay (LOS), readmission and cost, using different nutrition assessment tools and
evidence of them being controlled for confounders
Assessment Time of Prognostic Indicators
Participants, Duration of
Author, Tool and Assessment Controlled for
Country and n Mortality outcomes
year Indicator for and LOS Readmission Cost confounders
Type of Study /Survival tracking
Malnutrition Administrator
SUBJECTIVE GLOBAL ASSESSMENT
Well-
Malnourished: Presence of
Adult inpatients nourished:
Survival in 16.7 + 24.5 cancer and
randomly Within 72 US$138.00
hospital: days, median of infection, age During
(Correia & selected from 25 hours of per patient
SGA: Ratings Malnourished: 9 days, Well over 60 years hospital
Waitzberg, Brazilian admission, 709 Not Specified Malnourished:
B&C 212 (87.6%) nourished: 10.1 and those admission
2003) Hospital, Brazil, administrator US$228.00
Well nourished: + 11.7 days, undergoing
Retrospective not specified per patient
444 (95.3%) median of 6 clinical
Cohort Study (Increase of
days treatment
60.5%)
Malnourished:
Inpatients in two Mortality at 12
Median LOS =
(Middleton Sydney teaching During months:
17 days, Well
et al., hospitals, SGA: Ratings hospital stay, Malnourished =
819 nourished: Not Specified Not Specified Not Specified 1 year
2001) Australia, B&C administrator 29.7%
Median LOS =
Prospective not specified Well nourished =
11 days
Study 10.1%,
(p<0.0005)
Adult patients
Difference in Malnourished:
with cancer Within 30 days
mortality within Median LOS =
admitted to an of discharge:
30 days of 13 days
(Bauer et acute care SGA: Ratings SGA C: 17%,
Not Specified 71 discharge, Well nourished: Not Specified Not Specified 1 month
al., 2002) hospital, B&C SGA B: 52%
between SGA Median LOS =
Australia, SGA A: 59%
groups (p=0.305 7 days
Prospective (p= 0.037)
= NS) (p=0.024)
Study
Colorectal
During
cancer patients Median Survival
consultation
(Gupta et (Stages III & IV), SGA C: 6 mths
SGA: Ratings with a
al., 2005) United States, 234 SGA B: 8.8 mths Not Specified Not Specified Not Specified Not Specified 74 months
B&C dietitian,
Retrospective SGA A: 12.8
administrator
Epidemiologic mths
not specified
Study
SGA = Subjective Global Assessment, LOS = Length of hospital stay, mths = months
50
evidence of them being controlled for confounders (continued)
Participants, Assessment Time of Prognostic Indicators Duration
Controlled
Country and Tool and Assessment of
Lead Author n Mortality for
Type of Indicator for and LOS Readmission Cost confounders outcomes
Study Malnutrition Administrator /Survival tracking
SUBJECTIVE GLOBAL ASSESSMENT (continued)
Readmission
Hospitalised patients SGA: Undernutrition:
over 6 months:
to a university Ratings Within 48 hours 7.5 + 5.4 days,
(Planas et Undernutrition Not
hospital, B & C of admission by 400 Not Specified Normonutrition: Not Specified 6 months
al., 2004) 30% vs. Specified
Spain, a single dietitian 5.0 + 5.1 days
Normonutrition
Prospective study (p<0.05)
15% (p < 0.05)
SGA C:
Time of In-hospital
34.8 27.7 days
Inpatients with Acute assessment not mortality
SGA: SGA B: 30 days
(Fiaccadori Renal Failure, specified. SGA Not
Ratings 309 35.1 29.9 days Not Specified Not specified post
et al., 1999) Italy, assessed by 2 SGA C: 62% Specified
B&C SGA A: discharge
Prospective study trained SGA B: 20%
23.5 14.6 days
administrators SGA A: 18%
p< 0.01)
60-day or Postoperative LOS
SGA:
perioperative SGA Severe:
Mild,
mortality 16 9 days
Preoperative liver Moderat At the time of
SGA Moderate:
(Stephenson transplant patients, e or transplantation Not
99 SGA Severe: 10 5 days, Not Specified Not specified 60 days
et al., 2001) United States, Severe by a single Specified
= 15.6% (p = 0.0027);
Retrospective study Malnutri observer
SGA Mild to SGA Mild:
tion
moderate: 9 8 days,
3.0% (p = 0.10) (p = 0.0006)
Hospitalised patients Malnourished:
admitted < 2 weeks SGA: 13.3 10.5 days
(Gout et al., Not During
to a university Ratings Not specified 275 Not specified Well-nourished Not Specified Not specified
2009) Specified admission
hospital, Australia, B & C 8.8 8.8 days
Retrospective Study (p< 0.05)
SGA = Subjective Global Assessment, LOS = Length of hospital stay
51
Participants, Controlled Duration of
Lead Tool and Assessment
Country and Type n Mortality for outcomes
Author Indicator for and LOS Readmission Cost confounders tracking
of Study /Survival
SUBJECTIVE GLOBAL ASSESSMENT (Modified versions)
Patients
A 1 unit lower
commencing At the 7-point SGA
(Churchill SGA score
continuous 7-point SGA: commenceme Malnourished: 96
et al., was associated
peritoneal Ratings 1 to nt of days
1996) with a 25% Not Up to 2
dialysis in 14 5 peritoneal 680 Well nourished: Not Specified Not Specified
increase in the RR Specified years
centers in Canada dialysis, 13 days
of death
and United administrator
States, not specified
Prospective study
Within 48
Patients admitted Inpatient mortality: Malnourished = 8
hours of
(Martineau to an Acute Stroke PG-SGA: Malnourished = days,
admission, Not During
et al., Unit, Ratings B & 73 14%, Well nourished = Not specified Not specified
administrator specified admission
2005) Australia, C Well nourished = 3 days
was a single
Prospective study 2% (p <0.001)
dietitian
(p <0.092, NS)
(Thomas Within 48
Patients admitted Malnourished = 5
et al., hours of
2007) to an Acute PG-SGA: days,
admission, Not During
Assessment Unit, Ratings B & 64 Not specified Well nourished = Not Specified Not Specified
administrators Specified admission
Australia, C 4 days
were 2
Prospective study (p =NS)
dietitians
PG-SGA = Patient-Generated Subjective Global Assessment, LOS = Length of hospital stay, NS = Non significant
52
Participants, Duration of
Lead Tool and Assessment Controlled for
Country and n Mortality outcomes
Author Indicator for and LOS Readmission Cost confounders
Type of Study /Survival tracking
MINI NUTRITIONAL ASSESSMENT
LOS during the

Inpatient 2.7 years:
Hospitalised mortality: Malnourished =
MNA:
(Kagansky elderly patients Malnourished 59.9 77.0
At risk score Time and
et al., aged > 75 years = 38.7%, days, Well
= 17-23.5, administrator 414 Not Specified Not Specified Not Specified 2.7 years
2005) old in geriatric unit, Well nourished =
Malnourished not specified
Israel, nourished = 28.3 27.6
< 17
Prospective study 12.5% days
(p < 0.001) (p < 0.001)
Median LOS Number of

Hospitalised Major Disease
during the 1.5 hospital
elderly patients Classification
MNA: years: readmissions
(>65 years old) Mortality at admission,
At risk score Within 72 Malnourished = was not
(Charlton admitted to 2 sub- hazard rate = age, sex, 12 26
= 17-23.5 hours of 476 34 days, At risk associated Not Specified
et al., acute hospitals, 3.41 mobility and months
Malnourished admission, = 26 days with MNA
2012) Australia, (p = 0.038) LOS at
score < 17) Well nourished score
Retrospective index
= 20 days (r = -0.004,
Study admission
(p < 0.001) p = 0.45)
ANTHROPOMETRY
Patients admitted Within 48 BMI not a
(Martineau to an BMI: Cutoff hours of significant
During
et al., acute stroke unit, not specified admission by 73 Not specified predictor of Not specified Not Specified Not Specified
admission
2005) Australia, a single LOS
Prospective study dietitian
MNA = Mini Nutritional Assessment, LOS = Length of hospital stay, BMI = Body Mass Index, NS = Non significant
53
Assessment Time of Prognostic Indicators Duration
Participants,
Lead Tool and Assessment Controlled for of
Country and n
Author Indicator for and confounders outcomes
Type of Study Mortality /Survival LOS Readmission Cost
Malnutrition Administrator tracking
ANTHROPOMETRY (continued)
Chronic disease,
Hospitalised
MAC During Mean MAC = 28.4cm age, drugs,
(Gariballa patients aged
(cutoff for hospitalization (alive) vs. 26.2cm Not functional
& Forster, 65 years old. 445 Not Specified Not Specified 1 year
malnutrition by a single (dead). Specified capacity and
2007) United Kingdom,
not specified) observer acute-phase
Prospective Study
response
Within 4 days
(Neumann Older adults in CAMA: 133 At risk: 15
of admission, Assessment of During
et al., rehabilitation unit, Males < 21.4 adults days, Not at Not
2 administrator Not Specified Not Specified Quality of Life on admission
2005) Australia, cm , females > 65 risk: 15 days Specified
2 was 1 staff admission (for LOS)
Prospective Study < 21.6 cm years (p = NS)
nurse
Mortality at 8 years:
CAMA: Malnourished: HR = Age, gender,
Males < 21.4 1.94 marital status,
2
cm , females Upon Well nourished HR = smoking, self-
Older adults in the 2 1799
< 21.6 cm enrollment, 1.00 rated health,
(Miller et community, adults Not
administrators (p = 0.003) Not Specified Not Specified activities of daily 8 years
al., 2002) Australia, > 70 Specified
were trained Mortality at 8 years: living,
Prospective study years
Body Mass observers Malnourished: HR = comorbidity,
Index: < 20 1.36 cognition,
2
kg/m Well nourished: HR = depression
1.00 (p = NS)
BMI < 18.5
2 Undernutrition
Hospitalised kg/m or Within 48 Undernutrition:
30.7% vs.
patients to a BMI = 18.5 - hours of 7.8 + 7.2 days,
2 Normonutrition Not
university hospital, 20 kg/m and admission by 400 Not Specified Normonutrition Not Specified 6 months
(Planas et 20.7% Specified
Spain, TST or a single : 6.2 + 8.0
al., 2004) th (p = NS)
Prospective study MAMC < 15 dietitian days
percentile (p = NS)
SGA = Subjective Global Assessment, LOS = Length of hospital stay, BMI = Body Mass Index, MAC = Mid arm circumference, MAMC = Mid arm muscle
circumference, TST = Triceps skinfold thickness, CAMC = Corrected arm muscle area, HR = Hazard ratio, OR = Odds ratio, NS = Non significant
54
Table 1.3: Impact of malnutrition on mortality, length of hospital stay (LOS), readmission and cost, using different nutritional assessment tools and
Participants, Controlled Duration of
Lead Tool and Assessment
Country and n for outcomes
Author Indicator for and Mortality
Type of Study LOS Readmission Cost confounders tracking
Malnutrition Administrator /Survival
BIOCHEMICAL
Adjusted for
chronic
Hospitalised
Albumin During Mean albumin disease, age,
(Gariballa patients aged
(cutoff for hospitalization = 38 g/L drugs,
& Forster, 65 years old, 445 Not Specified Not Specified Not Specified 1 year
malnutrition by a single (alive) vs. 36 functional
2007) United Kingdom,
not specified) observer g/L (dead). capacity and
Prospective Study
acute-phase
response
Patients
A 1 g/L lower
commencing At the
serum albumin
continuous commenceme Albumin
concentration
(Churchill peritoneal Albumin nt of < 35g/L:
was Up to 2
et al., dialysis in 14 (cutoff for peritoneal 680 59 days Not Specified Not Specified Not Specified
associated years
1996) centers in Canada malnutrition dialysis, 35g/L:
with an
and United not specified) administrator 20 days
6% increase in
States, not specified
the RR of
Prospective study
death.
Patients admitted Within 48 No association

(Martineau to an hours of between serum
During
et al., acute stroke unit, Albumin: < admission by 73 Not Specified albumin and Not specified Not Specified Not Specified
admission
2005) Australia, 35g/L a single LOS (r =0.:013,
Prospective study dietitian p = 0:893)
SGA = Subjective Global Assessment, MNA = Mini Nutritional Assessment, LOS = Length of hospital stay, RR = Relative risk, NS = non-significant
55
1.1.5 Potential Confounders for Malnutrition Outcomes

malnutrition and clinical outcomes, the confounding effects of age, gender,
disease and its complexity have seldom been taken into consideration (refer to
Table 1.3).
Previous studies have shown that malnutrition is associated with advancing age
(Rauscher, 1993; Middleton et al., 2001). As people age, they tend to have more
illness, more admissions to hospital and potentially longer lengths of hospital stay.
Hence, in any clinical outcome study, it is important to control the results for age.
Despite this, many studies on malnutrition outcomes did not control for this
(Middleton et al., 2001; Planas et al., 2004; Gupta et al., 2005; Thomas et al.,
2007; Gout et al., 2009). Similarly, outcomes can be affected by gender. For
example, Pirlich (2006) showed that hospitalised females are more likely to be
malnourished than males (Pirlich et al., 2006).
Prospective descriptive studies on the outcomes of malnutrition which control for

the nature and severity of medical condition in adult hospitalised patients are
scarce and none have controlled for the complexity and severity of disease. One
study on adult hospitalised patients controlled for medical conditions but disease
severity was not taken into consideration. In a retrospective study on 709 newly
hospitalised patients in Brazil, Correia & Waitzberg (2003) controlled for
presence of cancer and infection, age over 60 years and those undergoing
clinical treatment. However controlling for just two medical conditions such as
cancer and infection is grossly inadequate as there are different stages of cancer
and severity of infection, as well as many other medical conditions which may
affect patient outcomes. Three other studies have focused on elderly patients,
which controlled for the confounders of age, gender and disease (Miller et al.,
2002; Gariballa & Forster, 2007; Charlton et al., 2012). The study by Miller (2002)
on 1799 older adults (75 years old) in the Australian community controlled for
56
age, gender, marital status, smoking, self-rated health, activities of daily living,
comorbidity, cognition and depression (Miller et al., 2002). The authors controlled
for disease using the 10 most prevalent comorbid conditions which are cancer,
arthritis, heart condition, heart attack, hypertension, ulcers, diabetes mellitus,
respiratory disease, hernia and stroke (Miller et al., 2002). Severity of disease
was also not taken into consideration in this study. The other two studies were
also on elderly65
( years old) and despite controlling for disease, did not
mention about severity of disease (Gariballa & Forster, 2007; Charlton et al.,
2012).
Controlling an outcome for disease or medical specialty alone is inadequate as

there are different levels of severity and complexities within each disease. For
example, Stage 4 cancer cases differ in complexity and outcome in comparison
to Stage 1 cancer. The diagnoses and complexities of disease based on the
resources used can actually be determined using DRG (Robinson et al., 1987).
The DRG is a system widely used by many countries to cluster patients with a
variety of diagnoses and procedures into diagnostic groups on the basis that
cases within a group will have a similar level of complexity and their treatment is
expected to utilise a similar level of hospital resources and hence incur similar
costs (Thompson, 1988). They are used as a basis for hospital funding allocation
(Thompson, 1988). Each DRG has a specific numerical code as assigned by the
Australian National Diagnosis-Related Groups (Commonwealth Department of
Health and Family Services and 3M Health Information Systems, 1996) or similar
groups. By nature of its widespread adoption across many countries, albeit with
some variation and customization specific to each country, DRG is a convenient
way to collect and compare data of patients with similar disease diagnoses and
complications, to study the effect or association of disease on various end-points
of interest. In order to show that the outcomes of malnutrition are independent of
the underlying disease and its severities, it is possible to adjust the results for
57
DRG. So far no studies have used DRG to control for the confounding effect of
diagnosis on malnutrition outcomes in a multidisciplinary setting.
Besides DRG, there are a variety of other methods which can be used to control
for disease and its severity or complexities such as the Charlson Comorbidity
Index (CCI) (Charlson et al., 1987), Cumulative Illness Rating Scale (CIRS) (Linn
et al., 1968), Index of Coexisting Disease (ICED) (Cleary et al., 1991), Burden of
Disease (BOD) index (Mulrow et al., 1994), Kaplan Index (Kaplan & Feinstein,
1974) and Incalzi Index (Incalzi et al., 1997). All these methods are labour-
intensive, and require clinical assessments or review of patients clinical case-
notes to provide a summative weightage of 13-59 items. In addition, the tools
contain limited number of disease categories resulting in some conditions not
being able to be scored, for example the CCI does not list hip fracture, short
bowel syndrome and infection as comorbidities, among others. Kaplan Index was
developed specifically for people with diabetes, and the BOD and Incalzi Index
for the elderly. Despite the limitations of various comorbidity tools, Baldwin et al.
(2006) in a study on four comorbidity methods reported that each is fairly robust
in predicting outcomes, although some comorbidity measures have minor
advantages over others. Therefore, investigators should select a suitable method
based on its availability, staff comfort with the methodology, and outcomes of
interest (Baldwin et al., 2006). Again, no studies have used measures of
comorbidities tools to control for the outcomes of malnutrition.
58
Summary of Issues on Clinical Outcomes of Malnutrition
In summary, the adverse outcomes of malnutrition in hospitalised patients are

well-established. Although some studies controlled for effect of disease on
outcomes, there is no published study on the consequences of malnutrition that
has adjusted for disease type and complexity in adult hospitalised patients. It is
important to determine the cost and clinical outcomes arising from malnutrition
and more so to control the outcomes for the confounders that have been
discussed in this section. If healthcare institutions and professionals are
convinced that malnutrition has an impact on clinical outcomes and costs
independent of medical diagnosis, they are more likely to proactively prevent and
manage malnutrition. A prospective study that controls for all the confounders
mentioned above will provide a strong evidence base from which to advocate for
additional resources if it does indeed show an independent association between
malnutrition and outcomes.
59
1.2 NUTRITION SCREENING
Given that malnutrition is commonly unidentified and untreated, (McWhirter &

Pennington, 1994) and increases morbidity and mortality risk, it is important to
identify patients who are at risk and make an accurate diagnosis for these
patients so that early nutrition intervention can be administered. A number of
methods have been developed to both identify risk and/or to diagnose
malnutrition (Guigoz et al., 1996; Lyne & Prowse, 1999). Nutrition screening and
nutrition assessment are typically used for these purposes; however the
delineation between them should be clearly set (Charney, 2008).
Nutrition screening is a process of efficiently identifying characteristics known to

be associated with malnutrition risk (American Dietetic Association., 1994;
American Society for Parenteral and Enteral Nutrition., 2012). Skipper et al.
(2012) defined nutrition screening as a process to identify patients, clients, or
groups who may have a nutrition diagnosis and benefit from nutrition assessment
and intervention by a registered dietitian (Skipper et al., 2012). Its purpose is to
identify those at risk of malnutrition to facilitate nutrition assessment and early
delivery of nutritional intervention (American Dietetic Association., 1994;
American Society for Parenteral and Enteral Nutrition., 2012). A key aspect of
screening is that further information is required to make a diagnosis or treatment
decision (Charney, 2008). However, there is some controversy regarding ideal
nutrition screening, including choice of screening tool, who should administer
screening, timing of screening and cutoff points to define a positive screen
(Charney, 2008). As nutrition screening is to be performed on a large number of
patients, typically all newly admitted patients, the tool has to be simple, quick,
reliable, valid and cost effective (Ferguson et al., 1999a; Charney, 2008).
Acceptable nutrition screening tools are those with a high level of validity,
reliability, efficiency and cost-effectiveness (Lyne & Prowse, 1999; Charney,
2008). In 2006, a group of dietitians from United States, Europe, Australia, and
the Middle East developed a set of guidelines for nutrition screening which
60
stated: (1) it can be used in any practice setting, (2) it should be quick, easy to
use, valid, and reliable for the patient population or setting, (3) the tool and
content should be developed by dietitians and (4) it should be conducted within
an appropriate time frame for the setting (Skipper et al., 2012).
It is recommended that hospitalised patients be screened for nutritional risk

within 24 hours of admission using a nutrition screening tool (Joint Commission
International, 2008). Data collected for nutrition screening should be easily
obtained to minimise incompletion (Charney, 2008). Parameters usually included
in nutrition screening are weight change, adequacy of oral intake and nutrition-
focused physical examinations (Charney, 2008).
There are a number of established nutrition screening tools, including the

Malnutrition Universal Screening Tool (MUST) (Stratton et al., 2004), Nutrition
Risk Index (NRIa) (Veterans Affairs Total Parenteral Nutrition Cooperative Study
Group, 1991) and Malnutrition Screening Tool (MST) (Ferguson et al., 1999a).
Table 1.4 presents the components, strengths and limitations of nutrition
screening tools developed for hospitalised patients. Common limitations of
current tools are high levels of missing data (for BMI, weight and biochemical
data), ambiguity (for appetite, nutritional intake, pressure ulcers and disease),
length of questionnaire, time-consuming to administer, no cutoff for referral to
dietitians and no option for patients who are unsure of their weight loss.
61
Table 1.4: Components, strengths and limitations of nutrition screening

tools for hospitalised patients (in alphabetical order)
Tools Components Strengths Limitations
Derby Nutritional a) Body weight for height Non-invasive Missing data for patients unfit
b) Mobility/capability to be weighed and measured
Score (Goudge et c) Gastrointestinal (GI) Symptoms Indicates score for for height
al., 1998) d) Skin type action plan
e) Appetite and dietary No mention of length of
f) Intake period for GI symptoms or
g) Psychological state poor appetite
h) Age
Questionnaire of 7 items may
be deemed too lengthy for
busy staff
Malnutrition a) Lost weight recently without trying Easy to use Malnutrition cutoffs may be
b) Quantity of weight loss Option of unsure for confounded by ethnicity/
Screening Tool c) Eating poorly because of decreased patient who are not population
(MST) appetite sure of weight loss
(Ferguson et al., Quick
Non-invasive
1999a)
Indicates cutoff score
for referral process
Malnutrition a) BMI Easy to use Missing data for patients unfit
b) % Unplanned weight loss in 3-6 Non-invasive to be weighed and measured
Universal months Indicates cutoff score for height
Screening Tool c) No or likely to be no nutritional intake for referral process
(MUST) for >5 days Directs clinicians in Many nurses do not carry
nutrition care plan calculator to calculate BMI
(Stratton et al.,
(reference to BMI charts an
2004) extra step)
Results inaccurate for

subjects with fluid retention or
dehydration
High incomplete screening
No option for patients who are

unsure of weight loss
Malnutrition cutoffs may be

confounded by ethnicity/
population
Mini Nutritional a) Loss of appetite Specifically designed Limitations of BMI as
b) Weight loss in the past 3 months for use in the elderly mentioned above
Assessment-Short c) Mobility community population
Form (MNA-SF) d) Psychological stress or acute Limited to the use in the
(Rubenstein et al., disease in the past 3 months elderly and community
e) Neuropsychological problems
1999)
f) Body Mass Index Need to use a different tool
for younger patients
Nutrition Risk Index a) Serum albumin Objective parameter Results are inaccurate as
b) Current weight albumin is easily affected by
(NRIa) c) Usual weight inflammation, fluid retention
(Veterans Affairs Indicates cutoff score or dehydration
Total Parenteral for referral process
Missing data for patients unfit
Nutrition
to be weigh or measured for
Cooperative Study height
Group, 1991)
Does not direct clinicians in
nutrition care plan
Validation of nutrition screening tools is shown in Table 1.6
62

tools for hospitalised patients (continued)
Nutritional Risk a) Dentures One of the pioneer Lengthy 16-items question
b) Prescriptive medications nutrition screening tools
Index c) Non-prescriptive medications Time-consuming to
(NRIb) d) Abdominal operation Comprehensive complete
(Wolinsky et al., e) Bowel problems
f) Food intolerance Non-invasive Limited to the use in
1985)
g) Mastication problems elderly
h) Conditions affecting food intake Indicates cutoff score for
i) Smoking nutritional risk
j) Special diet
k) Anemia
l) Stomach or abdominal pain
m) Illness leading to loss of appetite
n) Swallowing problems
o) Vomiting
p) Weight changes
Nursing Nutritional a) BMI Non-invasive Lengthy 7-items question

b) Appetite
Assessment Tool c) Swallowing, Indicates cutoff score for Time-consuming to
(Scanlan et al., d) Diet action plan and referral to complete
1994) e) Weight change dietitian
f) Medical/physical condition Limitations for BMI as
g) Postoperative status mentioned above
Nutrition Risk a) Weight loss in the last 3 months Non-invasive No cutoff score within the
b) BMI tool for referral process
Score c) Appetite One single tool for use in
(NRS) d) Ability to eat or retain food both paediatric and adult Limitations for BMI as
(Reilly et al., 1995) e) Stress factor patients mentioned above
Results may not truly

indicate malnutrition as it is
marred by severity of
illness
Nutrition Risk a) Ideal body weight Non-invasive Missing data for patients
b) Weight loss history unfit to be measured for
Classification c) Alterations in dietary intake height to calculate ideal
(Kovacevich et al., d) Gastrointestinal function body weight
1997)
Many nurses do not carry
calculator to calculate ideal
body weight (reference to
charts an extra step)
Nutrition Risk a) Weight loss Non-invasive Lengthy questionnaire and

b) Appetite action plans
Check c) Ability to eat
(Rawlinson, 1998) d) Gut function Time-consuming to
e) Medical condition Indicates staggered cutoff complete
f) Pressure ulcer assessment score for action plan and
referral to dietitian Limited details provided on
pressure ulcer assessment
Nurses may not know how

to score other medical
conditions not listed are
present
Validation of nutrition screening tools shown in Table 1.6
63

Nutritional Screening Tool a) Body weight for height Non-invasive Limitations of BMI as
b) Skin type mentioned above
(Scott & Hamilton, 1998) c) Appetite and dietary intake
d) Ability to eat Takes time to
e) Symptoms complete
f) Psychological state
g) Age Skin type
assessment is
subjective and vague
Nutrition Assessment Score a) Appetite Non-invasive Nurses may not
b) Weight loss, know how to score
(Oakley & Hill, 2000) c) General condition Indicates score for other medical
d) Clinical state action plan conditions not listed
e) Increased requirements because of are present i.e. renal
disease states failure, hip fracture,
trauma.
Nutrition Screening Tool a) Age Non-invasive Patient with co-
b) Mental condition morbidities (mental
(Burden et al., 2001) c) Weight status Indicates score for status, medical
d) Dietary intake action plan condition and gut-
e) Ability to eat function) may be
f) Medical condition over-rated. Hence,
g) Gut function results may not truly
indicate malnutrition
risk
Patient on tube
feeding but receiving
adequate nutrition
are scored poorly
Nutritional Risk Screening a) BMI Non-invasive Some staff may find
b) Weight loss in the last 3 months it complicated as it
2002 (NRS 2002) c) Reduced dietary intake in the last Indicates cutoff score has 2 sections
(Kondrup et al., 2003b) week for referral process
d) Severity of illness Limitations for BMI
e) Age Directs clinicians in as mentioned above
nutrition care plan
No option for patients
who are unsure of
weight loss
Results may not truly

indicate malnutrition
as it is marred by
severity of illness
Nutrition Screening Tool a) Unintentional weight loss in the last 6 Non-invasive No option for patients
months who are unsure of
(Weekes et al., 2004) b) Unintentionally eating less in the last Indicates cutoff score weight loss
6 months for action plan
c) Nil by mouth or unable to eat > 5 days Confusing and
complicated if the
With option for the nurse to complete nurses also need to
BMI and mid arm circumference (MAC) complete BMI and
MAC
64

Screening Nutritional Twelve-item questionnaire to be Can be used for The 3 step procedure
completed by admission staff / both adults and requires the admission staff,
Profile patient followed by anthropometric children nurses and dietitians to
(Hunt et al., 1985a) by nurses and albumin by dietitian: complete the form.
a) Weight loss Invasive and costly

b) Appetite requires blood test
c) Vomiting / Diarrhoea
d) Medical condition Albumin is easily affected by
e) Surgery inflammation, fluid retention
f) Fever or dehydration
g) Diet restrictions
h) Medications High rate of missing data for
i) Weight weight, height and albumin
j) Height
k) Head circumference (children)
l) Albumin
Screening Sheet a) BMI Non-invasive Takes time to complete

b) Weight loss
(Thorsdottir et al., 1999) c) Age Tedious due to need for
d) Symptoms information on surgery type
e) Length of hospital stay, and length of hospital stay in
f) Surgery type the past 2 months
Limitations of BMI as
mentioned above
Geared toward surgical

patients
Cutoff score for action not

stated
Short Nutritional a) Unintentional weight loss in 6 Easy to use Confusion on whether the tool
months is used for identifying risk or
Assessment b) Decreased appetite in 1 month Quick diagnosing malnutrition
Questionnaire c) Use of supplemental drinks or
(SNAQ) tube feeding over the last month Non-invasive No option for patients who are
unsure of weight loss
(Kruizenga et al., 2005a)
Indicates cutoff
score for referral Patient on supplemental drink
process or tube feeding but receiving
adequate nutrition are scored
poorly
Simple Screening Tool a) BMI Simple 2 question Confusing due to the two in
b) Weight loss (%) per screening one screening tool
(Laporte et al., 2001a)
OR Albumin is easily affected by
inflammation, fluid retention
a) BMI or dehydration
b) Albumin
Limitations of BMI as
mentioned above
65
1.2.1 Practical Considerations in Nutrition Screening
Nutrition screening must be practical to be of use in clinical practice. Practical

considerations before implementing a screening tool are discussed in the
following sections.
User-friendliness of nutrition screening tool
The ease of use of a nutrition screening tool is important. The tool must be user-
friendly in order to encourage compliance and completion of screening. The
lengthy questionnaires of Nutritional Risk Index (NRIb) (Wolinsky et al., 1985) and
Screening Nutritional Profile (Hunt et al., 1985a) decrease the applicability of the
tool, especially in the busy healthcare setting. Nutritional Risk Index contains 16
questions which incorporates nutritional intake, medications, smoking habits,
illness or medical procedures affecting food intake, changes in eating habits and
weight changes (Wolinsky et al., 1985; Wolinsky et al., 1986). Screening
Nutritional Profile consists of a twelve-item questionnaire to be completed by
admission staff, followed by anthropometric measurement by nurses and review
of albumin levels by dietitians (Hunt et al., 1985a).
Nutrition screening ideally should use easily obtainable data to increase

completion rate. The Nutrition Risk Index (NRIa) assesses nutritional risk using
serum albumin concentration and percentage weight loss from usual weight
(Veterans Affairs Total Parenteral Nutrition Cooperative Study Group, 1991). The
use of albumin level is potentially problematic, as not all inpatients have their
albumin levels routinely checked. In a Singapore study by Lim et al. (2009), only
46% of patients had their serum albumin tested within the first two days of
admission (Lim et al., 2009). If this is not included as part of routine medical care,
obtaining a blood sample for the purpose of nutrition screening could be
considered invasive and expensive (Lim et al., 2009). In addition, serum albumin
is confounded by inflammation (de Mutsert et al., 2009) and disease severity in
66
hospitalised patients and may be inappropriate as a measure of nutritional risk or

status per se (Klein, 1990; Haupt et al., 1999; Shenkin, 2006). In a prospective
study of 225 pre-surgery patients, abnormal concentrations of acute phase
proteins (indicating an acute phase response) were detected in 19% of patients.
The mean albumin concentration in these patients (35g/L) was lower than that of
patients who did not mount an acute phase response preoperatively (40g/L)
(Haupt et al., 1999). This study shows that a metabolic response to disease,
referred to as an acute phase response, can lead to low albumin concentrations
and hence confound the use of albumin to define nutritional risk or status (Haupt
et al., 1999). Therefore it can also lead to over-diagnosis of malnutrition
(Rosenthal et al., 1998).
Another parameter that is often used in nutrition screening tools but is difficult to
obtain is the BMI. There are many nutrition screening tools that require patients
weight and height or knee height to be measured to calculate BMI. Examples of
screening tools that contain BMI as one of their components include the NRS
(Reilly et al., 1995), MUST (Elia, 2003) and NRS 2002 (Kondrup et al., 2003a).
Weight and height measurements required in BMI pose challenges in patients
who are bed bound or old and frail. In an audit done on 526 hospital episodes,
only 41% had information on both weight and height (Campbell et al., 2002). An
audit in three English hospitals on the use of the MUST nutrition screening tool,
which includes BMI as one of three screening criteria, reported that one-third of
patients remained unscreened, even after specific training of clinical staff to
increase screening completions (Wong & Gandy, 2008; Porter et al., 2009). Even
in the research arena, where missing data has to be minimised, many
researchers faced difficulties in obtaining complete BMI data for their studies
(Kelly et al., 2000; Waitzberg et al., 2001; Correia & Campos, 2003; Wyszynski et
al., 2003). In one of these studies, BMI was not available in 35% of the subjects
even when the researchers actively tried to measure the weight and height
themselves (Kelly et al., 2000). In clinical practice, unavailable weight, height and
BMI records can be as high as 74-85% (Waitzberg et al., 2001; Correia &
67
Campos, 2003; Wyszynski et al., 2003; Neelemaat et al., 2011b). The

measurement challenges discussed above limit the utility of nutrition screening
tools which combine BMI with other parameters. This further strengthens the
argument that BMI is not a practical parameter for nutrition screening.
Realising the challenges of obtaining patients BMI, a few authors have come up
with alternative anthropometric measures or subjective assessment (Stratton et
al., 2006; Kaiser et al., 2009; Kaiser et al., 2011). For example, if weight could
not be measured in MUST, the author recommended the use of recalled weight.
If neither weight nor height could be obtained, subjective assessments of
physical appearance (very thin, thin etc.), were employed (Stratton et al., 2006).
Given the difficulties in obtaining weight and height data for many patients, the
use of surrogate anthropometric markers is an advantage for MUST. However,
the reliability of these methods to estimate BMI or physical appearance has not
been validated and their accuracy remains questionable. An alternative to
obtaining the weight and height of patients not able to sit or stand is to use the
bed weighing machine or the knee height measurement (Stratton et al., 2006;
Cereda et al., 2007). However each bed weighing machine costs about S$10,000
and not many hospitals can afford to purchase one for each of their wards. In
addition, these patients need to be transferred from their hospital bed to the
weighing bed and back, creating extra demand for resources and manpower.
Low nurse to patient ratios in Asian hospitals (specifically 1 nurse to 7 patients in
NUH in the year 2002) are a further barrier to such measurements. Therefore,
nutrition screening tools which require these measurements may not be suitable
in hospitals with low nurse to patient ratios.
Cutoff for risk of malnutrition
Body Mass Index (BMI) has often been used as a screening parameter for a
range of medical conditions (Foucan et al., 2002; Iseki et al., 2004; Pua & Ong,
2004; Norberg et al., 2006; Lloyd-Jones et al., 2007; Onalan et al., 2009; Sanada
68
et al., 2012). Although a high BMI has much supporting evidence for use as a
marker of cardiovascular disease risk such as heart disease, hypertension and
diabetes (Sung & Ryu, 2004; Hoad et al., 2010; Sanada et al., 2012), there is no
established cutoff point of the lower end of BMI to signify risk of malnutrition.
Body Mass Index and its associated cutoffs may be confounded by ethnicity,
changes in body composition, fluid retention commonly associated with illness
and reduction in height with ageing-related kyphosis (Nightingale et al., 1996;
Deurenberg-Yap et al., 2000; Kondrup et al., 2003b; Shirley et al., 2008). Body
Mass Index does not distinguish between major components of body weight (fat,
lean body mass and fluid). It is common for clinicians to consider underweight
patients as malnourished and patients with normal or high BMI as being well
nourished. However as a stand-alone marker, this can be misleading and may
result in over or under-diagnosis of malnutrition. Patients who have normal or
elevated BMI may have lost substantial body weight, placing them at nutritional
risk even if their BMI remains in the normal range. People of Asian ethnicity may
have a small frame or be classified as underweight based on BMI, yet are
healthy.
Different studies have used different BMI cutoffs for malnutrition (Kelly et al.,
2000; Thomas et al., 2002; Kyle et al., 2003). While Kelly et al (2000) rationalised
using BMI 18.5 kg/m2 as a cutoff for malnutrition based on World Health
Organisation recommendations, the remaining authors did not give a reason for
adopting a specific BMI cutoff for malnutrition. Deurenberg et al. (2001) found
that Asians have a different body composition from Caucasians and hence
require different cutoffs for BMI in overnutrition (Deurenberg-Yap et al., 2000;
Deurenberg & Deurenberg-Yap, 2001). Therefore it is likely that the cutoffs for
malnutrition risk will similarly differ for different populations even when the same
nutrition screening tool is used.
69
Confusion between screening and assessment
There is some lack of clarity in the literature regarding screening and

assessment. The naming convention of a few screening tools poses some
confusion to healthcare staff. For example, Nutrition Assessment Score (Oakley
& Hill, 2000), Nursing Nutritional Assessment Tool (Scanlan et al., 1994) and
Short Nutritional Assessment Questionnaire (SNAQ) (Kruizenga et al., 2005a)
which were developed as screening tools but were named as assessment tools.
The SNAQ uses a score of 2 to indicate moderate malnutrition and 3 as
severe malnutrition, adding to the confusion between risk and diagnosis of
malnutrition (Kruizenga et al., 2005a). Some investigators have used screening
tools to determine the nutritional status of patients (Bruun et al., 1999; Vanis &
Mesihovic, 2008; Gur et al., 2009), whereas in best practice, a nutrition
assessment tool should be used. Bruun et. al (1999) used BMI and weight loss to
assess nutritional status of surgical gastrointestinal and orthopaedic patients,
while Gur et. al. (2009) used NRS-2002 to determine the prevalence of
malnutrition in newly admitted surgery patients. Another example is the BMI,
which is used in some studies as part of a screening tool (Kondrup et al., 2003b;
Stratton et al., 2004) and used in others as part of an assessment tool to
determine nutritional status (McWhirter & Pennington, 1994; Landbo et al., 1999;
Kelly et al., 2000; Thomas et al., 2002; Kyle et al., 2003; Pirlich et al., 2006).
Body Mass Index has been incorporated into many screening tools, supporting
the fact again that by using BMI alone one cannot determine the nutritional status
of an individual (Kondrup et al., 2003a).
A single screening tool for an institution
In very large acute care hospitals, it is not practical to have different nutrition
screening tools for different disease and age groups. Notwithstanding, it is widely
agreed to be necessary to take a different approach for adult versus paediatric
patients. The aim of a single adult screening tool is to minimise confusion among
70
staff, standardise practice and conserve resources in training staff when they are
deployed from one ward to another. However, a number of age or disease
specific tools have been developed. For example, MNA-SF has been developed
for use with the elderly (Rubenstein et al., 1999) and Malnutrition-Inflammation
Score (MIS) has been developed for patients on haemodialysis (Kalantar-Zadeh
et al., 2001). The MNA-SF (Rubenstein et al., 1999) was developed from the
original 18-item MNA (Guigoz et al., 1994; Guigoz et al., 1996) by Nestl in
France as a screening tool specifically for the elderly ( 65 years old) and its use
has been limited mainly to this group of patients (Anthony, 2008). Both the MNA-
SF and MIS have been validated and are frequently used in nursing homes and
long term care facilities and dialysis centres respectively; and in these settings
have served their purpose well (Rubenstein et al., 1999; Kalantar-Zadeh et al.,
2001; Kalantar-Zadeh et al., 2003; Guigoz, 2006). Such targeted tools may be
more sensitive to the needs of their specific populations. However in the hospital
setting where there are patients with a wide age range of age and disease
conditions, it is not practical to carry out MNA-SF on the elderly, MIS on dialysis
patients and use a separate tool for the rest of the patients. At most, there should
only be separate nutrition screening tools for adult and paediatric patients, where
a small change in a nutritional parameter may have a more significant implication
for a child compared to an adult. For example a weight loss of one kilogram in a
two-month-old infant is highly significant whereas in an adult it may not be.
Similarly, failure to gain weight in a certain period of time is of significant concern
in a growing child. Screening tools developed for specific medical conditions or
age groups are summarised in Table 1.5.
For the reasons mentioned above, nutrition screening tools which are disease- or
age-specific will be excluded from literature review henceforth, unless prior
validation study in general adult hospitalised patients across different age and
disease groups has been carried out, and the tool found to be suitable for use in
hospital adult patients as a whole.
71
Table 1.5: Nutrition screening tools developed for targeted groups of

people
Targeted groups Nutrition Screening Tool
Adult with learning Nutrition Screening Tool (Bryan et al., 1998)
difficulties
Critically ill Prognostic inflammatory and nutritional index
(Ingenbleek & Carpentier, 1985)
Nutrition Risk in the Critically ill Score (NUTRIC
Score) (Heyland et al., 2011)
Community Nutrition Screening Equation (Elmore et al.,

1994)
Nutrition Risk Score (Reilly et al., 1995)
Community Focused Nutritional Screening Tool
(Gilford & Khun Khun, 1996)
Nutrition Assessment Tool (Hickson & Hill,
1997)
Screening in Practice (Ward 1998)
Community Nutrition Risk Assessment
(Bartholomew et al., 2003)
Children with special Nutritional Screening Test (Henderson et al.,

healthcare needs 1992)
PEACH Survey (Campbell & Kelsey, 1994)
Nutrition and Feeding Risk Identification Tool
(Baroni & Sondel, 1995)
Pediatric Nutritional Risk Score (Sermet-
Gaudelus et al., 2000)
STRONG(kids) (Hulst et al., 2010)
Dialysis Dialysis Malnutrition Score (DMS) (Kalantar-

Zadeh et al., 1999)
Malnutrition-Inflammation Score (MIS)
(Kalantar-Zadeh et al., 2001)
Screening Tool (Gower, 2002)
Objective Score of Nutrition on Dialysis
(Beberashvili et al., 2010)
72
Table 1.5: Nutrition screening tools developed for targeted groups of

people (continued)
Targeted groups Nutrition Screening Tool
Elderly Nutrition Screening Tool (Noel & Wojnaroski,
1987)
Nutrition Risk Index (Veterans Affairs Total
Parenteral Nutrition Cooperative Study Group,
1991)
DETERMINE (White et al., 1992)
Nutrition Score (Charalambous, 1993)
Nutritional Risk Assessment Scale (Nikolaus et
al., 1995)
Nursing Nutritional Screening Tool (Cotton et
al., 1996)
Mini Nutritional Assessment Short Form
(MNA-SF) (Rubenstein et al., 1999)
Nursing Nutrition Screening Assessment
(Pattison et al., 1999)
Simple Screening Tool (Laporte et al., 2001b)
Nutritional Risk Screening 2002 (Kondrup et al.,
2003b)
Rapid Screen (Visvanathan et al., 2004)
New Screening Tool (Charlton et al., 2005)
Geriatric Nutritional Risk Index (Bouillanne et
al., 2005)
Oncology Nutritional Screen Form (Lundvick & Phillips,
1983)
General Nutritional Status Score (GNS) (Guo et
al., 1994)
Oncology Screening Tools (Latkany et al.,
1995)
Patient-generated Subjective Global Screening
(VSG-GP) (Gomez Candela et al., 2010)
Preoperative surgery Prognostic Nutritional Index (Buzby et al., 1980)
Maastricht Index (Kuzu et al., 2006)
Psychiatric St Andrew's Nutrition Screening Instrument
(SANSI) (Rowell et al., 2012)
Surgical Undernutrition Risk Score (Doyle et al., 2000)
Trauma Nutrition Checklist (Cooper, 1998)
73
1.2.2 Validity of Nutrition Screening Tools
For a nutrition screening tool to be accurate and useful, it has to be valid and
reliable (Charney, 2008). Any screening procedure must have an acceptable
level of sensitivity and specificity, relative to some definitive diagnostic procedure
or gold standard (Rush, 1993). Sensitivity (the ability to identify true cases) is
important where the consequences of an undetected case may be significant.
Specificity (the ability to classify correctly those without the condition) is important
to avoid labeling someone with an incorrect presumptive diagnosis, which may
result in anxiety for the patient, unnecessary procedures and costs (Rush, 1993).
Jones (2002) critically appraised 44 published reports on nutrition screening and

assessment tools and found that only two-thirds assessed validity. Of the 44
studies appraised, 8 were studies using tools developed for hospitalised adult
patients (all medical conditions), of which only 5 were validation studies.
However, the review did not compare the sensitivity and specificity of the tools
(Jones, 2002). The sensitivity and specificity of tools allow comparison between
the tools to determine which tool to adopt. A nutrition screening tool that has high
sensitivity but low specificity may cause too many patients to be referred to the
dietitians unnecessarily and waste resources. For example the South
Manchester University Hospitals Trusts Nutrition Screening Tool (NST) (Burden
et al., 2001) was 78% sensitive and 52% specific, which meant that about half of
well-nourished patients were identified as at risk of malnutrition. This is likely
caused by over-scoring of patients with co-morbidities, as mental status, medical
condition and gut-function are scored separately. Patients on tube feeding
receiving adequate nutrition were also scored poorly. Hence, the tool might have
identified too many patients who were not truly at malnutrition risk, resulting in
the low specificity of the tool. On the other hand, a tool that has low sensitivity but
high specificity is more worrying as many patients who are malnourished will not
receive the necessary intervention. A validation study on 995 inpatients in a
74
hospital in Switzerland showed that although the specificity of NRI was good
(89%), its sensitivity was only 43% when compared against SGA (Kyle et al.,
2006). This is a concern as many malnourished patients will be missed out if NRI
tool is used.
Green and Watson reviewed 35 papers on nutrition screening and assessment

tools from the year 1982 to 2002, of which 13 were studies on nutrition screening
tools developed for use on adult hospitalised patients, 7 were on nutrition
assessment tools (mostly SGA or the modified version of SGA) and 15 were on
screening tools developed for children, community care or specific medical
conditions (Green & Watson, 2005). The authors found that many of the
published tools did not go through rigorous testing (Green & Watson, 2005).
Similar to the study by Jones (2002), the review did not provide information on
the reliability, sensitivity and specificity of the tools or whether the studies were
conducted in a blinded manner (Green & Watson, 2005).
Table 1.6 presents the validity and reliability of nutrition screening tools in
hospitalised adult patients. For nutrition screening tests, van Venroij (2007)
recommended sensitivity and specificity of at least 65% as acceptable (van
Venrooij et al., 2007). However this would depend on the relative importance of
detecting at risk patients and the resources (dietitian) needed to attend to
patients who were identified at risk. The higher the sensitivity and specificity of a
tool, the better. The validity of a nutrition screening tool can be determined by
validating it against a gold standard (Jones, 2004b). There is no consensus on
the gold standard of a nutrition assessment tool to define malnutrition. For
example, MUST was validated against seven nutritional screening and
assessment tools in the United Kingdom (Stratton et al., 2004), NRS was
validated against NRIb (Reilly et al., 1995) and MST validated against SGA in
Australia (Ferguson et al., 1999a; Ferguson et al., 1999b; Isenring et al., 2009).
The agreement between MUST and the various reference standards varies
according to reference standards; from poor agreement (r=0.255) when validated
75
against Undernutrition Risk Score to excellent agreement (r=0.775) when

validated against NRS (Stratton et al., 2004). Subsequent validation studies on
MUST by different investigators showed low sensitivities ranging from 54% to
61% (Bauer & Capra, 2003; Kyle et al., 2006; Lawson et al., 2012). All validation
studies on MST have used SGA as the gold standard with sensitivity ranging
from 41% to 100% (Ferguson et al., 1999a; Ferguson et al., 1999b; Lawson et
al., 2012).
Although many nutrition screening tools have been developed in various

countries, none have yet been developed and published in Asia. Many screening
tools such as the Nursing Nutritional Assessment Tool (Scanlan et al., 1994),
Nutrition Risk Check (Rawlinson, 1998) and Nutritional Screening Tool (Scott &
Hamilton, 1998), have not been validated even years after being developed. Yet
others such as the Nutrition Risk Score (Reilly et al., 1995), Derby Nutritional
Score (Goudge et al., 1998) and Nutrition Screening Tool (Weekes et al., 2004),
did not perform analytical accuracy testing on their tools, hence the sensitivity
and specificity of the tools are not known. These and other validation studies
presented in Table 1.6 showed that some of the screening tools were validated in
various clinical and healthcare settings across different countries, using different
reference standards, with different results. The most commonly used reference
standard for validation of nutrition screening tools was SGA, which will be
discussed further in section 1.2.4.
Nutrition screening administered by nurses in validation studies
A survey carried out in the USA showed that only 10% of nutrition screening was
administered by nutrition services staff, in comparison to 84% by nursing staff
and 4% by a computerized system (Chima et al., 2008). Nurses have the first
and regular contact with every patient upon admission, hence they are the most
suited to carry out nutrition screening on patients upon admission (Johnstone et
al., 2006a; b). Having dietetic staff to conduct nutrition screening is costly and not
76
feasible in many institutions. Lyne and Prowse (1999) shared the need for a
simple, effective and valid nutrition screening tool for nurses to prevent non-
compliance (Lyne & Prowse, 1999).
Although in practice nutrition screening is commonly undertaken by nursing staff,

few studies have used nurses to validate nutrition screening tools. Most studies
have used researchers, dietitians or dietetic staff to implement nutrition screening
(Ferguson et al., 1999a; Kyle et al., 2006; Kim et al., 2011; Velasco et al., 2011).
Kim et al (2011) used dietitians to test the validity of the Malnutrition Screening
Tool for Cancer Patients (MSTC) against PG-SGA. The sensitivity and specificity
of MSTC was 94% and 84% with high agreement (k= 0.7) (Kim et al., 2011). The
Malnutrition Screening Tool (MST) was developed to screen acute care patients
and validated against SGA by one dietitian, with a sensitivity and specificity of
93% (Ferguson et al., 1999a). The same tool was shown to be an effective
screening tool in comparison to PG-SGA when administered by researchers, with
100% sensitivity and 92% specificity (Isenring et al., 2006). Previous validation
studies which used nursing staff to administer screening and a dietitian to
complete nutrition assessment were with the British Nutrition Screening Tool
(NST) (Mirmiran et al., 2011), MST (Nursal et al., 2005a), Screening Sheet
(Thorsdottir et al., 1999) and MUST (Lawson et al., 2012), with sensitivity and
specificity ranging from 49 -85% and 63-85%, respectively.
Ideally, the assessor for the tool and the reference standard should be
independent, and blinded to each others results to prevent bias in validation
studies. Table 1.6 shows that most validation studies on nutrition screening tools
did not utilise blinding of the assessors. In the original validation study on MUST,
none of the assessors were blinded and the study showed good agreement with
SGA (k = 0.783) (Stratton et al., 2004). In two validation studies on MUST and
MST, the assessors were blinded to each others results but the results showed
poor sensitivities (59%, 54% and 49%) when compared to SGA (Bauer & Capra,
2003; Lawson et al., 2012). Two other studies which conducted blinding of the
77
assessors found the specificity of the Nutrition Screening Tool to be poor (62%)
(Mirmiran et al., 2011) or did not report on the results (Weekes et al., 2004).
1.2.3 Reliability of Nutrition Screening Tools
Reliability means results have to be consistent and reproducible. Inter-rater

reliability is the variation in results or measurements when assessed by different
persons using the same tool at the same time (Jones, 2004a). Intra-rater
reliability is the variation when the measurement is repeated by the same person
using the same tool at different times. However reliability does not imply validity.
A reliable tool may be able to produce the same results consistently when used
by a different assessor, but it may not necessarily give correct or accurate
results. Most studies have used kappa coefficient as a statistical measure to
determine inter-rater agreement. Fleiss (2003) defined a kappa of above 0.75 as
excellent, 0.40 to 0.75 as fair to good and below 0.40 as poor (Fleiss et al.,
2003). Among the nutrition screening tools, inter-rater reliability studies have
been performed on many of them, including MST (Ferguson et al., 1999a), NRS
(Reilly et al., 1995) and NST (Burden et al., 2001).
Similar to validity studies, many inter-rater reliability tests have been performed
on nutrition screening tools that compared the results of dietitian with dietitian, or
dietitian with nursing staff (Reilly et al., 1995; Ferguson et al., 1999a; Burden et
al., 2001). These studies generally show good agreement. Inter-rater agreement
between dietetic staff using MST was found to be excellent at 96% (r = 0.93-0.96,
Kappa = 0.84-0.88) (Ferguson et al., 1999a), and similarly research and nursing
staff implementing MST also shows high inter-rater reliability (k = 0.82, p<0.001)
(Isenring et al., 2006). Nutrition Risk Screening (NRS) shows good inter-rater
reliability when administered by dietitians (r=0.91, p<0.001) or by dietitians and
nurses (r=0.80, p<0.001) (Reilly et al., 1995). In contrast, McCall and Cotton
found poor agreement between dietitian and nurse administering the Nursing
Nutritional Assessment (NNA) (McCall & Cotton, 2001).
78
A nutrition screening tool should also be tested for inter-rater reliability among the
end-users, which are often the nurses. Only two studies examining inter-rater
reliability between nurses have been identified (McCall & Cotton, 2001; Mirmiran
et al., 2011). In the study by Mirmiran et al. (2011), two nursing staff administered
the British Nutrition Screening Tool (NST) on 446 newly admitted patients, with a
good level of agreement (Kappa = 0.71) (Mirmiran et al., 2011). In contrast,
McCall and Cotton (2001) found consistent disagreement between nursing staff.
The reliability study on NNA was carried out by two nurses on 185 patients
admitted to an acute elderly ward and the coefficient for the tool as a whole was
0.53. The authors attributed the poor inter-rater reliability between dietitian and
nurse, and nurse and nurse, to subjective interpretation of questions within the
tool (McCall & Cotton, 2001).
Summary of Issues related to Nutrition Screening Tools
There is no consensus on the best method for nutrition screening. The selection,
reporting and interpretation, including cutoffs, of nutrition screening may differ
between different racial groups, healthcare systems and cultural contexts. In
addition, a nutrition screening tool should be user friendly and use easily
obtainable data, not time-consuming, non-invasive, simple (not too lengthy) and
inexpensive. The limitations of current tools are their complexities, high rate of
missing data, age specificity, specialty specificity, cultural barriers and being not
user-friendly. These warrant the development of a new screening tool for acute
hospitals in Singapore, suitable for use across different diagnostic and adult age
groups ( 21 years old). Despite the variation in culture and needs, there has not
been any nutrition screening tool developed and validated for acute care
hospitals in Asia. In the last decade, most validation studies on nutrition
screening tools in Australia, Europe and the United States have used SGA as the
reference standard. However these studies did not perform blinding of the
79
assessors nor use the intended or most likely assessor (nurses) to conduct the
screening in their studies.
80
Table 1.6: Reliability and validity of nutrition screening tools in hospitalised adult patients* and method of validation,
including blinding of assessor (excluding tools developed for older adults, community and specific medical conditions)
Tool Participants, Reliability Relative validity
(Lead author) Country Number of Number Inter-rater reliability Number Reference Assessor for tool and Sensitivity Specificity
subjects of raters of standard reference standard/
subjects Blinding
Malnutrition Screening Tool (MST)
MST Hospital inpatients 32 2/3 Agreement between 408 SGA 1 dietitian, not blinded 93% 93%
(Ferguson et al., (Australia) 2 dietitians = 96%
1999a) (k = 0.88),
Agreement between
a dietitian and
nutrition assistant =
93% (Kappa = 0.84)
MST Oncology patients Not evaluated 106 SGA 2 dietitians, blinding not 100% 81%
(Ferguson et al., undergoing specified
1999b) radiotherapy*
(Australia)
MST Hospital inpatients 2181 2 Agreement between 2211 SGA 1 dietitian and 1 nurse, 74% 76%
(Nursal et al., (Turkey) 1 dietitian and 1 blinding not specified
2005a) nurse (k = 0.72)
MST Hospitalised renal 23 2 Agreement between 145 SGA Dietitians conducted the 49% 86%
(Lawson et al., patients* (United 2 nurses (k = 0.33) SGA.
2012) Kingdom) Nurses or nursing students
conducted the MST.
Investigators mostly
blinded.
SGA = Subjective Global Assessment, MST = Malnutrition Screening Tool, k = kappa score
*Further validation studies conducted on patients with specific medical condition are presented, only if the tool has previously been validated in general hospitalised adult patients
81
including blinding of assessor (continued)
(Lead author) Country Number Number Inter-rater reliability Number Reference Assessor for tool and reference Sensitivity Specificity
of of raters of standard standard/ Blinding
subjects subjects
Malnutrition Universal Screening Tool (MUST)
MUST Hospitalised Not evaluated 75 NRS (Medical) 1 assessor conducted test tool k = 0.775, Not
(Stratton et al., patients 75 and reference standards except & 0.813 specified
2004) (United Kingdom) 86 & 85 MST (Medical) for MNA which had 2 assessors. k= 0.707
MNA (Elderly) All not blinded. Assessors were k = 0.551
50 MNA (Surgical) trained by dietetic research fellow k= 0.605
52 SGA (Medical) and physician. k = 0.783
URS (Surgical) k= 0.255
& 0.431
MUST Medical and Not evaluated 995 SGA 2 trained co-workers performed 61% 76%
(Kyle et al., 2006) surgical inpatients both MUST and SGA; not blinded
(Switzerland)
MUST Hospitalised Not evaluated 65 SGA 2 experienced dietitian performed 59% 75%
(Bauer & Capra, oncology patients* SGA and MUST independently;
2003) (Australia) blinded
MUST Hospitalised renal 29 2 Agreement between 147 SGA Dietitians conducted the SGA. 54% 78%
(Lawson et al., patients* (United 2 nurses: Nurses or nursing students
2012) Kingdom) Kappa = 0.58 conducted the MUST.
Investigators mostly blinded.
SGA = Subjective Global Assessment, MUST = Malnutrition Universal Screening Tool, NRS = Nutrition Risk Score, MST = Malnutrition Screening Tool, MNA = Mini Nutritional
Assessment, URS = Undernutrition Risk Score, k = Kappa score
*Further validation studies conducted on patients with specific medical condition are presented, only if the tool has previously been validated in general hospitalised adult patients
82

(Lead author) Country Number of Number of Inter-rater Number Reference standard Assessor for tool Sensitivity Specificity
subjects raters reliability of and reference
subjects standard/ Blinding
OTHERS (in chronological order)
Screening Hospital adult

Nutritional Profile patients Not evaluated Not evaluated
(Hunt et al., (United States)
1985a)
East Orange Medical and 94 Subsequently received total 95% 89%
Nutrition surgical patients parenteral nutrition or referral
Screening Form Not evaluated Not specified
on admission for dietary consultation
(Brown & (United States)
Stegman, 1988)
Nursing Hospitalised
Nutritional patients
Assessment Tool Not evaluated Not evaluated
(United Kingdom)
(Scanlan et al.,
1994)
Nutrition Risk Hospitalised 20 2 Correlation: r = 0.91 40 Nutritional Risk Index (NRIb) 10 dietitians, Correlation: r = 0.68
Score (Reilly et patients on (between dietitians) and Clinical Impression blinding not (NRIb)
al., 1995) admission r= 0.83 (between specified r= 0.83 (clinical
(United Kingdom) dietitian and nursing impression)
staff)
Nutrition Risk Hospitalised 186 1 nurse Inter-observer 56 Prealbumin levels Not specified who 84.6 62.7
Classification patients on and 1 agreement of nutrition retrieved the
(Kovacevich et admission nutritionist classification between prealbumin results,
al., 1997) (United States) nurse and nutritionist blinding not
was 97.3% (p = 0.95) specified
BMI = Body mass index, MAC = Mid arm circumference, TST = Triceps Skinfold Thickness, MAMC = Mid arm Muscle Circumference, TLC = Total Lymphocyte Count
83
(Lead author) Country Number Number Inter-rater Number of Reference standard Assessor for tool Sensitivity Specificity
of of raters reliability subjects and reference
Derby Nutritional Hospital patients Agreement Dietetic Assessment Scale based 1 nurse conducted Moderately good
Score (Goudge (United between on current actual weight, screening, 1 agreement (kappa=0.6)
et al., 1998) Kingdom) pairs of percentage weight change, if any, dietitian performed
70 5 73
nurses during the last 3 months, current reference standard,
(median appetite and current dietary blinding not
k=0.7) intake. specified
Nutrition Risk Hospitalised

Check patients
Not evaluated Not evaluated
(Rawlinson, (United
1998) Kingdom)
Nutritional (United
Screening Tool Kingdom) Not evaluated Not evaluated
(Scott &
Hamilton, 1998)
Screening Sheet Hospital patients 3 of these seven parameters 1 dietitian and
(Thorsdottir et from medical below reference value: nurses, blinding not
-2
al., 1999) and surgical BMI 20 kgm , Albumin < 38 specified
wards (Iceland) g/L, Total protein < 66 g/L
Not evaluated 82 Prealbumin < 180mg/L 69% 91%
Haemoglobin < 130 g/L
(women) or < 118 g/L (men)
TLC < 1.8 x 109/L
TST & MAC <10th centile
84
(Lead author) Country Number Number Inter- Number Reference standard Assessor for tool and Sensitivity Specificity
of of raters rater of reference standard/
subjects reliability subjects Blinding
Nutrition Medical and Dietetic Assessment Form: BMI < 20 1 dietitian conducted First week First week
Assessment surgical wards kgm-2, MAC, TST , MAMC <15th both the screening of study: of study:
Score patients centile, weight loss 10%, Dietary and assessment at 100% 65%
(Oakley & Hill, (United Kingdom) Not evaluated 118 assessment, appetite, problems with the same sitting, not Last 3 Last 3
2000)
eating, mental/social conditions, blinded weeks of weeks of
disease state study: study:
95.8% 78.5%
Simple Hospitalised acute At least 4 of these nutrition indicators 1 diet technician and 33.3% 81.3%
Screening care adult (<65 yrs Not evaluated 54 were abnormal: BMI, weight loss, TST, 2 dietitians, blinding
Tool #1 & # 2 old) (Canada) SSF mid arm muscle area, not specified 33.3% 97.9%
(Laporte et al., Hospitalised acute biochemical, dietary assessment and 78.6% 92.9%
2001a)
care elderly (65 Not evaluated 57 physical examination
yrs) (Canada) 76.7% 83.7%
Nutrition Hospital patients One or more markers of BMI < 20 1 nurse conducted
95%
Screening (Medical, surgical, kgm-2, MAC <15th centile, weight loss screening, 1 dietitian
level of
Tool renal and elderly) 100 2 100 >10%, and 25% of energy intake on performed reference 78% 52%
(Burden et al., agreeme
(United Kingdom) the first day in hospital as compared to standard, blinding
2001) nt
estimated energy requirements not specified
2
Nutrition Hospitalised BMI 18.520.0 kg/m or 510% weight 1 nurse & 1 dietitian,
k = 0.66
Screening patients to a loss in the previous 36 months blinded k = 0.717
(fair Not
Tool general medical 26 3 nurses 100 or nil by mouth for >5 days (good
agreeme specified
(Weekes et ward or TSF, MAC, or agreement)
al., 2004) nt)
(United Kingdom) MAMC <5th centile with no weight loss
85
(Lead author) Country Number of Number of Inter-rater Number Reference standard Assessor for tool Sensitivity Specificity
subjects raters reliability of and reference
Short Nutritional Hospitalised 47 Not k = 0.69 291 BMI < 18.5 kgm-2 and Number of 86% 89%
Assessment patients specified (between nurse unintentional weight loss assessors and
Questionnaire (Netherlands) and nurse) 5% in the last 6 months blinding not
(SNAQ) k = 0.91 specified
(Kruizenga et al.,
(between nurse
2005a)
and dietitian)
Nutrition Risk Medical and Not evaluated 995 SGA 2 trained co- 43% 89%
Index (Kyle et al., surgical workers, not
2006) inpatients blinded
(Switzerland)
Nutritional Risk Medical and Not evaluated 995 SGA 2 trained co- 62% 93%
Screening 2002 surgical workers, not
(Kyle et al., 2006) inpatients blinded
(Switzerland)
Nutrition Hospitalised 150 2 nurses k = 0.68 & 0.74 414 BMI 18.520.0 kg/m2 or 5 1 nurse & 1 86.7% 61.7%
Screening Tool patients on 10% weight loss in the nutritionist, blinded
(Mirmiran et al., admission previous 36 months
2011) (Iran) or decreased dietary
intake with <5% weight loss
in the previous 36 months
or TST, MAC, or
MAMC <5th centile with no
weight loss
86
1.2.4 Reference Standard to Validate a Nutrition Screening Tool
Nutrition screening tools are often validated using nutrition assessment tools. For
example, Malnutrition Screening Tool (MST) has been validated against SGA in
several studies (Ferguson et al., 1999a; Ferguson et al., 1999b; Isenring et al.,
2006). Some components contained within a nutrition screening tool are also
utilised as part of assessment, for example components of weight loss and loss
of appetite are present in MST and Nutrition Risk Score (NRS) (both nutrition
screening tools) as well as in SGA (nutrition assessment tool) (Ferguson et al.,
1999a; Corish et al., 2004). Body mass index is required in Malnutrition Universal
Screening Tool (MUST), Nutrition Risk Score (NRS) and Mini Nutritional
Assessment - Short Form (MNA-SF) (Rubenstein et al., 1999; Corish et al., 2004;
Stratton et al., 2004). Charney (2008) suggested that nutrition assessment
continues the data gathering process initiated during nutrition screening
(Charney, 2008). Assessment allows the clinician to gather more information to
determine if there is indeed a nutrition problem, to name the problem, and to
determine the severity of the problem (Charney, 2008). It is recommended that
patients identified to be at nutritional risk have a nutrition assessment carried out
to determine if the patient is malnourished (Lochs et al., 2006).
Nutrition assessment is a comprehensive approach to determine the nutritional

status of individuals, whether well nourished or malnourished (American Dietetic
Association., 1994). Nutrition assessment is designed to systematically and
accurately identify those individuals with clinically significant malnutrition that
require nutrition intervention (Capra, 2007). It is also used to monitor and
evaluate outcomes of nutritional interventions (Feldblum et al., 2010). It involves
more in-depth data collection, usually including a combination of physical,
objective and biochemical data (Charney, 2008). Validity of nutrition assessment
tools is usually established against clinical outcomes (Table 1.3) or more definite
measurements such as body composition (Bannerman & Ghosh, 2000; Lawson
et al., 2001; Gupta et al., 2005; Norman et al., 2008b). ESPEN specified that
87
nutrition assessment should be conducted by an expert clinician, dietitian, or

nutrition nurse (Lochs et al., 2006).
In medical practice, screening tests such as mammogram and fecal occult blood
test are always validated against diagnostic tests, such as biopsy of breast tissue
and colonoscopy respectively (Wilhelm et al., 1986; Parente et al., 2009; Oxner
et al., 2012; Muinuddin et al., 2013). Similar to this concept, a nutrition screening
tool should always be validated against nutrition assessment which can provide a
diagnosis.
The decision on which nutrition assessment tool to validate a nutrition screening

tool can be based on these criteria:
Able to diagnose malnutrition (refer to Table 1.2)
Can be used for monitoring the progress of patients and effectiveness of
nutrition intervention (Beattie et al., 2000; Paton et al., 2004; Vermeeren et
al., 2004; Ravasco et al., 2005a; Norman et al., 2008a; Ha et al., 2010;
Paccagnella et al., 2010)
Validated against clinical outcomes (refer to Table 1.3)
Reliable (results repeatable) (refer to Table 1.8)
Easy to use for dietitians
Non-invasive for patients
Inexpensive
Table 1.7 presents the various nutrition assessment tools, their components, and
compares them based on the above criteria. Subjective Global Assessment is the
only nutrition assessment tool able to fulfill all the criteria above and will be
discussed in the next section.
88
Table 1.7: Comparison of nutrition assessment methods to validate a nutrition screening tool
PROPOSED CRITERIA
Non-invasive
a
Nutrition assessment
Inexpensive
Easy to use
malnutrition
b
Diagnose
progress
c
Validated
Monitor
Reliable
methods based on
Components Limitations
previous studies
stated in Table 1.2
MAMC a) Mid arm circumference (MAC) Skinfold callipers required

No Yes Yes No No Yes Yes Results skewed with untrained and inexperience
(Bishop et al., 1981) b) Triceps skinfold thickness (TST)
staff
AMA / CAMA Mid arm muscle circumference (MAMC) Hassle of carrying callipers around
No Yes Yes No No Yes Yes Percentile chart and equations not validated for
(Heymsfield et al., 1982)
non-Caucasian
SGA a) History of weight loss Subjective assessment
Yes Yes Yes Yes Yes Yes Yes
(Baker et al., 1982a) b) Changes in dietary intake Requires trained and experienced staff to perform
c) Gastrointestinal symptoms
7- Point SGA d) Functional status Yes Yes Yes Yes Yes Yes Yes
(Churchill et al., 1996) e) Disease state affecting nutritional
requirements
The first part of the assessment needs to be
f) Muscle wastage
PG-SGA completed by the patient. Literacy and
g) Fat stores Yes Yes Yes Yes No Yes Yes
(Ottery, 1994) understanding level may be a challenge for
h) Ankle or sacral oedema or ascites
patients
Mini Nutritional Assessment g) BMI
h) Mid arm circumference
(MNA) i) Calf circumference
(Guigoz et al., 1996) j) Weight loss in pass 3 months
k) Living independently vs. nursing home
l) > 3 prescription drugs
m) Presence of psychological stress/ acute
disease
Lengthy questionnaire of 18 items takes time to
n) Mobility
complete
o) Neuropsychological problems Yes Yes Yes Yes No Yes Yes
Limited to the use in the elderly. Hence, need to
p) Pressure sores / skin ulcers
use a different tool for younger adult patients
q) Number of full meals per day
r) Frequency of protein foods consumption
s) Frequency of fruits and vegetables
t) Declining food intake in pass 3 months
u) Beverages consumption per day
v) Mode of feeding
w) Nutritional problems as scored by patient
x) Patients perception of his/her health
MAMC = Mid arm muscle circumference, MAC = Mid arm circumference, TST = Triceps skinfold thickness, AMA = Arm muscle area, CAMA = Corrected arm muscle area,
SGA = Subjective Global Assessment, PG-SGA = Patient-generated Subjective Global Assessment, BMI = Body Mass Index, MNA = Mini Nutritional Assessment
a.) Refer to Table 1.2, b.) Refer to Table 1.3, c.) Refer to Table 1.8
89
Table 1.7: Comparison of nutrition assessment methods to validate a nutrition screening tool
PROPOSED CRITERIA
Non-invasive
a
Nutrition assessment
Inexpensive
Easy to use
malnutrition
b
Diagnose
progress
c
Validated
Monitor
Reliable
methods based on
Components Limitations
previous studies
stated in Table 1.2
Invasive and costly need blood test

Biochemical
Blood test for serum albumin, prealbumin or Results not immediately available and is marred by
(Albumin / Prealbumin / TLC) No Yes Yes No Yes No No
Total Lymphocyte Count inflammation, illness and fluid balances
(Doweiko & Nompleggi, 1991)
Overdiagnosis of malnutrition (low specificity)
BMI Missing data for patients unfit to be weighed
a) Weight
(World Health Organisation, No Yes No Yes Yes Yes Yes Results inaccurate for subjects with fluid retention
b) Height
1995) or dehydration. Cutoffs confounded by ethnicity.
BMI < 20 with either TST or MAMC below the
15th percentile
or
BMI 18.5 (age 1865 y) or 20 (age >.65 y);
Combined method s
or Cutoff for malnutrition not standardised
(Edington et al., 2000; Meijers
Unintentional weight loss ( 6 kg in the last 6 No No No No No Yes Yes Not well accepted
et al., 2009)
months or 3 kg in the last month); BMI has limitations mentioned above.
or
No nutritional intake for 3 days or reduced
intake for >10 days combined with a BMI of
18.520 (age 1865 y) or 2023 (age >65 y)
TLC = Total Lymphocyte Count, MAMC = Mid arm muscle circumference, MAC = Mid arm circumference, TST = Triceps skinfold thickness, BMI = Body Mass Index
a.) Refer to Table 1.2, b.) Refer to Table 1.3, c.) Refer to Table 1.8
90
Subjective Global Assessment (SGA)
Subjective Global Assessment (SGA) is well accepted as an assessment tool in

clinical practice (Baker et al., 1982a; Detsky et al., 1987; Lochs et al., 2006).
Subjective Global Assessment was developed by Baker et al to predict outcomes
in surgical patients and the tool was first published in 1982 (Baker et al., 1982b).
This validated and widely used nutrition assessment tool involves evaluation of
weight and dietary intake changes, gastrointestinal symptoms, functional capacity,
metabolic stress level from disease state and physical examination for evidence of
fat depletion, muscle wasting and nutritional related oedema. There is no
numerical weighting or scheme for combining the history and physical
examination into SGA; instead a subjective assessment of nutritional status is
made (Baker et al., 1982a; Detsky et al., 1987; Ek et al., 1996). The final SGA
rank is based on the subjective weighting of these features to classify patients into
three categories (3-point): A = well nourished, B = moderately malnourished and
C = severely malnourished (Baker et al., 1982b).
Subjective Global Assessment has been regarded by ASPEN as the best nutrition
assessment tool and one of the only two tools (the other one being MNA)
recognised as an assessment tool (A.S.P.E.N. Board of Directors and the Clinical
Guidelines Taskforce., 2002; Mueller et al., 2011). It has been incorporated into
clinical practice guidelines as a tool for assessing nutritional status in acute care
setting by the Dietitians Association of Australia (DAA) and ESPEN (Lochs et al.,
2006; Watterson et al., 2009). It is an easy, non-invasive and inexpensive tool for
widespread use by trained clinicians or dietitians (Keith, 2008).
Table 1.8 presents validation studies carried out with SGA. Subjective Global
Assessment was initially validated against the clinical judgment of clinicians in a
blinded manner (Baker et al., 1982a; Detsky et al., 1984). Subsequent studies
tested it against various anthropometry, biochemical and functional parameters
(Hirsch et al., 1991; Norman et al., 2005; Devoto et al., 2006; Pham et al., 2007).
91
For consistency in results, a nutrition assessment tool has to be reliable. The SGA
has good reliability if carried out by experienced clinicians and dietitians
(Steenson et al., 2013). A recent review paper on the inter-rater reliability of SGA
presented 11 published studies that showed agreement level of between 74-91%
(kappa = 0.34-0.88) (Steenson et al., 2013). It has been proposed that good inter-
rater reliability in SGA can be achieved with training and standardised data
collection techniques (Hirsch et al., 1991; Ek et al., 1996).
Subjective Global Assessment has been widely accepted and used as a

diagnostic tool for malnutrition, to track clinical outcomes and as a reference
standard to validate nutrition screening tools (Keith, 2008; Makhija & Baker, 2008;
Steenson et al., 2013). Among all the nutrition assessment tools, Subjective
Global Assessment had the most number of studies that used it to diagnose
malnutrition and track clinical outcomes, which equates to prognostic validation of
the tool (refer to Table 1.3) (Baker et al., 1982a; Detsky et al., 1984; Middleton et
al., 2001; Bauer & Capra, 2003; Correia & Waitzberg, 2003; Gupta et al., 2005;
Kyle et al., 2006; Kubrak & Jensen, 2007). The results of studies using SGA
consistently indicated poorer survival, higher length of hospital stay and higher
hospitalisation costs among malnourished patients (Middleton et al., 2001; Correia
& Waitzberg, 2003; Gupta et al., 2005). Correia and Middleton found that patients
rated as malnourished (categories B or C) using SGA stayed in the hospital 1.5-
1.7 times longer than well-nourished patients (Middleton et al., 2001; Correia &
Waitzberg, 2003). Middleton followed 819 patients for 12 months and found that
the mortality rate for Australian malnourished patients (rated using SGA) was
three times higher than well-nourished patients (Middleton et al., 2001). Using the
same tool for nutrition assessment, there was a 60% increased cost of
hospitalisation for malnourished patients in a study conducted in 25 Brazilian
hospitals (Correia & Waitzberg, 2003).
Due to its good prognostic value for a range of clinical outcomes, SGA has been
widely used as reference tool for validating screening tools (Table 1.6). The
92
Malnutrition Screening Tool (MST), MUST, NRS-2002 and NRI have all been
validated against SGA (Ferguson et al., 1999a; Ferguson et al., 1999b; Middleton
et al., 2001; Bauer & Capra, 2003; Correia & Waitzberg, 2003; Kyle et al., 2006;
Kubrak & Jensen, 2007; Lawson et al., 2012). Since the original 3-point SGA was
created, a number of modified versions have been developed, including the 7-
point SGA and Patient Generated SGA (PG-SGA), which will be discussed next.
7-point SGA
One of the disadvantages of the original SGA which consists of a 3-point scale is
that small differences in nutritional status during follow-up cannot be detected. To
overcome this problem, Churchill et al expanded the scale to a 7-point scale for
use in the well-known CANUSA study carried out on 680 patients commencing
peritoneal dialysis (Churchill et al., 1996). The ratings for nutritional status were
expanded to range from 1 to 7, in which ratings of 1-2 signify severely
malnourished, 3-5 signify moderately malnourished and 6-7 signify well nourished
(Churchill et al., 1996). Therefore, the results of nutrition status as assessed by
the 7-point scale will always be aligned with the conventional SGA, i.e. well
nourished, moderately malnourished or severely malnourished. The CANUSA
showed that a one unit lower score in the 7-point SGA was associated with 25%
increased relative risk of death (Churchill et al., 1996). Since then, the 7-point
SGA has been used widely to determine the nutritional status of renal patients in
clinical settings as well as in research (McCann, 1999; Visser et al., 1999;
Campbell et al., 2007; Steiber et al., 2007).
Nutrition assessment needs to be on-going in order to detect subtle decline in

patients or to track the patients response to nutritional intervention. Therefore it is
important to use a nutrition assessment tool that can detect small changes in
nutritional status and which is not confounded by hydration status of the patient.
Some authors have speculated that the 7-point SGA scale may be more sensitive
in identifying small changes in patients nutritional status (Jones et al., 2004;
93
Campbell et al., 2007). The candidate has been using the 7-point scale on her
patients for the past 14 years and her anecdotal experience is that by using this
tool, dietitians are able to detect changes faster in patients who improve or
deteriorate especially within the same category of nutritional status. For example,
in the original SGA assessment, a patient who is rated as B (malnourished) may
take 4 months to improve to A (well nourished). In the 7-point SGA scale, a
malnourished rating can be 3, 4 or 5. If a patient improves from a rating of 3 to
rating 4 (although still regarded as moderately malnourished), this signifies that
the nutrition care plan is on track. Vice versa, when a patient deteriorates across
the ratings, this can be picked up faster with the expanded 7-point SGA scale and
the nutrition intervention plan adjusted accordingly. There are no studies thus far
to support this opinion. However it is interesting that Campbells study showed an
increase in mean BCM from the lower rating of 3 to the highest rating of 7
(Campbell et al., 2007). The increase had a 2-stage pattern, whereby there was a
linear increase from SGA rating 3 to 5 and plateau from rating 5 to 7. This showed
that there was actually a body composition difference between the ratings (3, 4
and 5) within the same nutritional status (SGA B). To date the use of 7-point SGA
has been limited to renal patients.
To support the use of 7-point SGA in monitoring nutritional outcomes in an

intervention programme for malnourished patients, we conducted a study to
determine if 7-point SGA is able to detect small nutritional changes faster than the
conventional SGA. We have validated that the 7-point SGA scale is indeed more
time-sensitive in identifying small changes in patients nutritional status. That work
and research paper will support the use of 7-point SGA to monitor the nutritional
outcome of patients after discharge from the hospital and is not part of this
research programme (attached in Appendix E).
94
Patient-generated SGA (PG-SGA)
The Patient-generated Subjective Global Assessment (PG-SGA) was adapted

from SGA, specifically for use with oncology patients (Ottery, 1994; 1996). The
rationale for doing this was because SGA lacked sensitivity to detect
improvements in nutritional status observed over a short hospital admission. The
PG-SGA has all the components of the original SGA but requires more detailed
information. This tool is designed in a checkbox format for patients to complete
the first part of the assessment, which contains the history of weight change,
changes in food intake, gastrointestinal symptoms and functional activities. The
second part on medical condition, metabolic stress, muscle wastage, fat depletion
and oedema is completed by a healthcare professional. PG-SGA incorporates a
numerical score for each section, which is totalled to assist the assessor decide
on the next appropriate course of action.
The PG-SGA has been validated in oncology (Bauer et al., 2002; Isenring et al.,
2003a) and haemodialysis (Desbrow et al., 2005) patients and shown to have
good sensitivity, specificity and inter-rater reliability compared to the original SGA.
All the above studies were carried out in Australia. In a study on 71 oncology
patients admitted to an Australian private tertiary hospital, the PG-SGA score had
a sensitivity of 98% and a specificity of 82% when compared to SGA as reference
standard (Bauer et al., 2002). A prognostic validity study on PG-SGA was carried
out on 73 patients admitted to an acute stroke unit in an Australian private hospital
(Martineau et al., 2005). In this study, patients diagnosed as malnourished using
PG-SGA had longer lengths of stay (13 days vs. 8 days), increased complications
(50% vs. 14%), increased frequency of dysphagia (71% vs. 32%) and enteral
feeding (93% vs. 59%) (Martineau et al., 2005).
A limitation of PG-SGA is that the first part of the assessment is designed to be

completed by the patient. In Singapore, literacy level may be a challenge for some
patients in completing the form. On the other hand, the use of nurses to carry out
95
the first part of PG-SGA may be resource-intensive and not suitable for use in
Singapore hospitals. Another limitation of PG-SGA is that it is not always able to
achieve the same results as the conventional SGA (Bauer et al., 2002; Desbrow
et al., 2005).
Mini Nutritional Assessment (MNA)
Another well-recognised nutrition assessment tool is the MNA (Guigoz et al.,

1994; Guigoz et al., 1996). It consists of 18 items which are grouped into
anthropometric measurements, global evaluation of independent living,
medication intake, acute disease, psychological problems, mobility and pressure
ulcers, dietetic assessment on the number of full meals consumed per day, mode
of feeding, appetite, consumption of protein, vegetables, fruits and fluid, subjective
assessment of health status and whether the patients think they have malnutrition
(Guigoz et al., 1994; Guigoz et al., 1996). The MNA is based on a scoring system
to classify individuals into 3 categories: a)24 points: normal nutrition status, b)
17-23.5 points: borderline status/at risk, c) <17 points: malnutrition. It has been
widely used to assess the nutrition status of the elderly (Compan et al., 1999;
Gazzotti et al., 2000; Thomas et al., 2002; Visvanathan et al., 2004; Neumann et
al., 2005; Feldblum et al., 2007; Bauer et al., 2008; Tsai et al., 2009; Vanderwee
et al., 2010; Pereira Machado & Santa Cruz Coelho, 2011) It has also been
extensively validated and is able to predict outcomes in the older adults (Persson
et al., 2002; Hudgens et al., 2004; Kagansky et al., 2005; Neumann et al., 2005;
Charlton et al., 2007; Bauer et al., 2008; Cereda et al., 2008a; Wikby et al., 2008;
Chan et al., 2010; Charlton et al., 2012; Tsai et al., 2013). However MNAs use
has been limited to the elderly population and evidence of its validity in other age
groups has not been established. Other limitations of MNA are its lengthy
questionnaire and the requirement of patients BMI which may lead to difficulty in
completing the assessment on some patients. In a study of hospitalised geriatric
patients, MNA could only be completed in 66% of patients (Bauer et al., 2005).
96
Due to the above reasons, MNA was deemed not suitable for use as reference
standard for this research programme which encompasses all adult age groups.
Summary of Reference Standard to Validate Nutrition Screening Tool
From the literature review, there are various methods for nutrition assessment.
However, SGA is the most widely used and accepted assessment tool to
determine nutritional status and validate nutrition screening tools in adult
hospitalised patients due to its good prognostic value for a range of clinical
outcomes. It is non-invasive, inexpensive and has also been shown to be reliable
and easy to use by dietitians. It can be used across the adult age range. For
monitoring the progress of patients, the 7-point scale SGA is able to detect small
nutritional changes over shorter periods of time without differing from the final
results of the conventional SGA.
97
Table 1.8: Studies on reliability and validity of Subjective Global Assessment and other modified versions of SGA
(Lead author) Country Number Number Inter-rater Num Reference standard Assessor for Sensitivity Specificity
of of raters reliability ber of tool and
subjects subje Blinding
cts
Albumin, Total lymphocyte
Surgical patients Significant correlation between all
SGA count, Transferrin, % actual lean 2 assessors,
(Baker et al., (before surgery) 59 2 81% 48 reference standard except total
BW/ideal lean BW, % actual blinded
1982a) (Canada) lymphocyte count and transferrin
BW/ideal BW, %body fat/BW
Surgical patients
SGA 2 assessors,
(Detsky et al., (before surgery) 59 2 k = 0.72 59 Ability to predict infection 82% 72%
blinded
1984) (Canada)
Gastroenterology Significant difference in
Weight, mid arm circumference,
SGA patients within 4 Not measurements between well
(Hirsch et al., 175 2 79% 139 triceps skinfold, and serum
days of admission applicable nourished, moderately malnourished
1991) albumin
(United States) and severely malnourished
Patients (>70years)
with length of stay 2,
SGA at least 3 days and Internist
(Covinsky et al., 21 k = 0.71 Not evaluated
admitted to general and
1999)
medical unit nurse
(United States)
3 assessors,
2,
Patients with (2 dietitians Significant correlation between
SGA Dietitian Respiratory quotient, BMI,
(Scolapio et al., cirrhosis 15 >90% 15 and 1 respiratory quotient, albumin and
& Albumin, TSF, MAMC, CHI
2000) (United States) physician), CHI with SGA
physician
not blinded
SGA = Subjective Global Assessment, SGA A = Well nourished, SGA B = moderately malnourished, SGA C = severely malnourished, BMI = Body mass index, BW= Body weight, TST
= Triceps Skinfold Thickness, MAMC = Mid arm muscle circumference, MAC = Mid arm circumference CRP = C-reactive protein, PCR = Protein Catabolic Rate, k = kappa
98
(continued)
(Lead Country Number Number of Inter-rater Number Reference standard Assessor for tool Sensitivity Specificity
author) of raters reliability of and Blinding
subjects subjects
Hospital Hospital
Readmission,
Readmission= Readmission=
Mortality up to 3 months
Residents (65 50% 80%
after SGA, height, actual
years) admitted to Mortality= 75% Mortality= 84.4%
SGA 2, 85%, weight, usual weight, % 2 pharmacists,
(Sacks et al., long-term care 53 53 Significant difference between well
Pharmacists k = 0.75 usual weight, Ideal blinded
2000) facility nourished, moderately malnourished
weight, % ideal weight,
(United States) and severely malnourished for actual
BMI, Albumin,
weight, % usual weight, % ideal
Cholesterol
weight and , BMI.
Patients Hemoglobin, Albumin, Significant correlation between SGA
(>70years) Cholesterol, 3 assessors, (1 and TST, subscapular skinfold and
2,
SGA admitted to 73.6%, Lymphocyte, TST, medical student BMI.
(Duerksen et 87 Physicians 87
geriatric and k = 0.48 Subscapular skinfold, and 2 physician), Grip strength and laboratory data did
al., 2000)
rehabilitation units MAMA, BMI, Grip blinded not correlate significantly with clinical
(Canada) strength SGA.
78%,
60-75 k = 0.51 (not
Any available
SGA medical blinded)
(Duerksen, patients 30-37 Not evaluated
students and 65%,
2002) (Canada)
dietitians k = 0.34
(blinded)
= Triceps Skinfold Thickness, MAMC = Mid arm muscle circumference, MAMA = Mid arm muscle area, CRP = C-reactive protein, CHI = creatinine height index, k = kappa
99
(continued)
Tool Participants, Country Reliability Relative validity
(Lead author) Number Number Inter-rater Number Reference standard Assessor Sensitivity Specificity
of of raters reliability of for tool and
subjects subjects Blinding
Hospitalised patients
2,
except pregnant, had
Dietitian k = 0.88
SGA psychiatric conditions, 2199
(Nursal et al., and (p<0.001) Not evaluated
and intensive care unit
2005a) nurse
patients
(Turkey)
Inpatient at Medicine,
SGA Neurology, Long-term Detailed Nutritional Not
(Devoto et al., Not evaluated 108 77% 84%
care and Rehabilitation Assessment specified
2006)
wards (Italy)
k = 0.670 (good agreement)
BMI
Significant differences between SGA A
Weight, Weight loss, BMI,
Patients admitted for & B for Weight, Weight loss BMI,
Skinfold (TST), MAMC,
SGA elective major Not Skinfold (TST)
(Pham et al., Not evaluated 274 Albumin, Handgrip strength
abdominal surgery specified MAMC, Albumin & Haemoglobin
2007) Serum total protein,
(Vietnam) Significant differences between SGA B
Haemoglobin, CRP,
& C for all measures tested except for
Lymphocytes
CRP and lymphocytes
Hospitalised patients to
SGA an internal medicine 5, Kappa=
(Baccaro et al., 75 Not evaluated
service Physicians 0.75
2007)
(Argentina)
= Triceps Skinfold Thickness, MAMC = Mid arm muscle circumference, CRP = C-reactive protein, k = kappa
100
(continued)
(Lead author) Number of Number Inter-rater Number Reference Assessor for tool and Sensitivity Specificity
subjects of reliability of standard Blinding
raters subjects
Correlations with 7point
SGA scale :
BMI (r = 0.79, p <
BMI,
0.001), % fat (r = 0.77,
Mid arm
p <0.001),
circumference,
Mid arm
Haemodialysis and Mid arm
7-point SGA k = 0.72 4 nurses, blinding not circumference (r =
(Visser et al., peritoneal dialysis patients 22 4 22 muscle
k = 0.88 specified 0.71, p < 0.001)
1999) (Netherlands) Circumference,
Prealbumin (r = 0.60,
Serum
p = 0.004).
albumin,
Lower correlations were
Prealbumin
found with mid arm
muscle circumference
and serum albumin.
k = 0.5,
76 (inter- Spearmans Statistical difference in
7-point SGA
Haemodialysis patients rater) Rho=0.7 BMI 54 dietitians , blinding mean BMI (p<0.05) and
(Steiber et al., 54 154
2007) (Canada and United States) 111 (intra- k = 0.7 Albumin not specified albumin (p<0.001) across
rater) Spearmans Rho the 5 categories of SGA
= 0.8
SGA = Subjective Global Assessment, BMI = Body Mass Index , k = kappa
101
(continued)
(Lead author) Number Number Inter-rater Number of Reference standard Assessor for tool and Sensitivity Specificity
of of raters reliability subjects Blinding
subjects
Patient-
Oncology Inpatient
Generated SGA Not evaluated 71 SGA 1 assessor, not blinded 98% 82%
(Bauer et al., (Australia)
2002)
Patient- Dietitian, research
Haemodialysis SGA
Generated SGA Not evaluated 60 student and examiner 83% 92%
(Desbrow et al., inpatients (Australia)
(blinding not specified)
2005)
Transferrin, Albumin,
Dialysis Total protein,
Malnutrition
2, Cholesterol,
Score
Dialysis patients Dietitian Triglyceride, Creatinine, No correlation between the
(quantitative and 41 k = 0.83 41 1 assessor, not blinded
(United States) and Haematocrit, conventional SGA and any
modified from
other parameter.
SGA) physician Lymphocyte count,
(Kalantar-Zadeh MAC, MAMC, TST,
et al., 1999) Biceps, BMI, URR, PCR
Quantitative SGA Hospitalised patients

(Q-SGA) and except pregnant, had Q-SGA:
Q-SGA: 90%
Modified psychiatric conditions, 67%
Quantitative SGA Not evaluated 2197 SGA 1 dietitian, not blinded MQ SGA:
and intensive care unit MQ SGA:
(MQ-SGA) 90.9%,
patients 85.6%,
((Nursal et al.,
(Turkey)
2005b)
SGA = Subjective Global Assessment, BMI = Body Mass Index
102
1.2.5 Compliance with Nutrition Screening
For a nutrition screening tool to be effective, it must be completed fully and

accurately. However, the current literature reveals high rates of non-compliance
for nutrition screening. In a large Nutrition Day survey on 1217 organisational
units from 325 hospitals in 25 European countries and Israel (n = 21,007 patients),
a screening routine existed for only 52 % (range 21 - 73%) of the units surveyed
(Schindler et al., 2010). Self-reported surveys among healthcare professionals
indicate that routine screening for nutritional risk on admission is present in 13-
78% of the departments or institutions surveyed (Rasmussen et al., 1999; Mowe
et al., 2006; Lindorff-Larsen et al., 2007; Persenius et al., 2008; Ferguson et al.,
2010). In a survey on 68 healthcare institutions in Australia, 78% of the
respondents reported that nutrition screening occurs at their institutions (Ferguson
et al., 2010). However, out of those which had a nutrition screening process in
place, only half reported that all or almost all of their patients were screened
(Ferguson et al., 2010). In another study, Mowe et al. (2006) conducted a survey
by mail on the nutrition screening practices of 1753 doctors and 2759 nurses in
Denmark, Sweden and Norway. When asked if nutrition screening on admission
was carried out as a standard procedure, the results differed significantly between
the countries with a yes response of 40%, 21% and 16% from Denmark, Sweden
and Norway respectively (Mowe et al., 2006).
Specifically, there are a number of studies published on the compliance rate with
various nutrition screening tools (Table 1.9). Of the 15 studies found in the
literature, 4 were published as abstracts (Wong & Gandy, 2008; Voyce & Seager,
2009; Wong et al., 2009; Joyce et al., 2011) with limited information. The
incompletion rates of nutrition screening in Australia and Europe range from 28%
to 97% (Raja et al., 2008; Wong & Gandy, 2008; Neelemaat et al., 2011b; Geiker
et al., 2012). High levels of missing data especially on disease severity were
found in MUST (Neelemaat et al., 2011b). In addition, 25% of the patients had
incomplete data on weight, height and/or weight loss (Neelemaat et al., 2011b)
103
and 69% had no data on BMI (Toumi & Lawson, 2011). In a different study on
NRS 2002, 76% of the 2393 patients audited were not screened for nutritional risk
at all during their hospitalisation (Geiker et al., 2012). All the studies presented in
Table 1.9 except one (Geiker et al., 2012) did not look at whether the nutrition
screening was being completed correctly. Geiker et al. (2012) found that 92% of
the patients screened with NRS 2002 were either not assessed within 24 hours
from admission or had their nutritional status underestimated (Geiker et al., 2012).
Failure to achieve accurate and complete nutrition screening will affect the final
score allocated to a patient, which may result in a malnourished or at risk patient
not being referred for nutritional intervention.
To complete the process of nutrition screening, any patient identified from

screening to be malnourished or at risk of malnutrition should be referred to the
dietitians to receive a full nutrition assessment and intervention (American Dietetic
Association., 1994; Lochs et al., 2006; Joint Commission International, 2008). The
incidence of under-treatment of malnutrition is increased when patients screened
to be at risk are not referred to, evaluated and/or monitored by a nutrition-trained
professional. The incidence of at risk patients not referred to the dietitians ranges
from 16% to 66% (refer to Table 1.9). Kondrup et al (2002) showed that only 47%
of patients identified as at nutritional risk using NRS 2002 had a nutrition plan
implemented, and of these only 30% had dietary intake and/or body weight
monitored (Kondrup et al., 2002). In a study by McWhirter and Pennington (1994),
up to 70% of malnourished patients were not provided with any form of nutritional
intervention (McWhirter & Pennington, 1994).
The literature search yielded limited studies on strategies to improve the

compliance of nutrition screening, which are mostly confined to staff training. Two
studies showed that staff training was able to improve the completion rates of
MUST from 54% to 67% (Wong & Gandy, 2008) and 37% to 55% (Raja et al.,
2008), although the end results were still less than desirable. Similarly, Keller
(2007) found that training was essential for the success of nutrition screening in
104
community-dwelling older adults in Canada (Keller et al., 2007). The paper

reported on the issue of building capacity for nutrition risk screening among the
elderly in the community and did not report on whether the project had led to
improvement in the quality and take-up rate of screening. The author indicated
that the time lag between training and implementing screening should be as short
as possible. In addition, the content and depth of training is beneficial if
individualised to the level of expertise of the administrator (Keller et al., 2007).
Bailey (2006) reported on the compliance of nurses to nutrition screening after
implementation of MUST and training in a tertiary hospital in UK. The study
showed contrasting results in different wards (Bailey, 2006). Two wards
demonstrated good initial screening rates (87% and 73%), but the results
dwindled to 35% and 33% by the third audit cycle. Another ward started with
screening rate of 16% and achieved 32% by the third audit cycle. After refresher
training sessions were provided, the screening rates improved in one ward to 94%
but deteriorated to 16% in another ward (Bailey, 2006). This demonstrates the
challenges of implementing nutrition screening and sustaining good screening
rates.
Summary of Issues on Compliance with Nutrition Screening
Literature review indicates that compliance to nutrition screening is poor and

strategies to improve the compliance rate and sustain the results are needed.
Other than staff training, no published study has described the use of quality
improvement tools to facilitate improvements in the compliance to nutrition
screening and its referral process. It is important to conduct audit and process
evaluation of implemented nutrition screening tools as part of quality assurance to
improve the compliance rate to nutrition screening and to ensure that
improvements are sustained over time. An audit of this nature can identify gaps in
implementation as a precursor to improving the nutrition screening process, the
quality of the tool and staff behaviour. Quality improvement tools may be able to
105
determine the root causes of non-compliance, which when effectively addressed

will lead to substantial improvement and sustain the results.
106
Malnutrition in hospitalised patients and clinical outcomes: A missed opportunity
Table 1.9: Non-compliance and error rates with various nutrition screening tools in hospitalised adult patients
(sorted by countries)
Non-Compliance Rate
Country Nutrition
N Target Incompletion rate Error rate / At risk patients not
(Author, Year) Screening Tool
(%) referred (%)
United Kingdom 12% of at risk patients not referred to
48 Trauma inpatients Nutrition Checklist 33%
(Cooper, 1998) dietitians
78% (surgical
Medical emergency and
United Kingdom Name of tool not patients)
100 surgical elective inpatient Not specified
(Bell, 2007) mentioned 70% (medical
admitted 7 days
patients)
United Kingdom 432 46% (Audit in 2005) Not specified
(Wong & Gandy, Inpatient admitted 72hours MUST
2008) 392 33% (Audit in 2007) Not specified
92% of those screened were not done
United Kingdom
within 48 hours of admission
(Voyce & Seager, 301 Inpatient admitted 48hours MUST 69%
33% of at risk patients not referred to
2009)
dietitians.
United Kingdom
Inpatient aged > 16 years 55% of at risk patients not referred to
(Lamb et al., 328 MUST 31%
admitted 24hours dietitians
2009)
United Kingdom Spinal injury inpatient aged
SNST
(Wong et al., 81 18-80 years admitted 40% Not specified
2009) 72hours
United Kingdom
Cardiac and cardiothoracic NST 86% 66% of at risk patients not referred to
(Joyce et al., 140
inpatients 81% had no BMI dietitians
2011)
64% had no weight,
United Kingdom Weight, BMI and 69% had no BMI,
(Toumi & Lawson, 48 Inpatient admitted 3 days unintentional 52% had no Not specified
2011) weight loss information on weight
loss
MUST = Malnutrition Universal Screening Tool, SNST = Spinal Nutrition Screening Tool, NST = Nutritional Screening Tool, BMI = Body Mass Index,
NHS QIS = National Healthcare System Quality Improvement Scotland, NRS 2002 = Nutritional Risk Screening 2002, MST = Malnutrition Screening Tool,
SNAQ = Short Nutritional Assessment Questionnaire
107
Table 1.9: Non-compliance and error rates with various nutrition screening tools in hospitalised adult patients
(sorted by countries) (continued)
Non-Compliance Rate
Country Nutrition
N Target Incompletion rate Error rate / At risk patients not
(Author, Year) Screening Tool
(%) referred (%)
Australia 64% of malnourished patients were not
Inpatients admitted to 2 Name of tool not
(Middleton et al., 819 Not specified identified by the screening tool and
hospitals mentioned
2001) referred to the dietitian
47 96% (Audit 1)
Inpatient admitted MST
Australia 58 97% (Audit 2)
24hours in 4 wards Not specified
(Raja et al., 2008) 71 63% (Audit 1)
MUST
64 45% (Audit 2)
Australia
Inpatient admitted 39% (Ward A)
(Porter et al., 46 MUST Not specified
24hours in 2 wards 83% (Ward B)
2009)
Australia Inpatients admitted to a Name of tool not 64% of malnourished patients were not
275 Not specified
(Gout et al., 2009) public tertiary hospital mentioned referred to the dietitian.
Inpatients admitted under
Denmark
15 medical and surgical 86% of at risk patients did not have a
(Rasmussen et 590 NRS 2002 92%
department from 12 nutrition plan documented
al., 2004)
hospitals
Denmark 92% not assessed within 24hrs from
Inpatient admitted in
(Geiker et al., 2393 NRS 2002 76% admission or had an underestimation of
September 2008
2012) nutritional status
MUST 39%
Netherlands
Adult inpatients 18 years MST 30%
(Neelemaat et al., 275 Not specified
old NRS 2002 28%
2011b)
SNAQ 28%
MUST = Malnutrition Universal Screening Tool, SNST = Spinal Nutrition Screening Tool, NST = Nutritional Screening Tool, BMI = Body Mass Index,
NHS QIS = National Healthcare System Quality Improvement Scotland, NRS 2002 = Nutritional Risk Screening 2002, MST = Malnutrition Screening Tool,
SNAQ = Short Nutritional Assessment Questionnaire
108
1.3 INTERVENTIONS FOR PATIENTS WITH MALNUTRITION
The high prevalence and poor outcomes of hospital malnutrition are well
established, including longer length of hospital stay, higher mortality rate and
greater hospitalisation costs (Naber et al., 1997; Braunschweig et al., 2000;
Edington et al., 2000; Correia & Waitzberg, 2003) (see sections 1.1.2 and 1.1.4).
There is a need for hospital care to identify individuals at nutrition risk and put in
place a system to help improve the nutritional status of these patients. Nutrition
screening alone is insufficient to achieve beneficial effects and more research is
needed to explore interventions that will improve patient outcomes (Weekes et
al., 2009).
Possible ways to improve the outcomes of malnourished patients are to raise the
awareness of malnutrition amongst healthcare workers, and to put in place an
action plan that facilitates early identification of malnourished or at risk patients
so that appropriate medical nutrition therapy and adequate follow-up care can be
provided (Kruizenga et al., 2005b; O'Flynn et al., 2005).
The protocol for the management of malnourished patients needs to be

streamlined and evidence-based. The American Dietetic Association, recently
renamed as the Academy of Nutrition and Dietetics, has a general framework for
nutrition care process (Lacey & Pritchett, 2003). Based on this framework,
Campbell et al (2008) and Lim et al (2012) have published nutrition intervention
protocols for use in their studies on chronic kidney disease and haemodialysis
patients (Campbell et al., 2008; Lim, 2012). The results of these interventional
studies showed improvement in energy intake and nutritional outcomes for
patients (Campbell et al., 2008; Lim & Lye, 2012). However a nutrition care
process specifically for malnourished patients has not yet been developed.
Hence, a nutrition care process for the identification and management of
malnourished patients was developed based on the above protocols and the
candidates clinical experience (Figure 1.5).
109
Figure 1.5: Nutrition Care Process for Malnutrition
Nutrition Screening
Screen using a valid, reliable, quick, simple, effective and cost-efficient tool
Establish referral process for at risk patients
Nutrition Assessment and Diagnosis

Medical history and relevant laboratory tests, including biochemistry
Nutrition-focused assessment, including:
o Anthropometric data
o Biochemistry
o Detailed diet history leading to estimates of current macro and micronutrient intake,
especially calorie and protein intake
o Subjective assessment of nutritional status using Subjective Global Assessment (SGA)
o Physical activity
o Psychosocial and economic factors impacting on nutrition
o Readiness to change
Intervention
Individualised prescription for medical nutrition therapy
o Calculate nutrient requirements, including caloric and protein requirements.
o Determine deficits in nutrient intake by comparing current intake with requirements.
o Provide strategies to increase caloric, macro and/or micronutrient intake via diet
o Prescription of supplements (as required) if is anticipated that modification in diet alone
is not able to meet requirement
o Initiation of tube feeding nutrition support if oral nutrition support fails) or parenteral
nutrition if the gut is not working or there is contraindication to feed enterally
Self-management and family education
o Education on identifying macro and/or micronutrients
o Recipe modification
Behavioural interventions
Develop goals and set targets
Monitoring and Evaluation (Reassessment and Follow-up)

Anthropometry
Biochemistry
Dietary intake, including compliance with any supplements prescribed
Subjective Global Assessment
Behavioural change
Reinforce advice and provide further strategies for change if required
Expected outcomes
Body weight, muscle and fat stores improved
Biochemistry within expected range
Macro and micronutrient intake improved
Improvement in overall nutritional status or Subjective Global Assessment
Improved clinical outcomes i.e. length of stay, readmission, infection, quality of life,
survival rates
Adapted from: Lacey 2003, Campbell 2008, Lim 2012 (Lacey & Pritchett, 2003; Campbell et al., 2008; Lim,
2012; Lim & Lye, 2012); American Dietetic Association (ADA), Oncology evidence-based nutrition practice
guideline, Chicago (IL): American Dietetic Association (ADA), 2007 Oct; American Dietetic Association,
Chronic kidney disease evidence-based nutrition practice guideline, Chicago (IL): American Dietetic
Association; 2010 Jun. http://guideline.gov/content.aspx?id=23924&search=Renal+function+study
110
1.3.1 Nutrition Support for Malnourished Patients
The first line of treatment for malnourished patients is to increase nutrient intake
using food-based approach or through food fortification (Thomas, 2001). If
dietary intake is not able to meet requirements, oral supplementation is the next
line of treatment, followed by tube feeding and finally, if the gut is not working
and there is contraindication to feed enterally, parenteral nutrition can be
explored (Thomas, 2001).
Many studies have shown that patient outcomes can be improved with adequate
nutrition support (Kaminski, 1988; Otte et al., 1989; Beattie et al., 2000; Bourdel-
Marchasson et al., 2001; Isenring et al., 2003b; Isenring et al., 2004; Paton et al.,
2004; Smedley et al., 2004; Norman et al., 2008a; Ha et al., 2010; Paccagnella et
al., 2010). There are three Level I studies on nutrition support for malnourished
patients according to the NHMRC (National Health and Medical Research
Council, 2009) criteria. A Cochrane review consisting of 36 studies (n = 2714)
compared the effect of nutritional counselling to improve food-based nutrient
intake and supplements on the clinical and nutritional outcomes in malnourished
patients (Baldwin & Weekes, 2008). In the 2008 Cochrane review, the authors
concluded that a combination of nutritional counselling plus supplements is more
effective than supplements or dietary advice alone (Baldwin & Weekes, 2008).
Their recent review on 2123 participants stated that dietary counselling given
with or without supplements is effective at increasing the nutritional intake and
weight of patients (Baldwin & Weekes, 2012). The latter review reinforced that
individualised dietary counselling is the mainstay in helping malnourished
patients improve (Baldwin & Weekes, 2012).
A systematic review of 32 RCT papers (n=2286) on oral and enteral

supplementation showed that routine supplementation improved nutritional
indices (Potter et al., 1998). In the review paper, 17 out of the 32 trials reported
changes in anthropometric measures compared to control group. The treatment
group receiving routine nutritional supplementation showed consistently
111
improved changes in body weight and anthropometry compared with controls;

weighted mean difference 2.06% (95% CI: 1.63% to 2.49%) and 3.16% (94% CI:
2.43% to 3.89%) respectively. The benefits of supplementation were not
restricted to particular patient groups (Potter et al., 1998).
Milne et al. (2009) reviewed 62 randomised and quasi-randomised controlled

trials of oral protein and energy supplementation in older people, excluding those
recovering from cancer treatment or in critical care. They concluded that with the
10,187 randomised participants, supplementation produced a small but
consistent weight gain (mean difference for percentage weight change = 2.2%,
95% CI 1.8 - 2.5), in older people ( 65 years old) who were undernourished
(Milne et al., 2009).
In another review paper of 84 studies on patients with chronic diseases in the

community (45 randomized, 39 non-randomized, n=2570), oral nutritional
supplementation had a positive effect on body weight and functional status such
as improved muscle strength and walking distance (Stratton & Elia, 2000). In
addition, there was a reduction of falls and increased ability to perform activities
of daily living in older adults. Besides the review papers, a meta-analysis of five
randomised controlled trials on 1224 older adult patients showed that oral
nutritional supplementation significantly reduced the risk of developing pressure
ulcers by 25% (Stratton et al., 2005).
In summary, there is a large and growing body of evidence to show that nutrition
support is effective and leads to improved nutritional status and clinical outcomes
in malnourished patients. For this reason, the mode (oral diet, tube feeding or
parenteral nutrition) and content (nutrient content, diet counselling or
supplements) of nutrition support will not be the focus of the PhD research.
112
1.3.2 Post-discharge Follow-up of Malnourished Patients
The nutritional status of patients malnourished on admission often worsens

during the hospital stay, with a cumulative decline in status associated with
repeated readmissions (McWhirter & Pennington, 1994; Braunschweig et al.,
2000; Norman et al., 2008b; Cansado et al., 2009). This is, at least in part,
because the short length of stay of most inpatients limits the potential impact of
inpatient nutrition interventions that typically include nutrition supplements,
dietary fortification and patient education. Outpatient dietetic follow-up post-
discharge is commonly arranged in an attempt to extend the time frame and
potential effectiveness of these interventions (Kirkland et al., 2013). However,
the issue of patients becoming lost to follow-up is common, with a dropout rate
ranging from 54-58% of total dietetic outpatient attendances (Gallagher, 1984;
Finucane et al., 2007). A literature search yielded four studies that looked at
dietetics follow-up rate (Table 1.10). Only one small study focused on follow-up
rates of malnourished patients. Van Bokhorst-de van der Schuren et al. (2005)
observed that a dietitian saw only 54% of malnourished patients during
admission. Out of these, only 23% were followed-up by a dietitian after
discharge (van Bokhorst-de van der Schueren et al., 2005).
As discussed in the preceding section (1.3.1), dietary fortification and

counselling are often used as a first-line intervention for treating malnutrition in
the hospital. However, the effect of continuing care beyond hospital walls for
malnourished patients discharged from the hospitals has not yet been
thoroughly explored. The evidence base for this practice is weak and should be
addressed with well-designed trials that assess clinically relevant outcome
measures and costs.
At National University Hospital in Singapore, malnourished inpatients referred to

dietetics are given an appointment as part of discharge planning to come back
to the outpatient dietetic clinic for follow-up. However the candidates anecdotal
experience (which would be verified by data collection as part of this thesis) was
that most of these patients did not come back to the clinic for follow-up. This
113
was despite the fact that patients or caregivers were given reminders either by
an appointment letter sent to their home address or short messaging system via
mobile phone (according to their preference indicated in the hospital registration
system) two weeks before the appointment. Possible reasons for patients failing
to attend these appointments include prolonged waiting time, emotional state of
patients, patients too weak to come to clinic, unawareness of the importance of
nutrition support and the consequences of malnutrition, perception that nutrition
support was not useful, costs, transport problems and not wanting to cause
inconvenience to family members.
114
Table 1.10: Studies on follow-up attendance rate of patients seen by dietitians

Lead Author, Attendance/Follow-up Rate at Duration of follow-
Participants (Country) n Methods to increase follow-up rate
Year Outpatient Dietetics Clinic up/ observation
46% (out of 61 Diabetic
New outpatients referred outpatients)
(Gallagher, No specific method employed for this
to dietitians 209 42% (out of 117 Weight 12 months
1984) descriptive study
(United Kingdom) reduction outpatients)
8% (other than above)
(van Bokhorst-
Malnourished inpatients
de van der No specific method employed for this
discharged from hospital 24 23% Not specified
Schueren et descriptive study
(Netherlands)
al., 2005)
Telephone reminders 1 week before

Outpatients at Diabetes appointment. Up to 3 attempts were made
(Finucane et
Dietetics Clinic 432 59% before patients were deemed 4 months
al., 2007)
(Ireland) uncontactable (274 patients were
contactable and 158 were uncontactable)
Treatment group: A structured approach

Outpatients at intensive with six once-a-month appointments, a
96% returned for second
weight management signed agreement to attend, an initial
appointment
clinics (IWMC) 75 screening of readiness to change and
53% completed programme
(United Kingdom) consistent advice from one dietitian. (Only
patients who were motivated to lose weight
(Hickson et al.,
were enrolled). 6 months
2009)
Outpatients for weight 54% returned for second

loss at general dietetic appointment Control Group: Patients were offered 5
93
clinics 19% completed programme appointments in the outpatient clinic
(United Kingdom)
115
1.3.3 Strategies to Improve Post-discharge Follow-up
Given the adverse consequences of malnutrition and likelihood of poor rates of

follow-up post-discharge, new strategies are needed to effectively manage these
patients. One possible model of care is a telephone and/or home visit follow-up
programme. Telephone-delivered interventions have emerged as an increasingly
popular means of delivering health promotion and behaviour change intervention
for patients with chronic disease (McBride & Rimer, 1999; Kay et al., 2006;
VanWormer et al., 2009). This individualised and convenient service could
improve patient compliance to follow-up. To date, there have been limited studies
published on the efficacy of telephone care and home visits to improve the
nutritional outcomes of malnourished patients discharged from hospital, with
existing research based on geriatric population (Feldblum et al., 2010), or the use
of this model of care in other settings i.e. preventing the deterioration of
nutritional status in oncology outpatients receiving radiotherapy (Isenring et al.,
2004).
The study by Feldblum looked at the effectiveness of post-discharge home visits

for 259 hospitalised adults aged 65 and older who were at nutritional risk as
determined by the SF-MNA (Feldblum et al., 2010). The intervention group
received individualised nutritional treatment from a dietitian in the hospital and
three home visits after discharge. The control group received one meeting with a
dietitian in the hospital or standard care. After 6 months, the mean change in
MNA score was significantly higher in the intervention group than in the control
groups after adjusting for education level and hospitalisation ward (3.01 2.65
vs. 1.81 2.97, p = 0.004). Mortality was significantly lower in the intervention
group than in the control group (3.8% vs. 11.6%, p= 0.046) (Feldblum et al.,
2010).
Another study provided telephone reviews as part of intensive nutrition support

between nutrition counselling sessions for oncology outpatients receiving
116
radiotherapy to the gastrointestinal or head and neck area (Isenring et al., 2004).
In this randomised control trial, the intervention group received individualised
nutrition support in the form of regular and intensive nutrition counselling by a
dietitian. Nutrition counselling by the dietitian was provided within the first 4 days
of commencing radiotherapy and weekly for the course of radiotherapy
(approximately 6 weeks) and fortnightly for the remainder of the 12-week study
period. Telephone reviews were conducted between nutrition counselling
sessions. Individually tailored sample meal plans, recipe suggestions and
techniques to minimise the side effects of the tumour and therapy were provided.
Standard patient handouts were used, as well as high energy and snack ideas,
and protein exchange lists. If deemed appropriate, the dietitian would provide a
weekly supply of oral nutrition supplements for up to 3 months. The usual care
group received general nutrition advice provided by a nurse, no individualisation
of nutrition advice and less follow-up and referral to outpatient dietitians, which
was a maximum of two dietetic consultations. The intensive nutrition intervention
group had statistically smaller deteriorations in weight, nutritional status and
quality of life compared with those receiving usual care (Isenring et al., 2004).
Both the above studies are Level II studies (National Health and Medical
Research Council, 2009) which provided good evidence for the effectiveness of
either home visits or combination of outpatient visits and telephone follow-up up.
Such interventional studies with a control arm are not easy to carry out because
of ethical concerns, designing what the control group should receive and who
should be selected for which arm. In addition, home visits may not be possible for
many healthcare institutions and countries due to intensive resources needed
with regard to manpower, travelling expenses and time. If a face to face visit
with a dietitian is crucial in contributing to the improvement of nutritional status for
malnourished patients, another strategy could be to assign home visits only for
patients who had defaulted outpatient follow-up and who had been triaged as not
progressing well nutritionally. This modality of follow-up has not been explored.
117
Summary of Issues in Post-discharge Follow-up of Malnourished Patients
There have been limited studies published on the follow-up rate and the effect of
combined telephone, outpatient and home visit follow-up on the nutritional
outcomes of malnourished patients discharged from the hospital. Specifically,
limiting home visits to patients who really need them has not been studied. A
study carried out in this area will provide evidence to the feasibility and
effectiveness of this strategy of follow-up for malnourished patients.
118
1.4 OVERALL SUMMARY AND GAPS IN CURRENT RESEARCH
This literature review has revealed significant gaps in the current body of
knowledge, which can be presented as 4 broad topics.
Firstly, a critical step in managing malnutrition is to provide evidence that there is

indeed a problem. There have been scarce data on the prevalence and
outcomes of malnutrition in Singaporean hospitals. Specifically, there are no
published studies on the consequences of malnutrition that have adjusted for
disease complexities using DRG in adult hospitalised patients. Almost all studies
that compare the mortality rate of malnourished patients had not used the
national registry to track the mortality of patients post discharge. It is important to
determine the cost and clinical outcomes arising from malnutrition in hospitalised
patients and more so to use the national registry to track mortality and control the
outcomes for confounders which may likely affect the results.
Secondly, all patients admitted to hospital should be systematically screened,

using a nutrition screening tool that is simple, quick, reliable, valid and cost
effective. Although many nutrition screening tools exist, these have mostly been
validated in the Caucasian population. No nutrition screening tool has been
developed and validated for use in Singapore (or Asia) with its particular racial
mix. As Singapore consists of a unique multi-ethnic population, the applicability
of existing nutrition screening tools is uncertain, particularly the cutoffs used to
identify those at risk of malnutrition. The selection, reporting and interpretation,
including cutoffs, of nutrition screening may differ between different racial groups,
healthcare systems and cultural contexts (de Onis & Habicht, 1996; Chumlea,
2006). Hence development and validation of a malnutrition screening tool specific
to the Singaporean healthcare context is needed.
Thirdly, for a nutrition screening tool to be effective, it must be completed fully

and accurately. Studies have reported screening incompletion and error rates of
119
28-97% (Raja et al., 2008; Wong & Gandy, 2008; Neelemaat et al., 2011b;
Geiker et al., 2012). High levels of missing data (41- 47%) was found in
commonly used nutrition screening tools (Neelemaat et al., 2011b). Failure to
achieve accurate and complete nutrition screening will affect the final score
allocated to a patient, which may result in a malnourished or at risk patient not
being referred for nutritional intervention. To complete the process of nutrition
screening, any patient identified from screening to be malnourished or at risk of
malnutrition should be referred to receive a full nutrition assessment and
intervention (American Dietetic Association., 1994; Joint Commission
International, 2008). To date, there has not been any published study on the use
of quality improvement tools in addressing ways to improve the compliance rate
of nutrition screening and to improve the effectiveness of referral process for
patients identified as malnourished or at risk of malnutrition.
Fourthly, after a malnourished or at risk patient is identified and referred, nutrition

assessment has to be carried out to confirm and diagnose malnutrition, followed
by appropriate nutrition intervention by a dietitian. A review of the literature has
shown that SGA is a widely validated and accepted method to assess nutritional
status (Table 1.8), and that nutritional intervention has proven beneficial in
improving the outcomes of malnourished patients (section 1.3.1). However, to
comprehensively manage malnourished patients, ongoing monitoring and/or
intervention is required. From overseas studies, many patients are lost to follow-
up (Table 1.10) and we do not know as yet the rate of follow-up for malnourished
patients discharged from the hospitals in Singapore. In this area, there is also
limited evidence on effective methods of follow-up for post-discharge
malnourished patients.
120
1.5 CONCEPTUAL FRAMEWORK OF RESEARCH PROGRAMME
Based on the literature review which has been presented in this chapter, a
conceptual framework is derived (Figure 1.6). The conceptual framework
includes the overarching research questions which will be addressed by this
research programme.
In the conceptual framework, it is proposed that a validated nutrition screening

tool should be used to screen all newly admitted patients. Individuals identified to
be at risk of malnutrition through nutrition screening should be referred in a timely
manner for a detailed nutrition assessment to diagnose the presence and
severity of malnutrition as a prerequisite for treatment. Thereafter, confirmed
cases should be provided with appropriate nutrition intervention in the hospital.
Due to the short length of hospital stay, these patients should be followed up
post-discharge to ensure positive nutrition and clinical outcomes. Telephone
consult can be employed to follow-up on the patients. From the tele-consult, if
nutritional problems still persist, patients should be encouraged to return for
follow-up with the dietitians at the outpatient clinic so that the dietitians are able
to reassess the patients and modify the treatment plan as necessary. If these
patients fail to turn up at the outpatient clinic, home visits by the dietitians can be
effected. Patients should be reassessed and have their nutrition care plan
modified if needed in the outpatient setting or in the community. The same
assessment tool used to diagnose malnutrition should be used to monitor the
patients; and evaluate the treatment and outcomes of the nutrition intervention.
121
Figure 1.6: Conceptual Framework for the Identification and Management of Malnutrition in Hospitalised Patients
WHAT IS THE PREVALENCE OF MALNUTRITION IN A SINGAPORE HOSPITAL?
Nutrition Clinical Outcome:

Nutrition Screening Positive Intervention/ Follow-up Nutritional - Mortality rate
Assessment Post-
screening Treatment Outcome - Readmission rate
(At risk) Malnourished discharge Monitoring - Length of hospital stay
- Complications
Referral to (pressure ulcers /
Infections)
dietitian
Functional Outcome:
- Quality of life
- Handgrip strength
Poor
Intake Economic Outcome
Well - Cost of hospitalisation
Negative
nourished
screening Re-assess and modify
nutrition care if needed
Ongoing monitoring of intake by nurses
How do we best Is a validated Is nutrition What is the What is the

identify patients at nutrition support rate of dietetics effect
nutrition risk on assessment tool effective? follow-up of on outcomes?
admission to a Is referral used to diagnose malnourished
hospital in process for malnutrition? patients post-
Singapore? at risk discharge? Is
Is a validated & patients there a better
reliable nutrition effective? approach to
screening tool follow up?
used?
What is the
compliance to
nutrition
screening?
Signifies significant gaps in the current body of knowledge
122
From the conceptual framework, the following questions were raised:
1) What is the prevalence of malnutrition in a Singapore hospital?

2) How do we best identify patients at nutrition risk on admission to a hospital
in Singapore? Is a validated & reliable nutrition screening tool used? What
is the compliance to nutrition screening? Is the referral process for at risk
patients effective?
3) Is a validated nutrition assessment tool used to diagnose malnutrition?
4) Is nutrition support effective?
5) What is the rate of dietetics follow-up of malnourished patients post-
discharge? Is there a better approach to follow up?
6) What is the effect on outcomes?
Questions 1, 2 and 5 (circled) within the conceptual framework in Figure 1.6

represent significant gaps found in the current body of knowledge, which have
been covered in section 1.4. These led to five research questions which will be
presented in the next section.
123
1.6 RESEARCH QUESTIONS AND OBJECTIVES
The research questions address the significant gaps in current research, which
have been presented and discussed in sections 1.4 and 1.5. Each research
question is then addressed by a study with specific aims.
Research Question 1 (RQ 1):
What is the prevalence of malnutrition in a Singapore hospital and its

prospective impact on clinical outcomes?
Objective 1
To determine the prevalence of malnutrition on admission to a tertiary

stay, readmission and 3-year mortality.
How do we best identify patients at nutrition risk on admission to a hospital in

Singapore?
Objective 2
To develop and validate a new nutrition screening tool against Subjective

Global Assessment for use in the Singaporean adult population admitted
to an acute hospital.
124
Is the new nutrition screening tool valid and reliable when administered by the
intended users (nurses)?
Objective 3
To confirm the reliability and validity of the new nutrition screening tool
administered by nurses against Subjective Global Assessment in a new
cohort of patients.
What is the compliance of nurses with nutrition screening and are patients
screened as at risk referred to the dietitians? What measures can improve
compliance with nutrition screening and its referral rate?
Objective 4
To investigate the compliance rate of nurses in conducting nutrition

screening and referring at risk patients to the dietitians; and determine
the effect of quality improvement initiatives in improving the overall
performance of nutrition screening.
What is the rate of dietetics follow-up of malnourished patients post-discharge?

Is there a better approach to follow up?
Objective 5
To explore and determine the effectiveness of the current system and an

alternative model on dietetics follow-up rate of malnourished hospital
125
1.7 AIMS OF RESEARCH PROGRAMME
The proposed research programme will address significant gaps in the current
research and contribute to the Singaporean evidence-base. It will assist key
stakeholders to recognise the importance of managing malnutrition, and to
regard nutrition intervention as part of a holistic care plan for malnourished
patients or those at risk.
The aims of the research programme are to:

i) determine the prevalence of malnutrition on admission to a tertiary
stay, readmission and 3-year mortality,
ii) develop and validate a new nutrition screening tool for use in the
Singaporean adult population admitted to an acute hospital,
iii) confirm the reliability and validity of the new nutrition screening tool
administered by nurses in a new cohort of patients,
iv) investigate the compliance rate of nurses in conducting nutrition
screening and referring at risk patients to the dietitians; and determine
the effect of quality improvement initiatives in improving the overall
performance of nutrition screening and
v) explore and determine the effectiveness of the current system and an
alternative model on dietetics follow-up care for malnourished hospital
126
Chapter 2: LINKING THE RESEARCH

QUESTIONS
2.1 LINKING THE RESEARCH QUESTIONS AND OBJECTIVES
This thesis seeks to address major challenges and gaps in the healthcare
system to screen, manage and treat malnutrition in hospitalised patients. Is
malnutrition prevalent in newly hospitalised patients and what is the impact on
clinical outcome and cost? Is our screening system robust enough to identify
these patients? Following screening, is the referral process efficient enough to
channel these patients to the appropriate professionals? And finally is our
nutritional intervention effective beyond the hospital walls after the patient has
been discharged home?
Table 2.1 provides a brief overview of how the research questions and
objectives are addressed in the five research papers that make up this thesis.
The subsequent chapters will consist of separate research papers. All the
papers have either been published or accepted by peer reviewed indexed
journals.
2.2 HOW THE RESEARCH PROJECTS ARE RELATED
The five research projects are closely related to one another, and will address
the research questions initially set out under the broader conceptual framework.
As the papers follow the usual format of scientific publications; namely
comprising sections for introduction, methodology, results, discussion and
conclusion, they are not discussed in detail within each chapter in order to
minimise repetition. However each chapter will include an introduction and
conclusion to link the research papers. Due to the difference in elapsed time
127
needed for study implementation, data collection, data analysis, writing the
manuscripts and the turnaround time for successful publication, the papers were
not published in the chronological sequence of the research questions they
address. However for logical flow and ease of reading, the papers have been
arranged in their logical order. The format of each paper follows that of the
specific journal in which it was published, including referencing and English style
(British or American).
128
Table 2.1: Research questions, objectives and relevant publications

RQ # Research Questions Objectives of Research Research Papers
(RQ)
RQ 1 What is the prevalence Objective 1 Malnutrition and its
of malnutrition in a To determine the prevalence impact on cost of
Singapore hospital? of malnutrition on admission hospitalisation, length
What is the prospective to a tertiary hospital in of stay, readmission
impact on clinical Singapore and its impact on and 3-year mortality
outcomes? cost of hospitalisation, length (Lim et al., 2012; Lim
of stay, readmission and 3- & Daniels, 2013)
year mortality
RQ 2 How do we best Objective 2 Development and
identify patients at To develop and validate a validation of 3-Minute
nutrition risk on new nutrition screening tool Nutrition Screening
admission to a hospital against Subjective Global (3-MinNS) Tool for
in Singapore? Assessment for use in the acute hospital
Singaporean adult population patients in Singapore
admitted to an acute hospital (Lim et al., 2009)
RQ3 Is the new nutrition Objective 3 Validity and reliability
screening tool valid To confirm the reliability and of 3-Minute Nutrition
and reliable when validity of the new nutrition Screening (3-MinNS)
administered by the screening tool administered administered by
intended users by nurses against Subjective nurses
(nurses)? Global Assessment in a new (Lim et al., 2013a)
cohort of patients
RQ 4 What is the compliance Objective 4 Improving the
of nurses with nutrition To investigate the compliance performance of
screening and are rate of nurses in conducting nutrition screening
patients screened as nutrition screening and through continuous
at risk referred to the referring at risk patients to the quality improvement
dietitians? What dietitians; and determine the initiatives
measures can improve effect of quality improvement (In press, 2014)
compliance with initiatives in improving the
nutrition screening and overall performance of
its referral rate? nutrition screening
RQ 5 What is the rate of Objective 5 A pre-post evaluation
dietetics follow-up of To explore and determine the of an ambulatory
malnourished patients effectiveness of the current nutrition support
post-discharge? Is system and an alternative service for
there a better approach model on dietetics follow-up malnourished
to follow up? rate of malnourished hospital patients post hospital
patients post-discharge. discharge: a pilot
study
(Lim et al., 2013b)
129
2.3 OVERVIEW OF RESEARCH METHODOLOGY
The study programme was conducted in four phases and summarised in Table
2.2.
i. a cross-sectional 3-year prospective study phase in which the data was
used to determine the prevalence and outcomes of malnutrition. This
phase also involved development and validation of a new nutrition
screening tool called 3-Minute Nutrition Screening (3-MinNS) (CHAPTER
3 AND CHAPTER 4)
ii. a cross-sectional prospective study to confirm the validity and determine
the reliability of 3-MinNS when used by nursing staff, the intended users of
the tool (CHAPTER 5)
iii. a 6-year audit and quality improvement study to improve nurses
compliance with 3-MinNS (CHAPTER 6)
iv. a prospective interventional cohort study to evaluate the effectiveness of
an alternative ambulatory model of nutrition support for malnourished
patients discharged from the hospital (CHAPTER 7)
Table 2.2: Summary of study design, sample size and sampling strategy
Study Sampling
Phase Design n
(Chapter) strategy
1
Prospective
(Chapter 3)
1 818 Consecutive
2
Cross sectional
(Chapter 4)
3
2 Cross sectional 121 Consecutive
(Chapter 5)
4 Prospective quality
3 4467 Consecutive
(Chapter 6) improvement study
Retrospective (Year 2008) 261

5
4 Prospective pre-post study 163 Consecutive
(Chapter 7)
(Year 2010)
130
An overview of methodology for the research programme is presented in Figure

2.1, with each method explained in detailed within the published papers.
In phase one, nutrition assessment using Subjective Global Assessment (SGA)

was conducted on newly admitted patients aged 18 years and above, to
determine the prevalence of malnutrition in a Singapore acute hospital (n=818).
The clinical outcomes of these patients were prospectively tracked using a
Diagnosis-Related Groups (DRG) matched case control (well-nourished vs.
malnourished) design over 3 years and the results were adjusted for gender,
age and ethnicity. The same cohort of 818 patients was also screened using five
parameters that contribute to the risk of malnutrition. The parameters were
presence of unintentional weight loss in the past six months, intake in the past
one-week, body mass index, disease with nutrition risks and the presence of
muscle wasting in the temporalis and clavicular areas. The dietitian
administering SGA was blinded to the results of the nutrition screening
completed by a second dietitian. The sensitivity and specificity of individual
nutrition parameters as well as their different combinations were established
using the Receiver Operator Characteristics (ROC) curve. The best cutoff score
to identify malnourished patients or those at risk of malnutrition were determined
using SGA as a reference tool. The nutrition parameter (or its combination) with
the largest Area Under the ROC Curve (AUC) was chosen as the final screening
tool, which was named the 3-Minute Nutrition Screening (3-MinNS).
In phase two, three ward-based nurses administered the 3-MinNS to a new

group of patients within 24 hours of admission (n=121). A dietitian blinded to
these results conducted a nutrition assessment using SGA. To assess the
reliability of 3-MinNS, 37 patients screened by the first nurse were re-screened
by a second nurse within 24 hours, who was blinded to the results of the first
nurse. Receiver operator characteristic (ROC) curve analysis was performed to
determine the sensitivity and specificity of 3-MinNS when compared to SGA
(validity). The positive predictive value (PPV) and negative predictive value
131
(NPV) for 3-MinNS were determined. Kappa score was used to determine the
inter-rater reliability of 3-MinNS between the nurses.
In phase three, annual audits were carried out from 2008-2013 to investigate the
incompletion and error rates of 3-MinNS (n = 4467). Quality improvement tools
such as Value Stream Mapping and the Plan-Do-Check-Act cycle were applied
in this phase. Root cause analysis was used to determine the best action plan.
Hospital-wide action plan implemented from 2009-2011 included 1) nutrition
screening training incorporated as part of nurses orientation programme, 2)
Nutrition Screening Protocol was made accessible to all staff via the staff
intranet, 3) nurse empowerment for online dietetics referral of at risk cases, 4)
closed-loop feedback system and 5) removing a component in the nutrition
screening that caused the most error without compromising sensitivity and
specificity, a decision supported by phase 2 study results.
In phase four, the effectiveness of the current system in following up

malnourished patients discharged from the hospital was audited retrospectively
on a consecutive sample of malnourished inpatients referred to dietetics
(n=261). The results were used to develop a novel model of care called
Ambulatory Nutrition Support (ANS) to follow-up these patients post-discharge.
Ambulatory Nutrition Support provided a combination of outpatient review,
telephone calls and home visits. The effectiveness of ANS was evaluated via a
prospective cohort study of adult inpatients referred to dietetics and assessed as
malnourished using Subjective Global Assessment (n=163). All subjects
received inpatient nutrition intervention and four months of ANS. Subjective
Global Assessment, body weight, quality of life using the Euro Quality of Life - 5
Domain Visual Analogue Scale (EQ-5D VAS) and handgrip strength were
measured at baseline and five months post-discharge. Paired t-test was used to
compare pre- and post-intervention results.
132
Figure 2.1: Schematic diagram on the methodology of research programme
Phase 1 (n=818) Recruitment

(CHAPTER 3 & 4) Inclusion criteria:
18-74 years old
Not enrolled during previous admission
Not Paediatric, Psychiatry, ICU & Maternity cases
Nutrition Screening (Dietitian I) Nutrition Assessment (Dietitian II)

Testing of 5-items to determine Subjective Global Assessment (SGA) to determine
Blinded
nutritional risk patients nutritional status
within 24 hours of admission within 48 hours of admission
Not at risk At risk Well-nourished Malnourished
Development of 3-Minute Nutrition Screening Tool Prospective Outcomes Tracking

Compares with reference standard (SGA) Length of hospital stay (LOS)
Unplanned readmissions
Cost of hospitalization before government subsidy
Mortality at 1, 2 & 3 years from index admission
(Source: Hospital System & Singapore Death Registry)
Statistical Analyses Statistical Analyses

The sensitivity and specificity were Mixed Model Analysis to analyze difference between well- nourished
established using the Receiver Operator and malnourished groups with DRG as random effect
Characteristics (ROC) curve and the best Conditional Logistic Regression matching by DRG to evaluate the
cutoff scores determined. The nutrition association between nutritional status and outcomes
parameter with the biggest Area Under the (All results were controlled for gender, age and race and disease type and
ROC Curve (AUC) was chosen as the final complexity using DRG)
3-Minute Nutrition Screening tool (3-MinNS). (Participants who had been classified with a malnutrition sub-component in the
DRG were reclassified with the corresponding DRG without malnutrition so that
matching between similar codes of DRG can be performed)
Phase 2 (n=121)
(CHAPTER 5) Recruitment
Inclusion criteria:
21 years old
Oncology & Surgical wards
Not enrolled during previous admission
SGA (1 Dietitian)
3-MinNS (Nurse I)
Blinded
3-MinNS (Nurse II & III) Blinded Subjective Global Assessment (SGA) to determine
Within 24 hours of admission Within 24 hours of admission patients nutritional status
within 48 hours of admission
Statistical Analyses Statistical Analyses

Kappa score to determine the inter-rater ROC curve analysis to determine the sensitivity and specificity
reliability of 3-MinNS among the nurses. of 3-MinNS when compared to SGA done. Spearman rho to
determine the correlation between 3-MinNS and SGA.
Phase 3 (n=4467) Annual Audits on Nutrition Screening (Hospital-wide)

(CHAPTER 6) Inclusion criteria:
All patients admitted to National University Hospital during audit periods
21 years old, Not Paediatric cases
Year 2008 Year 2009 Year 2010 Year 2011 Year 2012 & 2013
Baseline data Baseline data CQI Post- CQI Post- Sustenance Post-
implementation implementation implementation
data data data
CQI = Continuous Quality Improvement Initiatives
133
Phase 4 (n= 261 & 163)

(CHAPTER 7)
Recruitment
Inclusion criteria:
Patients referred to dietitian, assessed to be malnourished using SGA and have been provided with inpatient
nutrition counselling and support,
Age 21 years
Not on tube feeding or total parenteral nutrition,
Not Psychiatry, Maternity, Palliative Care patients
Not patients discharged to nursing home or community hospital
2008 cohort (n = 261) 2010 cohort (n = 163)
Baseline measurements (4 days before discharge)

Outpatient follow-up
Weight and height
Dietetic outpatient appointment arranged for
SGA
one-month post-discharge
Mid arm anthropometry
Reminder sent by Short Messaging System EQ-5D Visual Analogue Scale (QOL)
(SMS) or letter one-week prior to Handgrip strength
appointment
4-month Ambulatory Nutrition Support

Telephone calls at 1 week, 2 and 4 months post-discharge
Dietetic outpatient appointments at 1 and 3 months post-
discharge
Patients failing to attend either outpatient appointments were
telephoned, with home visit provided if required
Data collection Data collection

Measurement of follow-up rate post- Measurements (same as at baseline) at 5 months post-
discharge from hospital discharge
Measurement of follow-up rate post-discharge from hospital
Statistical Analyses
Paired t-test to compare baseline and post-intervention results.
134
Chapter 3: PREVALENCE OF MALNUTRITION

AND ITS IMPACT ON CLINICAL
OUTCOMES AND COSTS
Publications:
readmission and 3-year mortality. Clinical Nutrition 2012. 31(3):345-350.
(IF = 1.362)
2. Lim SL, Daniels L. Reply - Malnutrition and its impact on cost of

hospitalisation, length of stay, readmission and 3-year mortality. Clinical
Nutrition 2013; 32(3):489-490. (IF = 1.362)
3.1 Introduction
There is a lack of evidence on the prevalence and prospective outcomes of

malnutrition in Singaporean hospitals. What is the evidence of there being a
problem of malnutrition in the hospital? What is the magnitude of the problem?
Providing evidence on the prevalence and outcomes of malnutrition in
hospitalised patients in Singapore is the first step in defining the problem.
Hence, this study was undertaken to establish the prevalence of malnutrition
and its impact on clinical outcomes and cost in hospitalised patients in
Singapore.
It is commonly believed that any association of malnutrition with poor outcomes

is due to underlying chronic diseases (Elia, 2006). Although many overseas
studies have been done on clinical outcomes of malnutrition, none have used
Diagnosis-Related Groups (DRG) to control for disease and its complexities.
The extent of malnutrition and its independent effect on outcomes must be
determined and understood prior to requesting for resources, planning and
developing any intervention.
135
The paper in this chapter underscores the prevalence of malnutrition in a tertiary

hospital in Singapore and its impact cost of hospitalisation, length of stay,
readmission and 3-year mortality, controlling for DRG.
3.2 Publication
Refer to paper or go to: 1) http://eprints.qut.edu.au/50643/

2) http://eprints.qut.edu.au/59962/
3.3 Conclusion
Malnutrition was evident in up to one third of inpatients in an acute hospital in

Singapore. After controlling for the potential confounders of age, gender,
ethnicity and diagnosis- related groups, malnourished patients stayed in the
hospital 1.5 times longer than well-nourished patients (p = 0.001) and were
almost twice as likely as well-nourished patients to be readmitted within 15 days
of discharge (p=0.025). They incurred 24% more cost than well-nourished
patients. Even after controlling for the confounders mentioned above,
malnutrition posed almost four-fold and three-fold increases in the risk of death
at 1-year and 3-year post-discharge, respectively (HR = 4.4, CI 3.3-6.0,
p<0.001). As such, this paper provides strong support that malnutrition is a
condition that must be addressed, and there is an urgent need for strategies to
prevent and treat malnutrition in hospitalised patients.
136
Due to copyright restrictions, the published version
of this journal article is not available here. Please
view the published version online at:
http://dx.doi.org/10.1016/j.clnu.2011.11.001
137
http://dx.doi.org/10.1016/j.clnu.2012.12.014
143
Chapter 4: DEVELOPMENT AND VALIDATION

OF 3-MINUTE NUTRITION
SCREENING (3-MinNS)
Publication:
Lim SL, Tong CY, Ang E, Lee EJ, Loke WC, Chen Y, Ferguson M, Daniels L.
Development and validation of 3-Minute Nutrition Screening (3-MinNS) Tool for
acute hospital patients in Singapore. Asia Pacific Journal of Clinical Nutrition.
2009;18(3):395-403. (IF = 1.438)
4.1 Introduction
From Chapter 3, we know that malnutrition is prevalent in a Singapore hospital

and leads to adverse outcomes, independent of illness and its complexity. About
one third of hospitalised patients are already malnourished upon admission.
Therefore, it is important to screen newly admitted patients so that those
identified to be at nutritional risk are systematically referred for timely nutrition
assessment and intervention.
In order to prevent and treat malnutrition, we must first know who is at risk.
Although there are a number of screening tools available, limitations exist for
each, as discussed in Chapter 1 and presented in Table 1.4. Most tools were
developed overseas and validated in the Caucasian population and their
suitability for use in the Singaporean multi-ethnic, large tertiary hospital has not
been established. The selection, reporting and interpretation, including cutoffs,
of nutrition screening may differ between different racial groups, healthcare
systems and cultural contexts (de Onis & Habicht, 1996; Chumlea, 2006). The
complexity and time-intensive nature of many nutrition screening tools such as
in NRIb (Wolinsky et al., 1985), Nursing Nutritional Assessment Tool (Scanlan et
al., 1994) and Screening Nutritional Profile (Hunt et al., 1985a) render them
impractical in Singapore, where patient to nurse ratio was high (7 patients to 1
145
nurse in the year 2002 and before). In addition to this, feedback sessions with
nursing staff identified ambiguity in their understanding of how to score a patient
when administering nutrition screening such as the Derby Nutritional Score
(Goudge et al., 1998). For example, what does loss of appetite mean? How
much of a decrease in food intake is considered significant? The MUST
(Stratton et al., 2004), SNAQ (Kruizenga et al., 2005a) and NRS 2002 (Kondrup
et al., 2003b) had not been published when the candidate developed the NUH
Nutrition Screening Tool in 2002. Although the Malnutrition Screening Tool
(MST) (Ferguson et al., 1999a) perhaps may have been a suitable tool for
Singapore, this was not readily accessible at that time. In addition, many
screening tools such as Derby Nutritional Score, MUST, NRS 2002, MNA-SF
and Nutrition Risk Score (Reilly et al., 1995), which require patients weight,
height and body mass index may not be completed in a sizeable proportion of
patients (Wong & Gandy, 2008; Porter et al., 2009). No nutritional screening tool
had yet been properly validated for hospitalised adults in Singapore. Hence,
development and validation of an effective nutrition screening tool specific to the
Singaporean healthcare context was mooted. The original screening tool was
created in 2002, and was subsequently revised as a result of a validation study
within this research programme, which is presented in this chapter.
The paper in this chapter addresses the research question How do we best
identify patients at nutrition risk on admission to a tertiary hospital in
Singapore? The aim of the study was to develop and validate a new nutrition
screening tool against Subjective Global Assessment for use in the Singaporean
adult population admitted to an acute hospital.
4.2 Publication
Refer to paper or go to: http://eprints.qut.edu.au/29117/
146
4.3 Conclusion
With the validation of 3-MinNS carried out on consecutive patients via an

independent, blinded comparison with SGA, which is a valid reference standard,
this study has provided the first Level II evidence under NHMRC (National
Health and Medical Research Council, 2009) criteria for nutrition screening in
hospitalised patients in Singapore. The 3-Minute Nutrition Screening tool was
both sensitive (86%) and specific (83%) in detecting nutritional risk in newly
admitted patients, with three as the best cutoff score. It has the added
advantage of summative scoring and cutoff points useful to define a protocol for
subsequent action. As the 3-MinNS is a simple and efficient tool to administer to
acute hospital patients in Singapore, it will assist healthcare workers to carry out
nutrition screening accurately and promptly so that patients who require nutrition
intervention can be managed appropriately.
147
http://apjcn.nhri.org.tw/server/APJCN/18/3/395.pdf
148
Chapter 5: VALIDITY AND RELIABILITY OF

3-MINUTE NUTRITION SCREENING
ADMINISTERED BY NURSES
Publication:
Lim SL, Ang E, Foo YL, Ng LY, Tong CY, Ferguson M, Daniels L. Validity and
reliability of nutrition screening administered by nurses. Nutrition in Clinical
Practice. 2013; 28:730-736. (IF = 1.594)
5.1 Introduction
The paper in Chapter 4 focused on determining the validity and reliability of 3-

Minute Nutrition Screening (3-MinNS), using one dietitian to administer nutrition
screening and another to conduct SGA as the reference standard (Lim et al.,
2009). Not only must the tool be valid but validation itself should ideally be
performed using the intended assessors. A survey carried out in the US,
showed that 84% of nutrition screening was administered by nursing staff
(Chima et al., 2008). Although in practice nutrition screening is commonly
undertaken by nursing staff, few studies have used nurses to validate nutrition
screening tools. Most studies have used researchers, dietitians or dietetic staff
to implement nutrition screening (Ferguson et al., 1999a; Kyle et al., 2006; Kim
et al., 2011; Velasco et al., 2011). As nutrition screening is usually administered
by nurses, it is critical that the reliability and validity of 3-MinNS be established
in this user group. Knowing that the 3-MinNS tool would eventually be used by
nurses to screen all newly admitted patients in NUH, it was only practical to let
the intended assessors (which would be the nurses) perform the screening tool
in this study. Hence, the paper in this chapter seeks to test the validity of 3-
MinNS when used by nurses on a new sample of patients. This paper also
provides new information on the reliability of the tool among nurses that was not
addressed in the earlier paper.
156
The paper in this chapter addresses the research question Is 3-MinNS valid
and reliable when used by the intended users (nurses)? The aim of the study
was to confirm the reliability and validity of the new nutrition screening tool
administered by nurses against SGA in a new cohort of patients.
5.2 Publication
5.3 Conclusion
The results of this study confirm the validity of 3-MinNS when used by nurses.
The sensitivity and specificity of 3-MinNS were even better than the earlier study
(89% and 88% respectively) when administered by the nurses. There was also
good agreement between nurses administering the tool (reliability = 78.3%, k =
0.58, p<0.001). In summary, the last two chapters (Chapters 4 and 5) have
provided an evidence-base that the 3-MinNS is a sensitive, specific and reliable
tool that can be administered by dietitians and nurses on newly admitted
hospitalised patients to identify patients at risk of malnutrition.
157
502812
research-article2013
NCPXXX10.1177/0884533613502812Nutrition in Clinical PracticeLim et al

http://dx.doi.org/10.1177/0884533613502812
158
Chapter 6: IMPROVING THE PERFORMANCE

OF NUTRITION SCREENING
Publication:
Lim SL, Ng SC, Lye J, Loke WC, Ferguson M, Daniels L. Improving the
performance of nutrition screening through continuous quality improvement
initiatives. The Joint Commission Journal on Quality and Patient Safety 2014. (In
press)
6.1 Introduction
The previous two chapters established that the 3-MinNS is a valid and reliable
nutrition screening tool for newly hospitalised patients in Singapore. However for
a nutrition screening tool to be effective, beyond being validated and reliable, it
must be completed fully and accurately. It will not serve its purpose well if not
done correctly or if patients identified as being at nutritional risk are not given
intervention. Addressing compliance with nutrition screening and ensuring that
malnourished patients are referred in a timely manner are critical steps to
ensure these patients receive early nutrition intervention. It is important to
conduct process evaluation of implemented nutrition screening tools as part of
quality assurance. The literature reveals high levels of screening incompletion
and error rates as discussed in section 1.2.5 and summarised in Table 1.9.
Incompletion and error rates range from 30-97% for a range of commonly-used
screening tools. No published studies have described the use of quality
improvement tools to facilitate improvements in the compliance to nutrition
screening and its referral process. Hence, this topic is ripe for research. The
paper in this chapter evaluates quality improvement initiatives to improve the
rate of compliance and referral, which is a considerable challenge for a large
tertiary hospital employing over 3200 nursing staff.
The paper seeks to address these research questions: What is the compliance
of nurses with nutrition screening and are patients screened as at risk referred
to the dietitians? What effective measures can improve compliance with nutrition
165
screening and its referral rate? It aims to investigate the compliance rate of
nurses in conducting nutrition screening and referring at risk patients to the
dietitians; and determine the effect of quality improvement initiatives in
improving the overall performance of nutrition screening.
6.2 Publication
Refer to paper or go to QUT ePrints (paper available from April 14 onwards)
6.3 Conclusion
This chapter shows that quality improvement initiatives were effective in

reducing the incompletion and error rates of nutrition screening, and led to
sustainable improvements in the referral process of patients at nutritional risk.
After the implementation of the quality improvement action plan, error rates were
reduced to from 33% in 2008 to 7% and 5% in 2012 and 2013 respectively.
From the audit database and root cause analysis, we were able to understand
the areas which caused the most errors. We were then able to devise action
plans, which when implemented, reduced the error rates substantially. Blank or
missing nutrition screening forms came down from 8% to 1%, with sustained
results for four consecutive years. Patients scored as being at risk of
malnutrition but were not referred to a dietitian was reduced from 10% in 2008 to
3% in 2012 and 2013. Direct online referral by the nurses to the dietitians was
the most effective initiative in improving the referral rates of patients at risk of
malnutrition.
166
Due to restrictions on availability prior to
print publication, this article cannot be made
available here.
Please view the journal's website at:
http://store.jcrinc.com/the-joint-commission-
journal-on-quality-and-patient-safety/
167
Chapter 7: FOLLOW-UP FOR MALNOURISHED

POST-DISCHARGED HOSPITAL
PATIENTS
Publication:
Lim SL, Lin XH, Chan YH, Ferguson M, Daniels L. A pre-post evaluation of an
ambulatory nutrition support service for malnourished patients post hospital
discharge: a pilot study. Annals Academy of Medicine Singapore. 2013; 42:507-
513. (IF = 1.362)
7.1 Introduction
The first study in this research programme has indicated that malnutrition is
prevalent in newly hospitalised patients and leads to adverse outcomes
(Chapter 3). A robust screening process has been put in place to identify these
patients so that they receive nutritional intervention (Chapters 3, 4 and 5).
However in Singapore and many other countries, the short length of hospital
stay (averaging 4 days) (OECD., 2011) limits the scope of inpatient nutrition
intervention for malnourished patients, with improvements in nutritional status
unlikely to be seen in this short time frame despite receive nutritional
intervention and follow up from dietitians in the wards. These patients potentially
return to the community malnourished, and are often readmitted, causing a
vicious cycle. Therefore, it is imperative that we follow-up these patients after
they are discharged for ongoing monitoring and treatment. Discharged patients
are often given follow-up appointments to return to the clinic to see the dietitian.
But the current literature showed that the issue of patients becoming lost to
follow-up is common, with a dropout rate ranging from 54-58% of total dietetic
outpatient attendances (refer to section 1.3.2 and Table 1.10). Only one small
study carried out in Netherlands on 24 patients focused on follow-up rates of
malnourished patients. The study reported that 77% of malnourished patients
seen in the wards were lost to Dietetics follow-up after discharge (van Bokhorst-
de van der Schueren et al., 2005). There is obviously a gap in practice and a
gap in the current literature.
191
Prospective studies are also lacking that show improvement in nutritional

outcomes if appropriate nutrition support and follow-up are provided for
malnourished discharged patients. Hence, the research questions in this chapter
are What is the rate of dietetics follow-up of malnourished patients post-
discharge? Is there a better approach to follow up? The paper aims to explore
and determine the effectiveness of the current system and an alternative model
on dietetics follow-up rate of malnourished hospital patients post-discharge.
7.2 Publication
7.3 Conclusion
This pilot research provides initial evidence that an Ambulatory Nutrition Support
(ANS) service consisting of clinic appointments, telephone calls and home visits
provides an effective model of follow-up for malnourished hospital patients post-
discharge, and is able to improve nutritional outcomes in this patient group. The
ANS service achieved 100% follow-up of malnourished inpatients within four
months of discharge from the hospital; a substantial improvement compared to
the 15% follow-up rate pre-ANS. More importantly, the nutritional indicators,
quality of life and functional status improved significantly in patients who
underwent ANS. Mean weight improved from 44.0 8.5kg to 46.3 9.6kg, EQ-
5D VAS from 61.2 19.8 to 71.6 17.4 and handgrip strength from 15.1 7.1
kg force to 17.5 8.5 kg force (p<0.001 for all). Seventy-four percent of patients
had an improvement in their SGA score and 67% improved in their quality of life.
192
Due to copyright restrictions, the published version of
this journal article is not available here.
Please view the publishers website online at:
http://www.annals.edu.sg/pastIssue.cfm?pastMonth=
10&pastYear=2013
193
Chapter 8: DISCUSSION AND

RECOMMENDATIONS
8.1 OVERVIEW OF RESEARCH QUESTIONS AND KEY FINDINGS
This chapter presents an overview of the results of the studies, highlights the
significance of the research, its contributions to scientific knowledge as well as
its impact to practice. The strengths and limitations of the research will also be
presented, followed by recommendations.
The over-arching research questions were:
1. What is the prevalence of malnutrition in a Singapore hospital and its

prospective impact on clinical outcomes and cost?
2. How do we best identify patients at nutrition risk on admission to a

hospital in Singapore?
3. Is the new 3-MinNS tool valid and reliable when used by the intended
users (nurses)?
4. What is the compliance of nurses with nutrition screening and are

patients screened as at risk referred to the dietitians? What measures
can improve compliance with nutrition screening and its referral rate?
5. What is the rate of dietetics follow-up of malnourished patients post-

discharge? Is there a better approach to follow-up malnourished hospital
patients post-discharge?
200
8.1.1 Prevalence of malnutrition and outcomes
The prevalence of malnutrition in a Singapore tertiary teaching hospital was one

third of newly hospitalised adult patients. In this study on 818 patients newly
admitted to the hospital, the highest prevalence of malnutrition was found in
patients from oncology (71%), endocrinology (48%) and respiratory medicine
(47%). Malnourished patients tended to be older (58 years vs. 49 years, p <
0.001, 95% CI 7-11) and were more likely to be male (32% vs. 26%, p=0.016).
There was no statistical difference in the prevalence of malnutrition among
different races in Singapore. After controlling for the potential confounders of
gender, age, race and disease complexities using DRG, analysis of the data
found that malnourished patients stayed in the hospital longer and were more
likely to be readmitted within 15 days of discharge. Malnutrition was a significant
predictor of overall mortality, and the effect was pronounced even at three years
post-discharge. The average cost of hospitalisation was higher for malnourished
patients.

malnutrition and clinical outcomes, the confounding effect of disease and its
complexity has seldom been taken into consideration (Chima et al., 1997;
Middleton et al., 2001). While it is widely agreed that disease and malnutrition
are closely linked (Jeejeebhoy, 2000; Pirlich et al., 2003), it has been argued
that LOS, mortality and hospitalisation costs are primarily determined by the
patients medical condition, and any association with malnutrition is due to
confounding (Elia, 2006). In this study, the matched case control design based
on DRG showed that malnutrition was an independent predictor of length of
hospital stay, readmission, hospitalisation cost and mortality.
To date, this is the only study which has tracked the mortality of hospitalised
patients for 3 years post-discharge using national death registry data and
matching for DRG. Malnourished patients had four times and three times higher
201
risk of death at one year and three year post-discharge, respectively (adjusted
hazard ratio = 4.4, 95% CI 3.3-6.0, p < 0.001). A similar study using national
registry data found a comparable mortality in malnourished patients at 1 year
(30% versus 34% in this study), but did not control for confounders (Middleton et
al., 2001). Other studies that associated malnutrition with long-term mortality
have been conducted mainly on elderly participants and did not use the national
death registry to track mortality (Sullivan & Walls, 1998; Miller et al., 2002;
Cereda et al., 2008b).
The current study found that malnourished patients stayed in the hospital one
and half times longer than well-nourished patients (6.9 7.3 days vs. 4.6 5.6
days, p < 0.001), which is consistent with other studies reporting longer LOS in
malnourished patients (Middleton et al., 2001; Correia & Waitzberg, 2003).
Malnourished patients cost an average of three times more than well-nourished
patients. This effect of malnutrition on hospitalisation cost was statistically
significant when matched for DRG, although this was not sustained when the
results were further adjusted for ethnicity, age and gender. These findings echo
previous studies which have linked malnutrition with higher hospitalisation costs
(Chima et al., 1997). These increased costs are indirectly attributed to longer
hospital stay (Correia & Waitzberg, 2003; Planas et al., 2004), increased use of
hospital resources (Correia & Waitzberg, 2003), higher rate of re-admission
(Planas et al., 2004), increased rates of infection (Rypkema et al., 2004) and
poor wound healing (Banks et al., 2010b).
Malnourished patients were found to have double the risk of readmission within
15 days in comparison to well-nourished patients (RR = 1.9, p < 0.001, 95% CI
= 1.1-3.2). However at 90 days and six months, even though there was
statistical significance when readmissions were controlled for age, gender and
ethnicity, the effect disappeared after matching for DRG. Planas et al (2004)
showed that malnourished patients were one and a half times more likely to be
readmitted within 6 months of discharge from the hospital but the results were
not controlled for any confounders (Planas et al., 2004).
202
Only a very small fraction of patients identified as malnourished in this study had
been classified under the DRG co-morbidity as malnourished (3 out of 235
patients). It is evident that the majority of malnourished patients in National
University Hospital were either not recognised, or not accurately documented
and coded for DRG. This scenario was also observed in Australia and Spain
whereby less than 2% of inpatients were coded as malnourished (Marco et al.,
2011; Rowell & Jackson, 2011), in contrast to many studies reporting the
prevalence of malnutrition in hospitals to be more than 30% (Norman et al.,
2008b). Accurate diagnosis and coding for malnutrition could result in a DRG
with a higher weighting, which would more accurately reflect the resource
requirement for these patients. In some countries this would increase the
amount of financial reimbursement received by the hospital (Delhey et al., 1989;
Funk & Ayton, 1995; Amaral et al., 2007). More importantly, the accurate
diagnosis of malnutrition is crucial so that nutrition support can be extended to
these patients to help improve patient outcome.
Impact on Practice
This study has provided the first published evidence that malnutrition is
prevalent in a Singapore hospital and independently leads to poor outcomes
regardless of the underlying disease and its complexities.
Strengths
The study utilised a large sequentially recruited sample representative of

patients admitted to a major Singaporean tertiary hospital. Prospective tracking
of hospitalisation outcomes and 3-year mortality based on national death
registry record are further strengths of this study. No prior studies have
controlled for the confounding effect of disease and its complexities on
malnutrition outcomes and only one study has used national death register data
to determine the 1-year mortality outcomes of patients (Middleton et al., 2001).
203
Limitations
This study was not able to monitor patients readmitted to other hospitals, and
thus readmission rates might be underestimated. However the likelihood of
readmission to other hospitals is low in our local experience, as most patients
prefer to return to the hospital closest to their homes, where their medical
records are held and where they attend post-discharge outpatient clinics. It
addition, it is the national policy in Singapore for public ambulance to take
emergency patients to the hospital nearest to their home.
Another limitation of this study is a lack of data on the number of study patients
referred for treatment of malnutrition or the outcome of that treatment. At the
time of this study (2006), access to inpatient dietetic services for assessment
and treatment of malnutrition was solely through medical referral by hospital
policy, and therefore it is likely that a substantial proportion of malnourished
patients were untreated (Bavelaar et al., 2008; Lamb et al., 2009). It is possible
that in a proportion of patients, nutritional status improved during or subsequent
to their admission. However, this proportion would likely be small and would
tend to attenuate associations between nutritional status at admission and
clinical outcomes.
8.1.2 Identifying patients at nutrition risk
This section discusses the two studies on the validation of 3-MinNS.
This first study developed and validated a new nutrition screening tool (3-
MinNS) for hospitalised adult patients in Singapore. Statistical analysis using the
ROC curve found that the most desirable among the five parameters tested in
terms of sensitivity and specificity was the combination of weight loss, nutritional
intake and muscle wastage. It was found that 3-MinNS was both sensitive (86%)
and specific (83%) in the determination of nutritional risk in newly admitted
patients. The 3-MinNS had good correlation with SGA, with the added
204
advantage of summative scoring and cutoff points useful to define a protocol for
subsequent action. It was able to differentiate patients at risk of moderate
malnutrition and severe malnutrition for prioritization and management
purposes. 3-Minute Nutrition Screening is non-invasive and does not require
expensive equipment or blood tests as is the case for a number of commonly
used nutrition screening tools such as the Nutrition Risk Index (NRI), Prognostic
Nutritional Index (PNI) and Maastrich Nutrition Index (MNI) (Schneider &
Hebuterne, 2000; Corish et al., 2004). Subsequent to the publication of 3-
MinNS, a validation study on NRS-2002 and MUST was published which
showed good sensitivity and specificity when compared to SGA (NRS-2002:
74% sensitive, 87% specific, MUST: 72% sensitive, 90% specific) (Velasco et
al., 2011). However the assessors were not blinded to the screening and
assessment results. Hence, observer bias might have occurred.
The second study confirms the validity of 3-MinNS when implemented by

nurses, the intended users of the tool, on a new cohort of hospitalised patients.
The sensitivity (89%) and specificity (88%) of 3-MinNS in this study were slightly
better than that of the earlier validation study, where 3-MinNS was implemented
by dietitians blinded to the screening results of another dietitian. The correlation
between 3-MinNS administered by nursing staff and SGA implemented by the
dietitian was good (r=0.78, p<0.001). Although in practice nutrition screening is
commonly undertaken by nursing staff, few studies have used nurses to validate
nutrition screening tools. Most studies have used researchers, dietitians or
dietetic staff to implement nutrition screening (Ferguson et al., 1999a; Kyle et
al., 2006; Kim et al., 2011; Velasco et al., 2011). Recent validation studies
using nursing staff to administer screening and a dietitian to complete nutrition
assessment in a blinded manner were conducted with the British Nutrition
Screening Tool (NST) (Mirmiran et al., 2011), MST and MUST (Lawson et al.,
2012). These studies showed that when the sensitivity was good, the specificity
was compromised and vice-versa (NST: 62% sensitive, 87% specific; MST: 45%
sensitive, 86% specific; MUST: 54% sensitive, 78% specific). The 3-MinNS
however was able to achieve good sensitivity and specificity at the same time
205
(89% and 88%) despite using nurses to conduct the screening and blinding to
the nutrition assessment results of the dietitian (Lim et al., 2013a).
In addition, this second study provided new data on the reliability of 3-MinNS,
which was not addressed in the first study. It shows good inter-rater reliability of
3-MinNS between nurse (agreement = 78.3%, k = 0.58, p<0.001)s. Several
inter-rater reliability tests have been performed on nutrition screening tools,
however many of these compared the results of dietitian with dietitian, or
dietitian with nursing staff (Reilly et al., 1995; Ferguson et al., 1999a; Burden et
al., 2001). These studies generally show good agreement, although MST and
Nutrition Risk Screening (NRS) have fared better than Nursing Nutritional
Assessment (NNA) (Reilly et al., 1995; Ferguson et al., 1999a; Isenring et al.,
2006). Only two studies examining inter-rater reliability between nurses have
been identified (McCall & Cotton, 2001; Mirmiran et al., 2011). In the Mirmiran et
al. (2011) study, a good level of agreement was found between nursing staff
administering the British Nutrition Screening Tool (NST) (Kappa = 0.71)
(Mirmiran et al., 2011). In contrast, McCall and Cotton (2001) found consistent
disagreement between nursing staff administering NNA, which they attributed to
subjective interpretation of questions within the tool (McCall & Cotton, 2001). In
this current study, each nurse was well trained in administering 3-MinNS, and
used the tool frequently in their daily practice. The scoring method and
objectivity of the questions within the 3-MinNS tool provided standardised
response options for the nurses. As 3-MinNS included assessment of muscles
at the temporalis and clavicular areas, nurses were trained using pictures of
different scorings of muscle wastage at these areas.
As 3-MinNS is simple and rapid to administer to acute hospital patients, and

enables healthcare workers to carry out nutrition screening accurately and
promptly, it is currently considered the best available nutrition screening tool to
identify patients at nutrition risk on admission to a hospital in Singapore.
206
Impact on Practice
More than 7000 nurses and dietitians have been trained on 3-MinNS in
Singapore and Malaysia. Through multiple platforms such as publications and
presentations at local and international conferences, 3-MinNS has been adopted
and implemented in the following organisations:
Singapore
National University Hospital
Mount Alvernia Hospital
Farrer Park Hospital
United Medicare Nursing Home
Man Fut Thong Nursing Home
Sree Narayanan Nursing Home
Khoo Teck Phuat Hospital
Alexandra Hospital
Gleneagles Hospital
Raffles Hospital
Malaysia
University Hospital Kuala Lumpur
Strengths
The 3-Minute Nutrition Screening tool was the first nutrition screening tool
developed in an Asian context; and specifically for Singapore with its particular
racial mix. The scoring system based on severity of nutritional indicators within
3-MinNS minimises ambiguity and the cutoff score for action guides the
healthcare professionals in their next course of action, i.e. whether to refer the
patient to the dietitian. The study was also the first validation study of a nutrition
screening tool in Singapore and Asia, which provides data on the sensitivity,
specificity, PPV and NPV of the tool. Further merits of this study are the large
sample size (n=818) consisting of a broad range of patients, recruitment of
207
subjects from a randomly determined sequential sample, and the blinded

reference method assessor. Prior to our study, most validation studies on
nutrition screening tools have not been conducted with blinding of the
assessors.
Both the validation studies on 3-MinNS used SGA as a reference standard.

Subjective Global Assessment is a well validated nutrition assessment tool and
prognostic indicator (Barbosa-Silva & Barros, 2006). For this reason, it is widely
used as a reference method for validating screening and assessment tools (Kyle
et al., 2006; Kim et al., 2011).
Limitations
In the first validation study, the study sample was representative of the hospitals
admission profile for gender and race but not for age. Although the study sample
was slightly older than the hospital population, this difference is unlikely to be
clinically significant or impact on results. The study protocol required that the
nutrition screening be completed within 24 hours and the reference standard
SGA within 48 hours, which may result in respondent bias as a patient may
respond differently or unable to answer based on their level of alertness and
medical condition at the time of nutrition screening and assessment. However,
the effect of this potential limitation was expected to be minimal as SGA was
completed within 24 hours for 90% of the study participants and changes in
nutritional status rarely occur over this short period of time.
In the second validation study, the study sample may not be representative of
the general hospital population as the participants were recruited from the
oncology and surgical wards. Oncology and surgical wards were chosen in
order to obtain a sufficient number of malnourished patients needed to
determine the sensitivity of the tool and both these specialties, especially
oncology, are known to have high prevalence of malnutrition. This is
advantageous as the nurses in these wards were used to working with
malnourished patients and nutrition may have higher profile on these wards than
208
in the general wards. In addition, the first study on the validation of 3-MinNS and
as presented in Chapter 5 had addressed the multidisciplinary setting (Lim et al.,
2009).
8.1.3 Compliance and referral process of nutrition screening
Even with the use of a valid and reliable nutrition screening tool, inaccurate
screening results may occur if the tool has missing data or is completed
erroneously. Studies have reported screening incompletion and error rates of
28-97% (Wong & Gandy, 2008; Neelemaat et al., 2011b). This may result in the
under-recognition and subsequent under-treatment of a malnourished or at-risk
patient. This problem is exacerbated if appropriate patients are not referred to a
nutrition-trained professional. A study by Kondrup et al (2002) showed that 53%
of patients identified as at nutritional risk did not have a nutrition plan
implemented (Kondrup et al., 2002).
In NUH, nutrition screening error rates were 33% and 31% in 2008 and 2009
respectively, with 5% and 8% blank or missing forms. From the study, we found
that despite the completion of nutrition screening within 24 hours of admission,
the entire process from nutrition screening to intervention by a dietitian took up
to 8.6 days, of which 7.5 days comprised of multiple activities (value added and
non-value added) preceding a dietetics referral. The mean time lag between
screening and referral was 4.3 1.8 days. Further drilling down of the process
identified the reasons for the long turnaround to be: (1) only doctors were
allowed to make dietetics referral for patients at nutritional risk, which resulted in
(2) the nurses having to constantly remind the doctors to make a dietetics
referral, which consumed a considerable amount of time.
Following root cause analysis, a list of quality improvement activities was

implemented. They included 1) establishing a nutrition screening protocol in the
hospital system, 2) nutrition screening training incorporated as part of the
compulsory nurses orientation programme, 3) nurse empowerment for online
209
dietetics referral of at risk cases, 4) closed-loop feedback system and 5)

removing a component of the nutrition screening that caused the most error
without compromising sensitivity and specificity. After the implementation of the
quality improvement action plan, nutrition screening error rates were
significantly reduced from 33% to 5% and blank or missing forms from 8% to
1%. The referral rate for patients at risk of malnutrition also improved from 10%
drop-referral to 3%. With the direct online referral system from the nurses, there
was a 92% reduction in turnaround time from nutrition screening to referral from
4.3 1.8 days to 0.3 0.4 days (p < 0.001). This study provided further
evidence on the capacity of the 3-MinNS to be completed accurately by nurses
so that newly admitted patients at nutritional risk can be provided with the
appropriate intervention as soon as possible.
Impact on Practice
This quality improvement study has significantly changed practice in a large

tertiary hospital which employs more than 3000 inpatient nurses, and has
resulted in excellent compliance to nutrition screening and referral of at risk
patients to the dietitians. The practical and effective action plan implemented
provides a positive case study for other organisations who are seeking ways to
improve on their nutrition screening and compliance rates.
Strengths
This study has several strengths, first of which is the large sample size of the
audits and consecutive sampling method. It is also amongst the first to report
the effectiveness of various quality improvement initiatives in increasing
compliance with nutrition screening, including the referral of care for appropriate
patients. The 93% compliance rate and 99% completion rate achieved in this
study are excellent when compared to other similar studies. Previous nutrition
screening audit studies have used only training as an intervention to improve the
compliance rate with limited success (Raja et al., 2008; Wong & Gandy, 2008).
The current study describes evidenced-based quality improvement measures
210
which were implemented successfully hospital-wide to improve the management

of malnourished or at risk patients.
Limitations
Given that this study is a quality improvement project, it did not include a control
group. Hence it could not be determined if the action plan was solely
responsible for the improvement in the compliance and referral rates. In
addition, the more efficient referral process has inevitably increased the
workload of dietetics, although the clinical and cost benefit to patients outweighs
the cost arising from additional resources required.
8.1.4 Follow-up of malnourished patients post discharge
In 2008, only 15% of malnourished patients discharged from NUH returned for
outpatient dietetic follow-up within four months of discharge from index
admission, and only 2% attended more than one dietetics outpatient clinic
appointment. This poor rate of follow-up is consistent with data from other
studies, which have shown that 54 to 58% of patients fail to attend scheduled
outpatient appointments (Gallagher, 1984; Finucane et al., 2007). In one study,
of the 54% of malnourished patients seen by a dietitian during admission only
23% were followed up after discharge (van Bokhorst-de van der Schueren et al.,
2005).
In response to the 2008 data, an Ambulatory Nutrition Support (ANS) service

was implemented in 2010, consisting of telephone calls, outpatient
appointments and home visits. If a patient failed to attend either of the
scheduled outpatient appointments they would receive a telephone call from the
dietetic assistant to review self-reported weight status and intake. Patients with
suboptimal intake or weight loss were visited at home by the study dietitian in
lieu of the missed scheduled outpatient appointment. This unique model of care
was able to achieve 100% follow-up of malnourished inpatients. There were
211
significant improvements in mean weight, triceps skinfold thickness and

handgrip strength for patients receiving ANS, with three out of four improving
their SGA score. In addition, two-thirds of patients had improved quality of life
(QOL). This is consistent with previous studies reporting an improvement in
QOL following nutritional intervention in malnourished patients (Ravasco et al.,
2005b; Rufenacht et al., 2010).
There is evidence that intensive dietetic monitoring and follow-up results in

higher nutrient intake, and this is a plausible reason for the improvements in
outcomes seen in this study (Kwon et al., 2004). These results are noteworthy,
as improvement in nutritional status has been shown to reduce readmissions,
rate of complications and mortality, which may result in long term cost-savings
for the individual, health-care institution and government (Stratton et al., 2005;
Koretz et al., 2007; Norman et al., 2008a; Gupta et al., 2010; Somanchi et al.,
2011; Starke et al., 2011).
The ANS service allowed for one of three modes of follow-up in the first four
months post-discharge, namely telephone calls, outpatient visits and home visits
for patients who did not attend outpatient appointments. Telephone calls made
up almost three quarters of overall contacts and were predominantly
administered by a trained dietetics assistant. They are relatively low cost and an
efficient form of patient follow-up which appear to generate positive nutritional
outcomes for patients as shown in this study. Telephone calls are less time-
intensive than outpatient reviews, and thus offer the benefit of reduced
manpower requirements and reduced costs for the healthcare provider. The
dietetic assistant was trained to ask a set of questions regarding appetite,
supplement usage (if prescribed), and to identify any new dietary issues or
questions. Detailed documentation on the advice given during telephone calls
ensured continuity of care when the dietitian saw the patient at the next follow-
up. The standard questions and advice administered by the dietetics assistant
via telephone calls are included in Appendix F. From the telephone calls, we
were able to review patients nutritional intake, reinforce the advice given during
the previous dietetic consultation and remind patients to attend the next
212
outpatient appointment with the dietitians. Patients who did not do well
nutritionally and missed their appointments were visited by the dietitian at their
homes. With this method, we were able to keep home visits (which is resource-
intensive) to patients who really needed it. To the best of our knowledge, there
have been no other studies that reserve home visits for this group of patients.
There are two studies published on the efficacy of combined outpatient visits
and telephone follow-up (Isenring et al., 2004; Neelemaat et al., 2011a) and one
study on home visits (Feldblum et al., 2010) to improve the nutritional outcomes
of malnourished patients discharged from hospital. These research were studies
conducted on geriatric population (Feldblum et al., 2010; Neelemaat et al.,
2011a) and oncology outpatients (Isenring et al., 2004). After the initiation of our
study, a pilot study on 12 patients was carried out to test the feasibility of
combined telephone follow-up and home visits on malnourished geriatric
patients ( 65 years old) discharged from the hospital (Mudge et al., 2012). As
the number of subjects was too small to provide any statistical significance, only
descriptive results were shared. The study provided proof of concept for post-
discharge nutritional management of malnourished patients using combined
telephone calls and home visits (Mudge et al., 2012).
Impact on Practice
This study provides promising initial evidence that a multi-modal ambulatory

nutrition support program is an effective way to follow-up malnourished hospital
patients post discharge, at relatively low cost and healthcare burden. The
Ambulatory Nutrition Support services arising from this research programme has
since been implemented as part of NUH Dietetics protocol for post-discharge
malnourished patients.
Strengths
There are a number of strengths for this study. It is the first study of its kind to
explore home visits for malnourished patients who did not attend follow-up and
213
includes adult malnourished patients across the age and disease spectrums. It
is also the first study specific to the Singaporean population. Inter-rater
differences were not present in this study as one dietitian measured all
nutritional outcomes, and this dietitian was trained in the use of all measurement
tools. The study protocol required that the baseline measurements (body weight,
SGA, mid-arm anthropometry, handgrip strength and QOL) carried out were no
more than 4 days before patient discharge regardless of whether they had been
done earlier during the admission. This ensures the currency of the baseline
data as it has been widely reported that patients weight and nutrition status tend
to deteriorate during hospitalisation (McWhirter & Pennington, 1994;
Braunschweig et al., 2000; Norman et al., 2008b; Cansado et al., 2009).
Limitations
A major limitation of this study is the pre-post design and lack of a control group
due to the nature of the funding which was for quality improvement purposes.
Hence, we were not able to use the randomised controlled trial (RCT) design.
There were ethical concerns as to what interventions would the control group
receive if RCT was conducted. It is possible that contact with a health
professional may have been responsible for the observed improvements. The
different time period of each cohort may have resulted in comparisons that were
not equally matched. Although there was no statistically significant difference
between the demographics of each cohort, there may have been other
characteristics which differed, such as socioeconomic status, family situation or
motivation level. The cost to the patient associated with outpatient review, in
comparison to free-of-charge telephone calls and home visits, may have
negatively impacted outpatient attendance rates.
214
8.2 RECENT DEVELOPMENTS
This section highlights recent relevant publications which post-date the

published papers arising from this thesis.
Malnutrition and outcomes
Subsequent to the publication of our paper (Lim et al., 2012), a large multicentre
study (n=2982) conducted in the Australasian region has been published
(Agarwal et al., 2013). In this study, malnutrition was assessed using SGA on a
single day assigned as NutritionDay (Agarwal et al., 2013). Disease severity
was determined using Patient Clinical Complexity Level (PCCL) and controlled
for in the analyses. The PCCL is a complex scoring system based on DRG to
determine the cumulative effect of a patients complications and comorbidities
(Department of Health and Ageing: Australian Government.). Agarwal et al.
(2013) found that malnourished patients had almost twice the risk of in-hospital
mortality and significantly higher readmission rate (35%) within 90 days of index
hospitalisation compared to well-nourished patients (27%). The median LOS for
malnourished patients was also 1.5 times higher than well-nourished patients
(Agarwal et al., 2013) These findings confirm the detrimental effect of
malnutrition in hospitalised patients independent of the underlying disease and
its complexities. This paper showed that even with a different approach to
controlling for underlying disease and its complexities, there is clear evidence of
an independent impact of malnutrition on clinical outcomes. As such, these
results confirm and strengthen the work of this thesis.
Nutrition screening
Post development and validation of 3-MinNS, the A.S.P.E.N. and the Academy
of Nutrition and Dietetics published a consensus statement which indicates that
a screening tool should include the identification of 2 or more of the following 6
characteristics to be used to diagnose malnutrition: 1) insufficient energy intake,
215
2) weight loss, 3) loss of muscle mass, 4) loss of subcutaneous fat, 5) localized

or generalized fluid accumulation and 6) diminished functional status as
measured by handgrip strength. (White et al., 2012). Coincidentally, the 3-
MinNS contains 3 of the 6 characteristics recommended. As the 3-MinNS is a
nutrition screening tool and not a diagnostic tool, having 3 characteristics would
suffice to determine patients at risk of malnutrition. This consensus statement
therefore provides further support to the work done on 3-MinNS under this
research programme. As such, there is potential for 3-MinNS to be used in other
countries if validated in the population.
Two recent studies which compare various nutrition screening tools in older
medical inpatients highlight commonalities between the tools (Poulia et al.,
2012; Young et al., 2013). In the study by Young et al. (2013), the nutrition
screening tools tested were MST, MNA-SF, MUST, NRS 2002, SNAQ,
Simplified Nutritional Appetite Questionnaire and Rapid Screen, whereas Poulia
et al. (2012) tested on NRI, Geriatric Nutritional Risk Index (GNRI), MNA-SF,
MUST and NRS 2002. Although the studies were specific to elderly inpatients,
the results from these two studies show that in general, all screening tools (if
accurately and completely implemented) are able to perform their function well.
Therefore, it is recommended that healthcare professionals choose a validated
screening tool that best matches their patient population and which they find
easiest to implement in practice (Young et al., 2013).
A recent study carried out in Norway, showed that quality improvement

initiatives were able to improve the nutrition screening rates of patients (Tangvik
et al., 2012) . The proportion of patients screened by NRS 2002 increased
significantly from the first to the last survey, with a range of 54% to 77% (p =
0.012). However the proportion of patients at nutritional risk who received
nutritional treatment did not improve with implementation and only 5% of the
patients at nutritional risk were evaluated and followed up by a dietitian. The
implementation plans were limited to creating nutritional guidelines and a staff
network. The staff in the network were educated for 2 days in basic clinical
nutrition and were tasked to introduce the guidelines to their units. The results
216
showed that besides implementing a nutrition screening protocol and involving

all the stake-holders, an effective quality improvement programme has to be
multi-pronged and continuous to ensure sustainable outcomes.
Post-discharge care
For the continuity of care for malnourished patients post-discharge, a recent

study on 38 malnourished elderly patients ( 60 years) reported that only 45%
received nutrition intervention within 6 weeks post-discharge (Rasheed &
Woods, 2013a). Of the group receiving nutrition support, 24% of patients
improved in body weight and 59% lost weight despite receiving nutrition
intervention. This reinforced an urgent need for a programme similar to our
Ambulatory Nutrition Support (ANS) service. With ANS, we managed to achieve
100% (n = 163) follow-up of malnourished patients post hospital discharge and
of these, an encouraging 70% gained weight in addition to significant
improvement in nutrition status, functional status and quality of life (Lim et al.,
2013b).
General comments
The conceptual framework in this thesis aligned with a recent recommendation

of a novel care model by the interdisciplinary Alliance to Advance Patient
Nutrition comprising of the Academy of Nutrition and Dietetics, the Academy of
Medical-Surgical Nurses, The American Society for Parenteral and Enteral
Nutrition, The Society of Hospital Medicine and Abbott Nutrition, to better
address hospital malnutrition (Tappenden et al., 2013). It emphasised 6
principles: (1) create an institutional culture where all stakeholders value
nutrition, (2) redefine clinicians' roles to include nutrition care, (3) recognize and
diagnose all malnourished patients and those at risk, (4) rapidly implement
comprehensive nutrition interventions and continued monitoring, (5)
communicate nutrition care plans, and (6) develop a comprehensive discharge
nutrition care and education plan. Prior to the research programme, NUH
Dietetics Department had already put in place a system to communicate
217
nutrition care plans with other healthcare professionals; especially the doctors,
nurses and caregivers (item 5 of the recommendation). This research
programme has successfully addressed principles 1, 3, 4 and 6.
Item 2 in redefining clinicians roles in nutrition care is challenging due to

professional silos and existing busy clinical workloads. However the publications
arising from this research programme have created multiple platforms to share
the results with clinicians and in local conferences. In addition, media interests
from the national newspapers (see page 18) on this research programme have
helped to raise awareness among the public and healthcare professionals.
These led to better awareness among clinicians on the importance to manage
malnutrition, with a few collaborations already taking place in Singapore, such
as the setting up of a Nutrition Support Team comprising doctors, dietitians and
pharmacists, blanket referral to dietitians for patients in the Intensive Care Unit
and implementing nutrition screening in other institutions.
In summary, the recent papers are in agreement with the work and results of the
research conducted under this PhD programme. Taken together, this PhD
research and the work of others have further built up the scientific knowledge
base and added value to current evidence-based practice in the screening and
management of at risk and malnourished patients.
218
8.3 RECOMMENDATIONS
Arising from the results and outcomes of this research programme, these are
the recommendations proposed in the areas of clinical practice, education and
research.
8.3.1 Clinical practice
It is imperative to recognise the significant prevalence and consequences of

malnutrition in hospitalised patients.
Malnutrition must be identified and treated concurrently with management of
a patients medical condition.
All hospitalised patients should be screened for malnutrition using a valid,
reliable, efficient tool relevant to the population on which it is used. The 3-
MinNS screening tool is recommended for use on Singaporean adult
inpatients.
Validity and reliability studies of a screening tool should ideally be carried out
with the intended assessors (usually nurses) in a blinded manner.
Compliance with nutrition screening, including completion rate, accuracy
(whether the tool is being used correctly) and referral rate for at risk patients
should be evaluated annually, at least.
If the audit results are undesirable, it is recommended to carry out quality
improvement activities to find out the root cause for the non-compliance to
nutrition screening and work collaboratively towards overcoming the
problems identified.
It is possible to achieve nutrition screening compliance rate of above 90%
with quality improvement initiatives effectively implemented. More
importantly, the positive results should be sustained through organisation
and system support.
219
Any patient identified as malnourished or at risk of malnutrition on screening

should be referred to the dietitians for nutrition assessment and nutrition
intervention as required. Creating an online referral system between the
professionals responsible for screening (nurses) and those responsible for
assessment (dietitians) could streamline this process.
Close collaboration between Dietetics, Nursing and Clinical Departments is
important to ensure continuous improvement and that the process of
screening, referral, intervention and monitoring are streamlined and adhered
to for the benefit of patients.
Post-discharge follow-up of malnourished patients is critical for monitoring
the effectiveness of inpatient nutrition intervention due to the short length of
hospital stay. Hence, community-based intervention and follow-up is a
feasible option for treating and preventing malnutrition. A model of follow-up
that includes outpatient review, telephone calls and home visits for patients
who do not attend outpatient appointments is effective in following up these
8.3.2 Education
Professional development education and/or orientation for healthcare

workers should include increasing the awareness of the prevalence and
impact of malnutrition.
Training in the implementation of nutrition screening should be incorporated
into orientation programmes for nurses and dietitians to ensure accurate
administration of screening and appropriate referral of patients. Periodic
refresher courses should be offered to existing staff to maintain competency.
The ongoing quality initiatives and results of the compliance to nutrition
screening should be incorporated into training programmes for nurses and
dietitians.
220
Dietitians on ward duties should capitalise on educational opportunities by

providing immediate feedback to nursing staff who have incorrectly
completed nutrition screening.
Malnourished patients and families or their caregivers need education to
understand their condition and be provided with resources and advice on
how to overcome malnutrition. The education on nutrition has to be targeted
and individualised.
Dietitians play an important role in empowering patients and caregivers to
manage malnutrition, and to rally family support for patients so as to
encourage them to be compliant with nutritional advice and treatment.
Close follow-up post hospitalisation is important to reinforce the education
that has been given to the patients, adjust treatment and to clarify any
misconception.
8.3.3 Research
Randomised controlled trials could be carried out to compare the

effectiveness of Ambulatory Nutrition Support services versus conventional
follow-up of malnourished patients via outpatient visits. These should include
robust cost analysis as well as cost feasibility if incorporate into practice.
Research investigating the prevention and management of malnutrition in the
outpatient or community settings in Singapore may be warranted as a first
step in reducing the prevalence of malnutrition in newly admitted hospital
patients. This may include nutrition screening at the community level and
treating those who are at risk or already malnourished before they
deteriorate further and require admission to the hospital.
There could be further study on the feasibility of electronic medical records to
facilitate data entry for nutrition screening and auto-referral of at-risk cases
to the dietitians.
221
It may also be worthwhile to study the feasibility of using electronic medical

records to overcome incompletion of screening by making it compulsory to
enter the nutrition screening field before allowing the user to continue.
As 3-MinNS has spread to nursing homes, it is recommended that a
validation study be carried out in this setting.
Research could be carried out to determine if the 3-Minute Nutrition
Screening Tool can be used in the community to identify malnourished or at
risk individuals in this setting. This may facilitate early intervention, prevent
the individual from further nutrition deterioration and avert hospitalization.
A randomised control trial could be conducted to determine if nutritional
intervention in the community can help prevent malnutrition in the first place,
leading to avoidance or reduction in hospitalisation.
222
8.4 CONCLUSION AND CONTRIBUTION TO KNOWLEDGE
This research programme has addressed a wide range of gaps in current

literature. It has made a significant contribution to the body of knowledge,
especially in Singapore, in the areas of malnutrition and its impact on outcomes,
nutrition screening, and improving dietetics support services for malnourished
patients. It has addressed the research questions that were set forth in the
beginning.
Malnutrition is a significant issue in Singapore, with almost one third of adult

patients malnourished on admission to hospital (RQ1). This has a significant
impact on clinical outcomes, resulting in longer length of stay, higher rates of
readmission and mortality, and greater healthcare costs (RQ1). The first critical
step in effective management of malnutrition is the use of nutrition screening to
accurately identify patients who are malnourished or at risk of malnutrition. The
3-MinNS is a quick, efficient and valid tool for nutrition screening of adult
inpatients in Singapore. It is able to identify patients requiring further nutrition
assessment and to differentiate between moderate and severe malnutrition
(RQ2). Most importantly, it is valid and reliable when administered by nursing
staff, the intended users of the tool (RQ3). Audits conducted post hospital-wide
implementation of 3-MinNS revealed fairly high rates of error and incompletion,
and failure to refer some cases for nutrition assessment (RQ4). Quality
improvement initiatives can successfully reduce error and incompletion rates,
and ensure appropriate referral of patients at nutritional risk (RQ4). Effective and
accurate nutrition screening and referral of appropriate patients are critical first
steps in the management of malnutrition, but given the short duration of most
inpatient admissions, post-discharge management strategies are critical to
holistically manage these patients. Many patients became lost to follow-up in
the traditional outpatient dietetic review model (RQ5). Conversely, an
Ambulatory Nutrition Support service is able to effectively follow-up patients
223
post-discharge, and results in improvements in nutritional and quality of life

outcomes (RQ5).
This research programme is amongst the first to examine the impact of

malnutrition on length of hospital stay, readmission, hospitalisation cost and
mortality in a large sample representative of patients admitted to a major
Singaporean tertiary hospital. It has provided clear evidence that the adverse
outcomes of malnutrition are not just a consequence of the disease process,
and lead to substantial increases in length of hospital stay, readmission rate,
mortality and hospitalisation cost when compared with well-nourished patients of
similar diagnoses and complexities. The research programme led to the
development and validation of a new nutrition screening tool (3-MinNS) and
confirmed that the 3-MinNS is a valid and reliable nutrition screening tool to be
used in Singaporean acute hospitals. Quality improvement methods employed
proved successful in improving the compliance of nurses to 3-MinNS and
ensuring referral of malnourished or at risk patients to dietitians. Finally, this
research programme has provided an evidence-based and effective method for
following up malnourished patients post-discharge, which resulted in improved
nutritional status of these patients.
In conclusion, the findings from the research programme have contributed to

positive changes in the nutrition screening process and the perception and
management of malnutrition in Singapore. It has contributed important new
knowledge and has demonstrated that it is possible to change practice in a large
and complex organisation employing thousands of staff, with many stakeholders
involved. Furthermore, this research programme has successfully delivered a
comprehensive model for managing hospital malnutrition, from screening on
admission and referral for assessment, to intervention and post-discharge
follow-up. Through these initiatives, we hope that many patients with or at risk of
malnutrition will receive the quality of care they need and deserve, leading to
improved outcomes and better quality of life.
224
REFERENCES
A.S.P.E.N. Board of Directors and the Clinical Guidelines Taskforce. (2002) Guidelines for the use of
parenteral and enteral nutrition in adult and pediatric patients JPEN J Parenter Enteral Nutr, 26,
1SA-138SA.
Agarwal, E., Ferguson, M., Banks, M., Batterham, M., Bauer, J., Capra, S. & Isenring, E. (2013)
Malnutrition and poor food intake are associated with prolonged hospital stay, frequent
readmissions, and greater in-hospital mortality: Results from the Nutrition Care Day Survey
2010. Clin Nutr, 32, 737-745.
Aliabadi, M., Kimiagar, M., Ghayour-Mobarhan, M., Shakeri, M.T., Nematy, M., Ilaty, A.A., Moosavi,
A.R. & Lanham-New, S. (2008) Prevalence of malnutrition in free living elderly people in Iran: a
cross-sectional study. Asia Pac J Clin Nutr, 17, 285-289.
Amaral, T.F., Matos, L.C., Tavares, M.M., Subtil, A., Martins, R., Nazare, M. & Sousa Pereira, N. (2007)
The economic impact of disease-related malnutrition at hospital admission. Clin Nutr, 26, 778-
784.
American Dietetic Association. (1994) Identifying patients at risk: ADA's definitions for nutrition
screening and nutrition assessment. Council on Practice (COP) Quality Management Committee.
J Am Diet Assoc, 94, 838-839.
American Society for Parenteral and Enteral Nutrition. (2012) Definition of terms, style, and conventions
used in A.S.P.E.N. Board of Directors-approved documents. Accessed 3 May 2013, Available
from: http://www.nutritioncare.org/Library.aspx.
Anthony, P.S. (2008) Nutrition screening tools for hospitalized patients. Nutr Clin Pract, 23, 373-382.
Baccaro, F., Moreno, J.B., Borlenghi, C., Aquino, L., Armesto, G., Plaza, G. & Zapata, S. (2007)
Subjective global assessment in the clinical setting. JPEN J Parenter Enteral Nutr, 31, 406-409.
Bailey, R. (2006) Implementing nutrition screening. Nurs Manag (Harrow), 13, 20-24. .
Baker, J.P., Detsky, A.S., Wesson, D.E., Wolman, S.L., Stewart, S., Whitewell, J., Langer, B. &
Jeejeebhoy, K.N. (1982a) Nutritional assessment: a comparison of clinical judgement and
objective measurements. N Engl J Med, 306, 969-972.
Baker, J.P., Detsky, A.S., Whitwell, J., Langer, B. & Jeejeebhoy, K.N. (1982b) A comparison of the
predictive value of nutritional assessment techniques. Hum Nutr Clin Nutr, 36, 233-241.
Baldwin, C. & Weekes, C.E. (2008) Dietary advice for illness-related malnutrition in adults. Cochrane
Database Syst Rev, CD002008.
Baldwin, C. & Weekes, C.E. (2012) Dietary counselling with or without oral nutritional supplements in
the management of malnourished patients: a systematic review and meta-analysis of randomised
controlled trials. J Hum Nutr Diet, 25, 411-426.
Baldwin, L.M., Klabunde, C.N., Green, P., Barlow, W. & Wright, G. (2006) In search of the perfect
comorbidity measure for use with administrative claims data: does it exist? Med Care, 44, 745-
753.
Banks, M., Ash, S., Bauer, J. & Gaskill, D. (2007) Prevalence of malnutrition in adults in Queensland
public hospitals and residential aged care facilities. Nutr Diet, 64, 172178.
Banks, M., Bauer, J., Graves, N. & Ash, S. (2010a) Malnutrition and pressure ulcer risk in adults in
Australian health care facilities. Nutrition, 26, 896-901.
225
Banks, M.D., Graves, N., Bauer, J.D. & Ash, S. (2010b) The costs arising from pressure ulcers
attributable to malnutrition. Clin Nutr, 29, 180-186.
Bannerman, E. & Ghosh, S. (2000) The prognostic value of the corrected arm-muscle area. Clin Nutr, 19,
70-71.
Barbosa-Silva, M.C. & Barros, A.J. (2006) Indications and limitations of the use of subjective global
assessment in clinical practice: an update. Curr Opin Clin Nutr Metab Care, 9, 263-269.
Baroni, M. & Sondel, S. (1995) A collaborative model for identifying feeding and nutrition needs in early
intervention Infant Young Child, 8, 15-25.
Barreto Penie, J. (2005) State of malnutrition in Cuban hospitals. Nutrition, 21, 487-497.
Bartholomew, C.M., Burton, S. & Davidson, L.A. (2003) Introduction of a community nutrition risk
assessment tool. Br J Nurs, 12, 351-358.
Bauer, J. & Capra, S. (2003) Comparison of a malnutrition screening tool with subjective global
assessment in hospitalised patients with cancer - sensitivity and specificity. Asia Pac J Clin Nutr,
12, 257-260.
Bauer, J., Capra, S. & Ferguson, M. (2002) Use of the scored Patient-Generated Subjective Global
Assessment (PG-SGA) as a nutrition assessment tool in patients with cancer. Eur J Clin Nutr, 56,
779-785.
Bauer, J.M., Kaiser, M.J., Anthony, P., Guigoz, Y. & Sieber, C.C. (2008) The Mini Nutritional
Assessment--its history, today's practice, and future perspectives. Nutr Clin Pract, 23, 388-396.
Bauer, J.M., Vogl, T., Wicklein, S., Trogner, J., Muhlberg, W. & Sieber, C.C. (2005) Comparison of the
Mini Nutritional Assessment, Subjective Global Assessment, and Nutritional Risk Screening
(NRS 2002) for nutritional screening and assessment in geriatric hospital patients. Z Gerontol
Geriatr, 38, 322-327.
Bavelaar, J.W., Otter, C.D., van Bodegraven, A.A., Thijs, A. & van Bokhorst-de van der Schueren, M.A.
(2008) Diagnosis and treatment of (disease-related) in-hospital malnutrition: the performance of
medical and nursing staff. Clin Nutr, 27, 431-438.
Baxter, Y.C., Dias, M.C., Maculevicius, J., Cecconello, I., Cotteleng, B. & Waitzberg, D.L. (2005)
Economic study in surgical patients of a new model of nutrition therapy integrating hospital and
home vs the conventional hospital model. JPEN J Parenter Enteral Nutr, 29, S96-105.
Beattie, A.H., Prach, A.T., Baxter, J.P. & Pennington, C.R. (2000) A randomised controlled trial
evaluating the use of enteral nutritional supplements postoperatively in malnourished surgical
patients. Gut, 46, 813-818.
Beberashvili, I., Azar, A., Sinuani, I., Yasur, H., Feldman, L., Averbukh, Z. & Weissgarten, J. (2010)
Objective Score of Nutrition on Dialysis (OSND) as an alternative for the malnutrition-
inflammation score in assessment of nutritional risk of haemodialysis patients. Nephrol Dial
Transplant, 25, 2662-2671.
Beghetto, M.G., Koglin, G. & de Mello, E.D. (2010) Influence of the assessment method on the
prevalence of hospital malnutrition: a comparison between two periods. Nutr Hosp, 25, 774-780.
Bell, J. (2007) Nutritional screening during hospital admission: 1. Nurs Times, 103, 3031. Accessed on
31 November 2011, Available from: www.nursingtimes.net.
Bishop, C.W., Bowen, P.E. & Ritchey, S.J. (1981) Norms for nutritional assessment of American adults
by upper arm anthropometry. Am J Clin Nutr, 34, 2530-2539.
226
Bouillanne, O., Morineau, G., Dupont, C., Coulombel, I., Vincent, J.P., Nicolis, I., Benazeth, S., Cynober,
L. & Aussel, C. (2005) Geriatric Nutritional Risk Index: a new index for evaluating at-risk
elderly medical patients. Am J Clin Nutr, 82, 777-783.
Bourdel-Marchasson, I., Joseph, P.A., Dehail, P., Biran, M., Faux, P., Rainfray, M., Emeriau, J.P.,
Canioni, P. & Thiaudiere, E. (2001) Functional and metabolic early changes in calf muscle
occurring during nutritional repletion in malnourished elderly patients. Am J Clin Nutr, 73, 832-
838.
Braunschweig, C., Gomez, S. & Sheean, P.M. (2000) Impact of declines in nutritional status on outcomes
in adult patients hospitalized for more than 7 days. J Am Diet Assoc, 100, 1316-1322.
Braunschweig, C.L., Gomez, S., Sheean, P., Tomey, K.M., Rimmer, J. & Heller, T. (2004) Nutritional
status and risk factors for chronic disease in urban-dwelling adults with Down syndrome. Am J
Ment Retard, 109, 186-193.
Brown, C.S. & Stegman, M.R. (1988) Nutritional assessment of surgical patients. QRB Qual Rev Bull, 14,
302-306.
Brugler, L., DiPrinzio, M.J. & Bernstein, L. (1999) The five-year evolution of a malnutrition treatment
program in a community hospital. Jt Comm J Qual Improv, 25, 191-206.
Bruun, L.I., Bosaeus, I., Bergstad, I. & Nygaard, K. (1999) Prevalence of malnutrition in surgical patients:
evaluation of nutritional support and documentation. Clin Nutr, 18, 141-147.
Bryan, F., Jones, J.M. & Russell, L. (1998) Reliability and validity of a nutrition screening tool to be used
with clients with learning difficulties. J Hum Nutr Diet, 11, 4150.
Burden, S.T., Bodey, S., Bradburn, Y.J., Murdoch, S., Thompson, A.L., Sim, J.M. & Sowerbutts, A.M.
(2001) Validation of a nutrition screening tool: testing the reliability and validity. J Hum Nutr
Diet, 14, 269-275.
Burton, L.A. & Sumukadas, D. (2010) Optimal management of sarcopenia. Clin Interv Aging, 5, 217-228.
Butterworth, C.E., Jr. (1994) The skeleton in the hospital closet. Nutrition, 10, 442.
Buzby, G.P., Mullen, J.L., Matthews, D.C., Hobbs, C.L. & Rosato, E.F. (1980) Prognostic nutritional
index in gastrointestinal surgery. Am J Surg, 139, 160-167.
Campbell, I.T. (1999) Limitations of nutrient intake. The effect of stressors: trauma, sepsis and multiple
organ failure. Eur J Clin Nutr, 53 Suppl 1, S143-147.
Campbell, K.L., Ash, S., Bauer, J.D. & Davies, P.S. (2007) Evaluation of nutrition assessment tools
compared with body cell mass for the assessment of malnutrition in chronic kidney disease. J
Ren Nutr, 17, 189-195.
Campbell, K.L., Ash, S., Davies, P.S. & Bauer, J.D. (2008) Randomized controlled trial of nutritional
counseling on body composition and dietary intake in severe CKD. Am J Kidney Dis, 51, 748-
758.
Campbell, M.K. & Kelsey, K.S. (1994) The PEACH survey: a nutrition screening tool for use in early
intervention programs. J Am Diet Assoc, 94, 1156-1158.
Campbell, S.E., Avenell, A. & Walker, A.E. (2002) Assessment of nutritional status in hospital in-
patients. QJM, 95, 83-87.
227
Cansado, P., Ravasco, P. & Camilo, M. (2009) A longitudinal study of hospital undernutrition in the
elderly: comparison of four validated methods. J Nutr Health Aging, 13, 159-164.
Capra, S. (2007) Nutrition assessment or nutrition screening--how much information is enough to make a
diagnosis of malnutrition in acute care? Nutrition, 23, 356-357.
Cereda, E., Limonta, D., Pusani, C. & Vanotti, A. (2007) Feasible use of estimated height for predicting
outcome by the geriatric nutritional risk index in long-term care resident elderly. Gerontology,
53, 184-186.
Cereda, E., Valzolgher, L. & Pedrolli, C. (2008a) Mini nutritional assessment is a good predictor of
functional status in institutionalised elderly at risk of malnutrition. Clin Nutr, 27, 700-705.
Cereda, E., Zagami, A., Vanotti, A., Piffer, S. & Pedrolli, C. (2008b) Geriatric Nutritional Risk Index and
overall-cause mortality prediction in institutionalised elderly: a 3-year survival analysis. Clin
Nutr, 27, 717-723.
Chan, M., Lim, Y.P., Ernest, A. & Tan, T.L. (2010) Nutritional assessment in an Asian nursing home and
its association with mortality. J Nutr Health Aging, 14, 23-28.
Charalambous, L. (1993) Ageing matters. A healthy approach. Nurs Times, 89, 59-60.
Charlson, M.E., Pompei, P., Ales, K.L. & MacKenzie, C.R. (1987) A new method of classifying
prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis, 40,
373-383.
Charlton, K., Nichols, C., Bowden, S., Milosavljevic, M., Lambert, K., Barone, L., Mason, M. &
Batterham, M. (2012) Poor nutritional status of older subacute patients predicts clinical outcomes
and mortality at 18 months of follow-up. Eur J Clin Nutr, 66, 1224-1228.
Charlton, K.E., Kolbe-Alexander, T.L. & Nel, J.H. (2005) Development of a novel nutrition screening tool
for use in elderly South Africans. Public Health Nutr, 8, 468-479.
Charlton, K.E., Kolbe-Alexander, T.L. & Nel, J.H. (2007) The MNA, but not the DETERMINE,
screening tool is a valid indicator of nutritional status in elderly Africans. Nutrition, 23, 533-542.
Charney, P. (2008) Nutrition screening vs nutrition assessment: how do they differ? Nutr Clin Pract, 23,
366-372.
Chima, C.S., Barco, K., Dewitt, M.L., Maeda, M., Teran, J.C. & Mullen, K.D. (1997) Relationship of
nutritional status to length of stay, hospital costs, and discharge status of patients hospitalized in
the medicine service. J Am Diet Assoc, 97, 975-978.
Chima, C.S., Dietz-Seher, C. & Kushner-Benson, S. (2008) Nutrition risk screening in acute care: a survey
of practice. Nutr Clin Pract, 23, 417-423.
Chumlea, W.C. (2006) Is the MNA valid in different populations and across practice settings? J Nutr
Health Aging, 10, 524-527.
Churchill, D.N., Taylor, D.W. & Keshaviah, P.R. (1996) Adequacy of dialysis and nutrition in continuous
peritoneal dialysis: association with clinical outcomes. Canada-USA (CANUSA) Peritoneal
Dialysis Study Group. J Am Soc Nephrol, 7, 198-207.
Cleary, P.D., Greenfield, S., Mulley, A.G., Pauker, S.G., Schroeder, S.A., Wexler, L. & McNeil, B.J.
(1991) Variations in length of stay and outcomes for six medical and surgical conditions in
Massachusetts and California. JAMA, 266, 73-79.
228
Collin, C., Wade, D.T., Davies, S. & Horne, V. (1988) The Barthel ADL Index: a reliability study. Int
Disabil Stud, 10, 61-63.
Commonwealth Department of Health and Family Services and 3M Health Information Systems (1996)
Australian National Diagnosis Related Groups (AN-DRGs) Version 3.1, Definitions Manual, Vol
1, 3M Australia Pty Limited and Commonwealth of Australia.
Compan, B., di Castri, A., Plaze, J.M. & Arnaud-Battandier, F. (1999) Epidemiological study of
malnutrition in elderly patients in acute, sub-acute and long-term care using the MNA. J Nutr
Health Aging, 3, 146-151.
Cooper, N. (1998) Audit in clinical practice: evaluating use of a nutrition screening tool for developed for
trauma nurses. J Hum Nutr Diet, 11, 403-410.
Corish, C.A., Flood, P. & Kennedy, N.P. (2004) Comparison of nutritional risk screening tools in patients
on admission to hospital. J Hum Nutr Diet, 17, 133-139.
Corish, C.A. & Kennedy, N.P. (2000) Protein-energy undernutrition in hospital in-patients. Br J Nutr, 83,
575-591.
Correia, M.I. & Campos, A.C. (2003) Prevalence of hospital malnutrition in Latin America: the
multicenter ELAN study. Nutrition, 19, 823-825.
Correia, M.I. & Waitzberg, D.L. (2003) The impact of malnutrition on morbidity, mortality, length of
hospital stay and costs evaluated through a multivariate model analysis. Clin Nutr, 22, 235-239.
Cotton, E., Zinober, B. & Jessop, J. (1996) A nutritional assessment tool for older patients. Prof Nurse, 11,
609-610, 612.
Covinsky, K.E., Covinsky, M.H., Palmer, R.M. & Sehgal, A.R. (2002) Serum albumin concentration and
clinical assessments of nutritional status in hospitalized older people: different sides of different
coins? J Am Geriatr Soc, 50, 631-637.
Covinsky, K.E., Martin, G.E., Beyth, R.J., Justice, A.C., Sehgal, A.R. & Landefeld, C.S. (1999) The
relationship between clinical assessments of nutritional status and adverse outcomes in older
hospitalized medical patients. J Am Geriatr Soc, 47, 532-538.
Davalos, A., Ricart, W., Gonzalez-Huix, F., Soler, S., Marrugat, J., Molins, A., Suner, R. & Genis, D.
(1996) Effect of malnutrition after acute stroke on clinical outcome. Stroke, 27, 1028-1032.
de Mutsert, R., Grootendorst, D.C., Indemans, F., Boeschoten, E.W., Krediet, R.T. & Dekker, F.W. (2009)
Association between serum albumin and mortality in dialysis patients is partly explained by
inflammation, and not by malnutrition. J Ren Nutr, 19, 127-135.
de Onis, M. & Habicht, J.P. (1996) Anthropometric reference data for international use: recommendations
from a World Health Organization Expert Committee. Am J Clin Nutr, 64, 650-658.
Delhey, D.M., Anderson, E.J. & Laramee, S.H. (1989) Implications of malnutrition and diagnosis-related
groups (DRGs). J Am Diet Assoc, 89, 1448-1451.
Department of Health and Ageing: Australian Government. Australian Casemix Glossary. AR-DRG
version 6.0. Accessed on 25 August 2013, Available from:
http://www.health.gov.au/internet/main/publishing.nsf/Content/health-casemix-glossary1.htm.
Desbrow, B., Bauer, J., Blum, C., Kandasamy, A., McDonald, A. & Montgomery, K. (2005) Assessment
of nutritional status in hemodialysis patients using patient-generated subjective global
assessment. J Ren Nutr, 15, 211-216.
229
Detsky, A.S., Baker, J.P., Mendelson, R.A., Wolman, S.L., Wesson, D.E. & Jeejeebhoy, K.N. (1984)
Evaluating the accuracy of nutritional assessment techniques applied to hospitalized patients:
methodology and comparisons. JPEN J Parenter Enteral Nutr, 8, 153-159.
Detsky, A.S., McLaughlin, J.R., Baker, J.P., Johnston, N., Whittaker, S., Mendelson, R.A. & Jeejeebhoy,
K.N. (1987) What is subjective global assessment of nutritional status? JPEN J Parenter Enteral
Nutr, 11, 8-13.
Deurenberg-Yap, M., Schmidt, G., van Staveren, W.A. & Deurenberg, P. (2000) The paradox of low body
mass index and high body fat percentage among Chinese, Malays and Indians in Singapore. Int J
Obes Relat Metab Disord, 24, 1011-1017.
Deurenberg, P. & Deurenberg-Yap, M. (2001) Differences in body-composition assumptions across ethnic

groups: practical consequences. Curr Opin Clin Nutr Metab Care, 4, 377-383.
Devoto, G., Gallo, F., Marchello, C., Racchi, O., Garbarini, R., Bonassi, S., Albalustri, G. & Haupt, E.
(2006) Prealbumin serum concentrations as a useful tool in the assessment of malnutrition in
hospitalized patients. Clin Chem, 52, 2281-2285.
Doweiko, J.P. & Nompleggi, D.J. (1991) The role of albumin in human physiology and pathophysiology,
Part III: Albumin and disease states. JPEN J Parenter Enteral Nutr, 15, 476-483.
Doyle, M.P., Barnes, E. & Moloney, M. (2000) The evaluation of an undernutrition risk score to be used
by nursing staff in a teaching hospital to identify surgical patients at risk of malnutrition on
admission: a pilot study. J Hum Nutr Diet, 13, 433-441.
Duerksen, D.R. (2002) Teaching medical students the subjective global assessment. Nutrition, 18, 313-
315.
Duerksen, D.R., Yeo, T.A., Siemens, J.L. & O'Connor, M.P. (2000) The validity and reproducibility of
clinical assessment of nutritional status in the elderly. Nutrition, 16, 740-744.
Dzieniszewski, J., Jarosz, M., Szczygiel, B., Dlugosz, J., Marlicz, K., Linke, K., Lachowicz, A., Ryzko-
Skiba, M. & Orzeszko, M. (2005) Nutritional status of patients hospitalised in Poland. Eur J Clin
Nutr, 59, 552-560.
Edington, J., Boorman, J., Durrant, E.R., Perkins, A., Giffin, C.V., James, R., Thomson, J.M., Oldroyd,
J.C., Smith, J.C., Torrance, A.D., Blackshaw, V., Green, S., Hill, C.J., Berry, C., McKenzie, C.,
Vicca, N., Ward, J.E. & Coles, S.J. (2000) Prevalence of malnutrition on admission to four
hospitals in England. The Malnutrition Prevalence Group. Clin Nutr, 19, 191-195.
Ek, A.C., Unosson, M., Larsson, J., Ganowiak, W. & Bjurulf, P. (1996) Interrater variability and validity
in subjective nutritional assessment of elderly patients. Scand J Caring Sci, 10, 163-168.
Elia, M. (2003) Screening for Malnutrition: A Multidisciplinary Responsibility. Development and Use of
the Malnutrition Universal Screening Tool (MUST) for Adults. Malnutrition Advisory Group
(MAG), A Standing Committee of BAPEN. Redditch, Worcs: BAPEN.
Elia, M. (2006) Nutrition and health economics. Nutrition, 22, 576-578.
Elmore, M.F., Wagner, D.R., Knoll, D.M., Eizember, L., Oswalt, M.A., Glowinski, E.A. & Rapp, P.A.
(1994) Developing an effective adult nutrition screening tool for a community hospital. J Am
Diet Assoc, 94, 1113-1118.
Feldblum, I., German, L., Castel, H., Harman-Boehm, I., Bilenko, N., Eisinger, M., Fraser, D. & Shahar,
D.R. (2007) Characteristics of undernourished older medical patients and the identification of
predictors for undernutrition status. Nutr J, 6, 37.
230
Feldblum, I., German, L., Castel, H., Harman-Boehm, I. & Shahar, D.R. (2010) Individualized nutritional
intervention during and after hospitalization: the nutrition intervention study clinical trial. J Am
Geriatr Soc, 59, 10-17.
Ferguson, M., Banks, M., Bauer, J., Isenring, E., Vivanti, A. & Capra, S. (2010) Nutrition screening
practices in Australian healthcare facilities: A decade later. Nutr Diet, 67, 213218.
Ferguson, M., Capra, S., Bauer, J. & Banks, M. (1999a) Development of a valid and reliable malnutrition
screening tool for adult acute hospital patients. Nutrition, 15, 458-464.
Ferguson, M.L., Bauer, J., Gallagher, B., Capra, S., Christie, D.R. & Mason, B.R. (1999b) Validation of a
malnutrition screening tool for patients receiving radiotherapy. Australas Radiol, 43, 325-327.
Fiaccadori, E., Lombardi, M., Leonardi, S., Rotelli, C.F., Tortorella, G. & Borghetti, A. (1999) Prevalence
and clinical outcome associated with preexisting malnutrition in acute renal failure: a prospective
cohort study. J Am Soc Nephrol, 10, 581-593.
Finucane, F.M., Gaffney, L., Hatunic, M., Burns, N. & Nolan, J.J. (2007) Attendance at an Irish diabetes
dietetic outpatient clinic. Diabetes Res Clin Pract, 77, 335-336.
Fleiss, J.L., Levin, B. & Paik, M.C. (2003) Statistical methods for rates and proportions The measurement
of interrater agreement. John Wiley & Sons, New Jersey, pp. 598-627.
Flood, A., Chung, A., Parker, H., Kearns, V. & O'Sullivan, T.A. (2013) The use of hand grip strength as a
predictor of nutrition status in hospital patients. Clin Nutr, (In press).
Foucan, L., Hanley, J., Deloumeaux, J. & Suissa, S. (2002) Body mass index (BMI) and waist
circumference (WC) as screening tools for cardiovascular risk factors in Guadeloupean women. J
Clin Epidemiol, 55, 990-996.
Freijer, K., Tan, S.S., Koopmanschap, M.A., Meijers, J.M., Halfens, R.J. & Nuijten, M.J. (2013) The
economic costs of disease related malnutrition. Clin Nutr, 32, 136-141.
Funk, K.L. & Ayton, C.M. (1995) Improving malnutrition documentation enhances reimbursement. J Am
Diet Assoc, 95, 468-475.
Gallagher, C. (1984) Next please: a review of dietetic out-patient attendance. Hum Nutr Appl Nutr, 38,
181-186.
Gariballa, S. & Forster, S. (2007) Malnutrition is an independent predictor of 1-year mortality following
acute illness. Br J Nutr, 98, 332-336.
Gazzotti, C., Albert, A., Pepinster, A. & Petermans, J. (2000) Clinical usefulness of the mini nutritional
assessment (MNA) scale in geriatric medicine. J Nutr Health Aging, 4, 176-181.
Geiker, N.R., Horup Larsen, S.M., Stender, S. & Astrup, A. (2012) Poor performance of mandatory
nutritional screening of in-hospital patients. Clin Nutr, 31, 862-867.
Gilford, A.K. & Khun Khun, R. (1996) Development of nutritional risk screening in the community. Br J
Community Health Nursing, 1, 335339.
Gomes Beghetto, M., Koglin, G. & Daniel de Mello, E. (2011) Influence of the assessment method on the
prevalence of hospital malnutrition: a comparison between two periods. Nutr Hosp, 25, 774-780.
Gomez Candela, C., Olivar Roldan, J., Garcia, M., Marin, M., Madero, R., Perez-Portabella, C., Planas,
M., Mokoroa, A., Pereyra, F. & Martin Palmero, A. (2010) [Assessment of a malnutrition
screening tool in cancer patients]. Nutr Hosp, 25, 400-405.
231
Goudge, D.R., Williams, A. & Pinnington, L.L. (1998) Development, validity and reliability of the Derby
Nutritional Score. J Hum Nutr Diet, 11, 411-421.
Gout, B.S., Barker, L.A. & Crowe, T.C. (2009) Malnutrition identification, diagnosis and dietetic
referrals: Are we doing a good enough job? Nutr Diet, 66, 206211.
Gower, T. (2002) Computers in clinical practice: screening renal patients for abnormal biochemistry and
malnutrition. J Ren Nutr, 12, 107-112.
Green, S.M. & Watson, R. (2005) Nutritional screening and assessment tools for use by nurses: literature
review. J Adv Nurs, 50, 69-83.
Guigoz, Y. (2006) The Mini Nutritional Assessment (MNA) review of the literature--What does it tell us?
J Nutr Health Aging, 10, 466-485.
Guigoz, Y., Vellas, B. & Garry, P.J. (1994) Mini Nutritional Assessment: a practical assessment tool for
grading the nutritional state of elderly patients. . Facts Res Gerontol, 4 (suppl 2), 15-59.
Guigoz, Y., Vellas, B. & Garry, P.J. (1996) Assessing the nutritional status of the elderly: The Mini
Nutritional Assessment as part of the geriatric evaluation. Nutr Rev, 54, S59-65.
Guo, C.B., Ma, D.Q. & Zhang, K.H. (1994) Applicability of the general nutritional status score to patients
with oral and maxillofacial malignancies. Int J Oral Maxillofac Surg, 23, 167-169.
Gupta, D., Lammersfeld, C.A., Vashi, P.G., Burrows, J., Lis, C.G. & Grutsch, J.F. (2005) Prognostic
significance of Subjective Global Assessment (SGA) in advanced colorectal cancer. Eur J Clin
Nutr, 59, 35-40.
Gupta, D., Lis, C.G., Vashi, P.G. & Lammersfeld, C.A. (2010) Impact of improved nutritional status on
survival in ovarian cancer. Support Care Cancer, 18, 373-381.
Gur, A.S., Atahan, K., Aladag, I., Durak, E., Cokmez, A., Tarcan, E. & Tavusbay, C. (2009) The efficacy
of Nutrition Risk Screening-2002 (NRS-2002) to decide on the nutritional support in general
surgery patients. Bratisl Lek Listy, 110, 290-292.
Ha, L., Hauge, T., Spenning, A.B. & Iversen, P.O. (2010) Individual, nutritional support prevents
undernutrition, increases muscle strength and improves QoL among elderly at nutritional risk
hospitalized for acute stroke: a randomized, controlled trial. Clin Nutr, 29, 567-573.
Hamerman, D. (2002) Molecular-based therapeutic approaches in treatment of anorexia of aging and

cancer cachexia. J Gerontol A Biol Sci Med Sci, 57, M511-518.
Haupt, W., Holzheimer, R.G., Riese, J., Klein, P. & Hohenberger, W. (1999) Association of low
preoperative serum albumin concentrations and the acute phase response. Eur J Surg, 165, 307-
313.
Haverkort, E.B., Binnekade, J.M., de Haan, R.J. & van Bokhorst-de van der Schueren, M.A. (2012)
Handgrip strength by dynamometry does not identify malnutrition in individual preoperative
outpatients. Clin Nutr, 31, 647-651.
Henderson, C.J., Lovell, D.J. & Gregg, D.J. (1992) A nutritional screening test for use in children and
adolescents with juvenile rheumatoid arthritis. J Rheumatol, 19, 1276-1281.
Heyland, D.K., Dhaliwal, R., Jiang, X. & Day, A.G. (2011) Identifying critically ill patients who benefit
the most from nutrition therapy: the development and initial validation of a novel risk assessment
tool. Crit Care, 15, R268.
232
Heymsfield, S.B., McManus, C., Smith, J., Stevens, V. & Nixon, D.W. (1982) Anthropometric
measurement of muscle mass: revised equations for calculating bone-free arm muscle area. Am J
Clin Nutr, 36, 680-690.
Hickson, M. & Hill, M. (1997) Implementing a nutritional assessment tool in the community: a report
describing the process, audit and problems encountered. J Hum Nutr Diet, 10, 373377.
Hickson, M., Macqueen, C. & Frost, G. (2009) Evaluation of attendance and weight loss in an intensive
weight management clinic compared to standard dietetic care. J Hum Nutr Diet, 22, 72-76.
Hirsch, S., de Obaldia, N., Petermann, M., Rojo, P., Barrientos, C., Iturriaga, H. & Bunout, D. (1991)
Subjective global assessment of nutritional status: further validation. Nutrition, 7, 35-37.
Hoad, V., Somerford, P. & Katzenellenbogen, J. (2010) High body mass index overtakes tobacco as the
leading independent risk factor contributing to disease burden in Western Australia. Aust N Z J
Public Health, 34, 214-215.
Hoffer, L.J. (2001) Clinical nutrition: 1. Protein-energy malnutrition in the inpatient. CMAJ, 165, 1345-
1349.
Hudgens, J., Langkamp-Henken, B., Stechmiller, J.K., Herrlinger-Garcia, K.A. & Nieves, C., Jr. (2004)
Immune function is impaired with a mini nutritional assessment score indicative of malnutrition
in nursing home elders with pressure ulcers. JPEN J Parenter Enteral Nutr, 28, 416-422.
Hulst, J.M., Zwart, H., Hop, W.C. & Joosten, K.F. (2010) Dutch national survey to test the STRONGkids
nutritional risk screening tool in hospitalized children. Clin Nutr, 29, 106-111.
Hunt, D.R., Maslovitz, A., Rowlands, B.J. & Brooks, B. (1985a) A simple nutrition screening procedure
for hospital patients. J Am Diet Assoc, 85, 332-335.
Hunt, D.R., Rowlands, B.J. & Johnston, D. (1985b) Hand grip strength - a simple prognostic indicator in
surgical patients. JPEN J Parenter Enteral Nutr, 9, 701-704.
Iizaka, S., Okuwa, M., Sugama, J. & Sanada, H. (2010) The impact of malnutrition and nutrition-related
factors on the development and severity of pressure ulcers in older patients receiving home care.
Clin Nutr, 29, 47-53.
Incalzi, R.A., Capparella, O., Gemma, A., Landi, F., Bruno, E., Di Meo, F. & Carbonin, P. (1997) The
interaction between age and comorbidity contributes to predicting the mortality of geriatric
patients in the acute-care hospital. J Intern Med, 242, 291-298.
Incalzi, R.A., Gemma, A., Capparella, O., Cipriani, L., Landi, F. & Carbonin, P. (1996) Energy intake and
in-hospital starvation. A clinically relevant relationship. Arch Intern Med, 156, 425-429.
Ingenbleek, Y. & Carpentier, Y.A. (1985) A prognostic inflammatory and nutritional index scoring
critically ill patients. Int J Vitam Nutr Res, 55, 91-101.
Iseki, K., Ikemiya, Y., Kinjo, K., Inoue, T., Iseki, C. & Takishita, S. (2004) Body mass index and the risk
of development of end-stage renal disease in a screened cohort. Kidney Int, 65, 1870-1876.
Isenring, E., Bauer, J. & Capra, S. (2003a) The scored Patient-generated Subjective Global Assessment
(PG-SGA) and its association with quality of life in ambulatory patients receiving radiotherapy.
Eur J Clin Nutr, 57, 305-309.
Isenring, E., Capra, S., Bauer, J. & Davies, P.S. (2003b) The impact of nutrition support on body
composition in cancer outpatients receiving radiotherapy. Acta Diabetol, 40 Suppl 1, S162-164.
233
Isenring, E., Cross, G., Daniels, L., Kellett, E. & Koczwara, B. (2006) Validity of the malnutrition
screening tool as an effective predictor of nutritional risk in oncology outpatients receiving
chemotherapy. Support Care Cancer, 14, 1152-1156.
Isenring, E.A., Bauer, J.D., Banks, M. & Gaskill, D. (2009) The Malnutrition Screening Tool is a useful
tool for identifying malnutrition risk in residential aged care. J Hum Nutr Diet, 22, 545-550.
Isenring, E.A., Capra, S. & Bauer, J.D. (2004) Nutrition intervention is beneficial in oncology outpatients
receiving radiotherapy to the gastrointestinal or head and neck area. Br J Cancer, 91, 447-452.
Jeejeebhoy, K.N. (2000) Nutritional assessment. Nutrition, 16, 585-590.
Jeejeebhoy, K.N. (2012) Malnutrition, fatigue, frailty, vulnerability, sarcopenia and cachexia: overlap of
clinical features. Curr Opin Clin Nutr Metab Care, 15, 213-219.
Jensen, G.L., Bistrian, B., Roubenoff, R. & Heimburger, D.C. (2009) Malnutrition syndromes: a
conundrum vs continuum. JPEN J Parenter Enteral Nutr, 33, 710-716.
Jensen, G.L., Mirtallo, J., Compher, C., Dhaliwal, R., Forbes, A., Grijalba, R.F., Hardy, G., Kondrup, J.,
Labadarios, D., Nyulasi, I., Castillo Pineda, J.C. & Waitzberg, D. (2010) Adult starvation and
disease-related malnutrition: a proposal for etiology-based diagnosis in the clinical practice
setting from the international consensus guideline committee. JPEN J Parenter Enteral Nutr, 34,
156-159.
Johnstone, C., Farley, A. & Hendry, C. (2006a) Nurses' role in nutritional assessment and screening
(second of a two-part series). Nurs Times, 102, 28-29.
Johnstone, C., Farley, A. & Hendry, C. (2006b) Nurses' role in nutritional assessment and screening: Part
one of a two-part series. Nurs Times, 102, 28-29.
Joint Commission International (2008) Accreditation Standards for Hospital. 3rd Edition Assessment of
Patients Standard AOP.1.6. . Department of Publications of Joint Commission Resources,
Oakbrook, IL, pp. 77-78.
Jones, C.H., Wolfenden, R.C. & Wells, L.M. (2004) Is subjective global assessment a reliable measure of
nutritional status in hemodialysis? J Ren Nutr, 14, 26-30.
Jones, J.M. (2002) The methodology of nutritional screening and assessment tools. J Hum Nutr Diet, 15,
59-71.
Jones, J.M. (2004a) Reliability of nutritional screening and assessment tools. Nutrition, 20, 307-311.
Jones, J.M. (2004b) Validity of nutritional screening and assessment tools. Nutrition, 20, 312-317.
Joyce, T., Mayre-Chilton, K.M. & Tabi, S. (2011) Audit of the completion of the nutrition screening tool
on cardiac and cardiothoracic wards (Abstract). J Hum Nutr Diet, 24, 392-393.
Kagansky, N., Berner, Y., Koren-Morag, N., Perelman, L., Knobler, H. & Levy, S. (2005) Poor nutritional
habits are predictors of poor outcome in very old hospitalized patients. Am J Clin Nutr, 82, 784-
791.
Kaiser, M.J., Bauer, J.M., Ramsch, C., Uter, W., Guigoz, Y., Cederholm, T., Thomas, D.R., Anthony, P.,
Charlton, K.E., Maggio, M., Tsai, A.C., Grathwohl, D., Vellas, B. & Sieber, C.C. (2009)
Validation of the Mini Nutritional Assessment short-form (MNA-SF): a practical tool for
identification of nutritional status. J Nutr Health Aging, 13, 782-788.
Kaiser, M.J., Bauer, J.M., Uter, W., Donini, L.M., Stange, I., Volkert, D., Diekmann, R., Drey, M.,
Bollwein, J., Tempera, S., Guerra, A., Ricciardi, L.M. & Sieber, C.C. (2011) Prospective
234
validation of the modified mini nutritional assessment short-forms in the community, nursing
home, and rehabilitation setting. J Am Geriatr Soc, 59, 2124-2128.
Kalantar-Zadeh, K., Kleiner, M., Dunne, E., Lee, G.H. & Luft, F.C. (1999) A modified quantitative
subjective global assessment of nutrition for dialysis patients. Nephrol Dial Transplant, 14, 1732-
1738.
Kalantar-Zadeh, K., Kopple, J.D., Block, G. & Humphreys, M.H. (2001) A malnutrition-inflammation
score is correlated with morbidity and mortality in maintenance hemodialysis patients. Am J
Kidney Dis, 38, 1251-1263.
Kalantar-Zadeh, K., McAllister, C.J., Lehn, R.S., Lee, G.H., Nissenson, A.R. & Kopple, J.D. (2003)
Effect of malnutrition-inflammation complex syndrome on EPO hyporesponsiveness in
maintenance hemodialysis patients. Am J Kidney Dis, 42, 761-773.
Kaminski, M.V., Jr. (1988) Impact of nutrition support on patient outcome and hospital costs. J Can Diet
Assoc, 49, 85-88.
Kaplan, M.H. & Feinstein, A.R. (1974) The importance of classifying initial co-morbidity in evaluating
the outcome of diabetes mellitus. J Chronic Dis, 27, 387-404.
Kay, D., Blue, A., Pye, P., Lacy, A., Gray, C. & Moore, S. (2006) Heart failure: improving the continuum
of care. Care Manag, 7, 58-63.
Keith, J.N. (2008) Bedside nutrition assessment past, present, and future: a review of the Subjective
Global Assessment. Nutr Clin Pract, 23, 410-416.
Keller, H.H., Haresign, H. & Brockest, B. (2007) Process evaluation of bringing nutrition screening to
seniors in Canada (BNSS). Can J Diet Pract Res, 68, 86-91.
Kelly, I.E., Tessier, S., Cahill, A., Morris, S.E., Crumley, A., McLaughlin, D., McKee, R.F. & Lean, M.E.
(2000) Still hungry in hospital: identifying malnutrition in acute hospital admissions. QJM, 93,
93-98.
Kim, J.Y., Wie, G.A., Cho, Y.A., Kim, S.Y., Kim, S.M., Son, K.H., Park, S.J., Nam, B.H. & Joung, H.
(2011) Development and validation of a nutrition screening tool for hospitalized cancer patients.
Clin Nutr, 30, 724-729.
Kinosian, B. & Jeejeebhoy, K.N. (1995) What is malnutrition? Does it matter? Nutrition, 11, 196-197.
Kirkland, L.L., Kashiwagi, D.T., Brantley, S., Scheurer, D. & Varkey, P. (2013) Nutrition in the
hospitalized patient. J Hosp Med, 8, 52-58.
Klein, S. (1990) The myth of serum albumin as a measure of nutritional status. Gastroenterology, 99,
1845-1846.
Kondrup, J., Allison, S.P., Elia, M., Vellas, B. & Plauth, M. (2003a) ESPEN guidelines for nutrition
screening 2002. Clin Nutr, 22, 415-421.
Kondrup, J., Johansen, N., Plum, L.M., Bak, L., Larsen, I.H., Martinsen, A., Andersen, J.R., Baernthsen,
H., Bunch, E. & Lauesen, N. (2002) Incidence of nutritional risk and causes of inadequate
nutritional care in hospitals. Clin Nutr, 21, 461-468.
Kondrup, J., Rasmussen, H.H., Hamberg, O., Stanga, Z. & Group., A.H.E.W. (2003b) Nutritional risk
screening (NRS 2002): a new method based on an analysis of controlled clinical trials. Clin Nutr,
22, 321-336.
235
Koretz, R.L., Avenell, A., Lipman, T.O., Braunschweig, C.L. & Milne, A.C. (2007) Does enteral nutrition
affect clinical outcome? A systematic review of the randomized trials. Am J Gastroenterol, 102,
412-429.
Kovacevich, D.S., Boney, A.R., Braunschweig, C.L., Perez, A. & Stevens, M. (1997) Nutrition risk
classification: a reproducible and valid tool for nurses. Nutr Clin Pract, 12, 20-25.
Kruizenga, H.M., Seidell, J.C., de Vet, H.C., Wierdsma, N.J. & van Bokhorst-de van der Schueren, M.A.
(2005a) Development and validation of a hospital screening tool for malnutrition: the short
nutritional assessment questionnaire (SNAQ). Clin Nutr, 24, 75-82.
Kruizenga, H.M., Van Tulder, M.W., Seidell, J.C., Thijs, A., Ader, H.J. & Van Bokhorst-de van der
Schueren, M.A. (2005b) Effectiveness and cost-effectiveness of early screening and treatment of
malnourished patients. Am J Clin Nutr, 82, 1082-1089.
Kubrak, C. & Jensen, L. (2007) Malnutrition in acute care patients: a narrative review. Int J Nurs Stud, 44,
1036-1054.
Kuzu, M.A., Terzioglu, H., Genc, V., Erkek, A.B., Ozban, M., Sonyurek, P., Elhan, A.H. & Torun, N.
(2006) Preoperative nutritional risk assessment in predicting postoperative outcome in patients
undergoing major surgery. World J Surg, 30, 378-390.
Kwon, J., Suzuki, T., Kim, H., Yoon, H. & Lee, S. (2004) Effects of home-visit nutrition education on
nutritional status improvement of an urban community-dwelling elderly women in Korea. Nippon
Koshu Eisei Zasshi, 51, 391-402.
Kyle, U.G., Kossovsky, M.P., Karsegard, V.L. & Pichard, C. (2006) Comparison of tools for nutritional
assessment and screening at hospital admission: a population study. Clin Nutr, 25, 409-417.
Kyle, U.G., Pirlich, M., Schuetz, T., Luebke, H.J., Lochs, H. & Pichard, C. (2003) Prevalence of
malnutrition in 1760 patients at hospital admission: a controlled population study of body
composition. Clin Nutr, 22, 473-481.
Lacey, K. & Pritchett, E. (2003) Nutrition Care Process and Model: ADA adopts road map to quality care
and outcomes management. J Am Diet Assoc, 103, 1061-1072.
Lamb, C.A., Parr, J., Lamb, E.I. & Warren, M.D. (2009) Adult malnutrition screening, prevalence and
management in a United Kingdom hospital: cross-sectional study. Br J Nutr, 102, 571-575.
Landbo, C., Prescott, E., Lange, P., Vestbo, J. & Almdal, T.P. (1999) Prognostic value of nutritional status
in chronic obstructive pulmonary disease. Am J Respir Crit Care Med, 160, 1856-1861.
Lang, P.O., Heitz, D., Hedelin, G., Drame, M., Jovenin, N., Ankri, J., Somme, D., Novella, J.L., Gauvain,
J.B., Couturier, P., Voisin, T., De Waziere, B., Gonthier, R., Jeandel, C., Jolly, D., Saint-Jean, O.
& Blanchard, F. (2006) Early markers of prolonged hospital stays in older people: a prospective,
multicenter study of 908 inpatients in French acute hospitals. J Am Geriatr Soc, 54, 1031-1039.
Laporte, M., Villalon, L. & Payette, H. (2001a) Simple nutrition screening tools for healthcare facilities:
development and validity assessment. Can J Diet Pract Res, 62, 26-34.
Laporte, M., Villalon, L., Thibodeau, J. & Payette, H. (2001b) Validity and reliability of simple nutrition
screening tools adapted to the elderly population in healthcare facilities. J Nutr Health Aging, 5,
292-294.
Latkany, L., Lloyd, M.E. & Schaeffer, A. (1995) Development of adult and pediatric oncology nutrition
screening tools Top Clin Nutr, 10.
236
Law, M. & Letts, L. (1989) A critical review of scales of activities of daily living. Am J Occup Ther, 43,
522-528.
Lawson, C.S., Campbell, K.L., Dimakopoulos, I. & Dockrell, M.E. (2012) Assessing the validity and
reliability of the MUST and MST nutrition screening tools in renal inpatients. J Ren Nutr, 22,
499-506.
Lawson, J.A., Lazarus, R. & Kelly, J.J. (2001) Prevalence and prognostic significance of malnutrition in
chronic renal insufficiency. J Ren Nutr, 11, 16-22.
Lazarus, C. & Hamlyn, J. (2005) Prevalence and documentation of malnutrition in hospitals:A case study
in a large private hospital setting. Nutr Diet, 41-47.
Lim, S.L. (2012) Using Expedited 10-g Protein Counter (EP-10) for Meal Planning. J Ren Nutr, 22, e55-
56.
Lim, S.L., Ang, E., Foo, Y.L., Ng, L.Y., Tong, C.Y., Ferguson, M. & Daniels, L. (2013a) Validity and
reliability of nutrition screening administered by nurses. Nutr Clin Pract 28, 730 - 736.
Lim, S.L. & Daniels, L. (2013) Reply - Malnutrition and its impact on cost of hospitalization, length of
stay, readmission and 3-year mortality. Clin Nutr, 32, 489-490.
Lim, S.L., Lin, X.H., Chan, Y.H., Ferguson, M. & Daniels, L. (2013b) A pre-post evaluation of an
ambulatory nutrition support service for malnourished patients post hospital discharge: a pilot
study. Ann Acad Med Singapore, 42, 507-513.
Lim, S.L. & Lye, J. (2012) Nutritional intervention incorporating expedited 10 g protein counter (EP-10)
to improve the albumin and transferrrin of chronic hemodialysis patients. International Scholarly
Reserch Network - Nutrition, 2013.
Lim, S.L., Ong, K.C., Chan, Y.H., Loke, W.C., Ferguson, M. & Daniels, L. (2012) Malnutrition and its
impact on cost of hospitalization, length of stay, readmission and 3-year mortality. Clin Nutr, 31,
345-350.
Lim, S.L., Tong, C.Y., Ang, E., Lee, E.J., Loke, W.C., Chen, Y., Ferguson, M. & Daniels, L. (2009)
Development and validation of 3-Minute Nutrition Screening (3-MinNS) tool for acute hospital
patients in Singapore. Asia Pac J Clin Nutr, 18, 395-403.
Lindorff-Larsen, K., Hojgaard Rasmussen, H., Kondrup, J., Staun, M. & Ladefoged, K. (2007)
Management and perception of hospital undernutrition-a positive change among Danish doctors
and nurses. Clin Nutr, 26, 371-378.
Linn, B.S., Linn, M.W. & Gurel, L. (1968) Cumulative illness rating scale. J Am Geriatr Soc, 16, 622-
626.
Lloyd-Jones, D.M., Liu, K., Colangelo, L.A., Yan, L.L., Klein, L., Loria, C.M., Lewis, C.E. & Savage, P.
(2007) Consistently stable or decreased body mass index in young adulthood and longitudinal
changes in metabolic syndrome components: the Coronary Artery Risk Development in Young
Adults Study. Circulation, 115, 1004-1011.
Lochs, H., Allison, S.P., Meier, R., Pirlich, M., Kondrup, J., Schneider, S., van den Berghe, G. & Pichard,
C. (2006) Introductory to the ESPEN Guidelines on Enteral Nutrition: Terminology, definitions
and general topics. Clin Nutr, 25, 180-186.
Lundvick, J. & Phillips, R. (1983) Nutritional screening of the oncology patient Nutr Support Serv, 3, 21-
24.
237
Lyne, P.A. & Prowse, M.A. (1999) Methodological issues in the development and use of instruments to
assess patient nutritional status or the level of risk of nutritional compromise. J Adv Nurs, 30,
835-842.
Mahoney, F.I. & Barthel, D.W. (1965) Functional Evaluation: The Barthel Index. Md State Med J, 14, 61-
65.
Makhija, S. & Baker, J. (2008) The Subjective Global Assessment: a review of its use in clinical practice.
Nutr Clin Pract, 23, 405-409.
Marco, J., Barba, R., Zapatero, A., Matia, P., Plaza, S., Losa, J.E., Canora, J. & Garcia de Casasola, G.
(2011) Prevalence of the notification of malnutrition in the departments of internal medicine and
its prognostic implications. Clin Nutr, 30, 450-454.
Marshall, J.K., Gadowsky, S.L., Childs, A. & Armstrong, D. (2005) Economic analysis of home vs
hospital-based parenteral nutrition in Ontario, Canada. JPEN J Parenter Enteral Nutr, 29, 266-
269.
Martineau, J., Bauer, J.D., Isenring, E. & Cohen, S. (2005) Malnutrition determined by the patient-
generated subjective global assessment is associated with poor outcomes in acute stroke patients.
Clin Nutr, 24, 1073-1077.
Matos, L.C., Tavares, M.M. & Amaral, T.F. (2007) Handgrip strength as a hospital admission nutritional
risk screening method. Eur J Clin Nutr, 61, 1128-1135.
McBride, C.M. & Rimer, B.K. (1999) Using the telephone to improve health behavior and health service
delivery. Patient Educ Couns, 37, 3-18.
McCall, R. & Cotton, E. (2001) The validation of a nursing nutritional assessment tool for use on acute
elderly wards. J Hum Nutr Diet, 14, 137-148.
McCann, L. (1999) Using subjective global assessment to identify malnutrition in the ESRD patient.
Nephrol News Issues, 13, 18-19.
McWhirter, J.P. & Pennington, C.R. (1994) Incidence and recognition of malnutrition in hospital. BMJ,
308, 945-948.
Meijers, J.M., Candel, M.J., Schols, J.M., van Bokhorst-de van der Schueren, M.A. & Halfens, R.J. (2009)
Decreasing trends in malnutrition prevalence rates explained by regular audits and feedback. J
Nutr, 139, 1381-1386.
Middleton, M.H., Nazarenko, G., Nivison-Smith, I. & Smerdely, P. (2001) Prevalence of malnutrition and
12-month incidence of mortality in two Sydney teaching hospitals. Intern Med J, 31, 455-461.
Miller, M.D., Crotty, M., Giles, L.C., Bannerman, E., Whitehead, C., Cobiac, L., Daniels, L.A. &
Andrews, G. (2002) Corrected arm muscle area: an independent predictor of long-term mortality
in community-dwelling older adults? J Am Geriatr Soc, 50, 1272-1277.
Milne, A.C., Potter, J., Vivanti, A. & Avenell, A. (2009) Protein and energy supplementation in elderly
people at risk from malnutrition. Cochrane Database Syst Rev, CD003288.
Mirmiran, P., Hosseinpour-Niazi, S., Mehrabani, H.H., Kavian, F. & Azizi, F. (2011) Validity and
reliability of a nutrition screening tool in hospitalized patients. Nutrition, 27, 647-652.
Morley, J.E., Argiles, J.M., Evans, W.J., Bhasin, S., Cella, D., Deutz, N.E., Doehner, W., Fearon, K.C.,
Ferrucci, L., Hellerstein, M.K., Kalantar-Zadeh, K., Lochs, H., MacDonald, N., Mulligan, K.,
Muscaritoli, M., Ponikowski, P., Posthauer, M.E., Rossi Fanelli, F., Schambelan, M., Schols,
238
A.M., Schuster, M.W. & Anker, S.D. (2010) Nutritional recommendations for the management
of sarcopenia. J Am Med Dir Assoc, 11, 391-396.
Mowe, M., Bosaeus, I., Rasmussen, H.H., Kondrup, J., Unosson, M. & Irtun, O. (2006) Nutritional
routines and attitudes among doctors and nurses in Scandinavia: a questionnaire based survey.
Clin Nutr, 25, 524-532.
Mudge, A.M., Denaro, C.P. & O'Rourke, P. (2012) Improving hospital outcomes in patients admitted
from residential aged care: results from a controlled trial. Age Ageing, 41, 670-673.
Mueller, C., Compher, C. & Ellen, D.M. (2011) A.S.P.E.N. Board of Directors Clinical Guidelines:
Nutrition screening, assessment, and intervention in adults. JPEN J Parenter Enteral Nutr, 35,
16-24.
Muinuddin, A., Aslahi, R., Hopman, W. & Paterson, W.G. (2013) Relationship between the number of
positive fecal occult blood tests and the diagnostic yield of colonoscopy. Can J Gastroenterol,
27, 90-94.
Mulrow, C.D., Gerety, M.B., Cornell, J.E., Lawrence, V.A. & Kanten, D.N. (1994) The relationship
between disease and function and perceived health in very frail elders. J Am Geriatr Soc, 42,
374-380.
Naber, T.H., Schermer, T., de Bree, A., Nusteling, K., Eggink, L., Kruimel, J.W., Bakkeren, J., van
Heereveld, H. & Katan, M.B. (1997) Prevalence of malnutrition in nonsurgical hospitalized
patients and its association with disease complications. Am J Clin Nutr, 66, 1232-1239.
National Health and Medical Research Council (2009) NHMRC levels of evidence and grades for
recommendations for developers of guidelines. Stage 2 Consultation. Canberra: National Health
and Medical Research Council.
Neelemaat, F., Bosmans, J.E., Thijs, A., Seidell, J.C. & van Bokhorst-de van der Schueren, M.A. (2011a)
Post-discharge nutritional support in malnourished elderly individuals improves functional
limitations. J Am Med Dir Assoc, 12, 295-301.
Neelemaat, F., Meijers, J., Kruizenga, H., van Ballegooijen, H. & van Bokhorst-de van der Schueren, M.
(2011b) Comparison of five malnutrition screening tools in one hospital inpatient sample. J Clin
Nurs, 20, 2144-2152.
Neumann, S.A., Miller, M.D., Daniels, L. & Crotty, M. (2005) Nutritional status and clinical outcomes of
older patients in rehabilitation. J Hum Nutr Diet, 18, 129-136.
Nightingale, J.M., Walsh, N., Bullock, M.E. & Wicks, A.C. (1996) Three simple methods of detecting
malnutrition on medical wards. J R Soc Med, 89, 144-148.
Nikolaus, T., Bach, M., Siezen, S., Volkert, D., Oster, P. & Schlierf, G. (1995) Assessment of nutritional
risk in the elderly. Ann Nutr Metab, 39, 340-345.
Noel, M.B. & Wojnaroski, S.M. (1987) Nutrition screening for long-term care residents. J Am Diet Assoc,
87, 1557-1558.
Norberg, M., Eriksson, J.W., Lindahl, B., Andersson, C., Rolandsson, O., Stenlund, H. & Weinehall, L.
(2006) A combination of HbA1c, fasting glucose and BMI is effective in screening for
individuals at risk of future type 2 diabetes: OGTT is not needed. J Intern Med, 260, 263-271.
Norman, K., Kirchner, H., Freudenreich, M., Ockenga, J., Lochs, H. & Pirlich, M. (2008a) Three month
intervention with protein and energy rich supplements improve muscle function and quality of
life in malnourished patients with non-neoplastic gastrointestinal disease--a randomized
controlled trial. Clin Nutr, 27, 48-56.
239
Norman, K., Kirchner, H., Lochs, H. & Pirlich, M. (2006) Malnutrition affects quality of life in
gastroenterology patients. World J Gastroenterol, 12, 3380-3385.
Norman, K., Pichard, C., Lochs, H. & Pirlich, M. (2008b) Prognostic impact of disease-related
malnutrition. Clin Nutr, 27, 5-15.
Norman, K., Schutz, T., Kemps, M., Josef Lubke, H., Lochs, H. & Pirlich, M. (2005) The Subjective
Global Assessment reliably identifies malnutrition-related muscle dysfunction. Clin Nutr, 24,
143-150.
Norman, K., Stobaus, N., Gonzalez, M.C., Schulzke, J.D. & Pirlich, M. (2011) Hand grip strength:
Outcome predictor and marker of nutritional status. Clin Nutr, 30, 135-142.
Nursal, T.Z., Noyan, T., Atalay, B.G., Koz, N. & Karakayali, H. (2005a) Simple two-part tool for
screening of malnutrition. Nutrition, 21, 659-665.
Nursal, T.Z., Noyan, T., Tarim, A. & Karakayali, H. (2005b) A new weighted scoring system for
Subjective Global Assessment. Nutrition, 21, 666-671.
O'Flynn, J., Peake, H., Hickson, M., Foster, D. & Frost, G. (2005) The prevalence of malnutrition in
hospitals can be reduced: results from three consecutive cross-sectional studies. Clin Nutr, 24,
1078-1088.
O'Keefe, S.J., Dicker, J. & Delport, I. (1986) Incidence of malnutrition in adult patients at Groote Schuur
Hospital, 1984. S Afr Med J, 70, 16-20.
Oakley, C. & Hill, R. (2000) Nutrition Assessment Score validation and the implications for usage. J Hum
Nutr Diet, 13, 343-352.
OECD. (2011) "Average length of stay in hospitals, in Health at a Glance. OECD Indicators, OECD
Publishing, Accessed on 21 December 2012, Available from:
http://dx.doi.org/2010.1787/health_glance-2011-2033-en.
Onalan, R., Onalan, G., Tonguc, E., Ozdener, T., Dogan, M. & Mollamahmutoglu, L. (2009) Body mass
index is an independent risk factor for the development of endometrial polyps in patients
undergoing in vitro fertilization. Fertil Steril, 91, 1056-1060.
Otte, K.E., Ahlburg, P., D'Amore, F. & Stellfeld, M. (1989) Nutritional repletion in malnourished patients
with emphysema. JPEN J Parenter Enteral Nutr, 13, 152-156.
Ottery, F.D. (1994) Rethinking nutritional support of the cancer patient: the new field of nutritional
oncology. Semin Oncol, 21, 770-778.
Ottery, F.D. (1996) Definition of standardized nutritional assessment and interventional pathways in
oncology. Nutrition, 12, S15-19.
Oxner, C.R., Vora, L., Yim, J., Kruper, L. & Ellenhorn, J.D. (2012) Magnetic resonance imaging-guided
breast biopsy in lesions not visualized by mammogram or ultrasound. Am Surg, 78, 1087-1090.
Paccagnella, A., Morello, M., Da Mosto, M.C., Baruffi, C., Marcon, M.L., Gava, A., Baggio, V., Lamon,
S., Babare, R., Rosti, G., Giometto, M., Boscolo-Rizzo, P., Kiwanuka, E., Tessarin, M.,
Caregaro, L. & Marchiori, C. (2010) Early nutritional intervention improves treatment tolerance
and outcomes in head and neck cancer patients undergoing concurrent chemoradiotherapy.
Support Care Cancer, 18, 837-845.
Parente, F., Marino, B., DeVecchi, N., Moretti, R., Ucci, G., Tricomi, P., Armellino, A., Redaelli, L.,
Bargiggia, S., Cristofori, E., Masala, E., Tortorella, F., Gattinoni, A., Odinolfi, F. & Pirola, M.E.
240
(2009) Faecal occult blood test-based screening programme with high compliance for
colonoscopy has a strong clinical impact on colorectal cancer. Br J Surg, 96, 533-540.
Paton, N.I., Chua, Y.K., Earnest, A. & Chee, C.B. (2004) Randomized controlled trial of nutritional
supplementation in patients with newly diagnosed tuberculosis and wasting. Am J Clin Nutr, 80,
460-465.
Pattison, R., Corr, J., Ogilvie, M., Farquhar, D., Sutherland, D., Davidson, H.I.M. & Richardson, R.A.
(1999) Reliability of a qualitative screening tool versus physical measurements in identifying
undernutrition in an elderly population. J Hum Nutr Diet, 12, 133140.
Pennington, C.R. & McWhirter, J.P. (1997) Patients go hungry in British hospitals. Malnutrition is
common, unrecognised, and treatable in hospital patients. BMJ, 314, 752.
Pereira Machado, R.S. & Santa Cruz Coelho, M.A. (2011) Risk of malnutrition among Brazilian
institutionalized elderly: a study with the Mini Nutritional Assessment (MNA) questionnaire. J
Nutr Health Aging, 15, 532-535.
Persenius, M.W., Hall-Lord, M.L., Baath, C. & Larsson, B.W. (2008) Assessment and documentation of
patients' nutritional status: perceptions of registered nurses and their chief nurses. J Clin Nurs, 17,
2125-2136.
Persson, M.D., Brismar, K.E., Katzarski, K.S., Nordenstrom, J. & Cederholm, T.E. (2002) Nutritional
status using mini nutritional assessment and subjective global assessment predict mortality in
geriatric patients. J Am Geriatr Soc, 50, 1996-2002.
Pham, N.V., Cox-Reijven, P.L., Wodzig, W.K., Greve, J.W. & Soeters, P.B. (2007) SGA and measures
for muscle mass and strength in surgical Vietnamese patients. Nutrition, 23, 283-291.
Pirlich, M., Schutz, T., Kemps, M., Luhman, N., Burmester, G.R., Baumann, G., Plauth, M., Lubke, H.J.
& Lochs, H. (2003) Prevalence of malnutrition in hospitalized medical patients: impact of
underlying disease. Dig Dis, 21, 245-251.
Pirlich, M., Schutz, T., Kemps, M., Luhman, N., Minko, N., Lubke, H.J., Rossnagel, K., Willich, S.N. &
Lochs, H. (2005) Social risk factors for hospital malnutrition. Nutrition, 21, 295-300.
Pirlich, M., Schutz, T., Norman, K., Gastell, S., Lubke, H.J., Bischoff, S.C., Bolder, U., Frieling, T.,
Guldenzoph, H., Hahn, K., Jauch, K.W., Schindler, K., Stein, J., Volkert, D., Weimann, A.,
Werner, H., Wolf, C., Zurcher, G., Bauer, P. & Lochs, H. (2006) The German hospital
malnutrition study. Clin Nutr, 25, 563-572.
Planas, M., Audivert, S., Perez-Portabella, C., Burgos, R., Puiggros, C., Casanelles, J.M. & Rossello, J.
(2004) Nutritional status among adult patients admitted to an university-affiliated hospital in
Spain at the time of genoma. Clin Nutr, 23, 1016-1024.
Porter, J., Raja, R., Cant, R. & Aroni, R. (2009) Exploring issues influencing the use of the Malnutrition
Universal Screening Tool by nurses in two Australian hospitals. J Hum Nutr Diet, 22, 203-209.
Potter, J., Langhorne, P. & Roberts, M. (1998) Routine protein energy supplementation in adults:
systematic review. BMJ, 317, 495-501.
Poulia, K.A., Yannakoulia, M., Karageorgou, D., Gamaletsou, M., Panagiotakos, D.B., Sipsas, N.V. &
Zampelas, A. (2012) Evaluation of the efficacy of six nutritional screening tools to predict
malnutrition in the elderly. Clin Nutr, 31, 378-385.
Pua, Y.H. & Ong, P.H. (2004) Anthropometric indices as screening tools for cardiovascular risk factors in
Singaporeans: receiver operating characteristic curves analysis. Ann Acad Med Singapore, 33,
S28-30.
241
Raja, R., Gibson, S., Turner, A., Winderlich, J., Porter, J., Cant, R. & Aroni, R. (2008) Nurses views and
practices regarding use of validated nutrition screening tools. Aust J Adv Nurs 26, 26-33.
Raja, R., Lim, A.V., Lim, Y.P., Lim, G., Chan, S.P. & Vu, C.K. (2004) Malnutrition screening in
hospitalised patients and its implication on reimbursement. Intern Med J, 34, 176-181.
Rasheed, S. & Woods, R.T. (2013a) Malnutrition and associated clinical outcomes in hospitalized patients
aged 60 and older: an observational study in rural Wales. J Nutr Gerontol Geriatr, 32, 71-80.
Rasheed, S. & Woods, R.T. (2013b) Malnutrition and quality of life in older people: A systematic review
and meta-analysis. Ageing Res Rev, 12, 561-566.
Rasmussen, H.H., Kondrup, J., Ladefoged, K. & Staun, M. (1999) Clinical nutrition in danish hospitals: a
questionnaire-based investigation among doctors and nurses. Clin Nutr, 18, 153-158.
Rasmussen, H.H., Kondrup, J., Staun, M., Ladefoged, K., Kristensen, H. & Wengler, A. (2004)
Prevalence of patients at nutritional risk in Danish hospitals. Clin Nutr, 23, 1009-1015.
Rauscher, C. (1993) Malnutrition among the elderly. Can Fam Physician, 39, 1395-1403.
Ravasco, P., Monteiro-Grillo, I., Marques Vidal, P. & Camilo, M.E. (2005a) Impact of nutrition on
outcome: a prospective randomized controlled trial in patients with head and neck cancer
undergoing radiotherapy. Head Neck, 27, 659-668.
Ravasco, P., Monteiro-Grillo, I., Vidal, P.M. & Camilo, M.E. (2005b) Dietary counseling improves
patient outcomes: a prospective, randomized, controlled trial in colorectal cancer patients
undergoing radiotherapy. J Clin Oncol, 23, 1431-1438.
Rawlinson, D. (1998) Audit of nutritional practice and knowledge. Prof Nurse, 13, 291-294.
Reilly, H.M., Martineau, J.K., Moran, A. & Kennedy, H. (1995) Nutritional screening - evaluation and
implementation of a simple Nutrition Risk Score. Clin Nutr, 14, 269-273.
Robinson, G., Goldstein, M. & Levine, G.M. (1987) Impact of nutritional status on DRG length of stay.
JPEN J Parenter Enteral Nutr, 11, 49-51.
Rosenthal, A.J., Sanders, K.M., McMurtry, C.T., Jacobs, M.A., Thompson, D.D., Gheorghiu, D., Little,
K.L. & Adler, R.A. (1998) Is malnutrition overdiagnosed in older hospitalized patients?
Association between the soluble interleukin-2 receptor and serum markers of malnutrition. J
Gerontol A Biol Sci Med Sci, 53, M81-86.
Rowell, A., Long, C., Chance, L. & Dolley, O. (2012) Identification of nutritional risk by nursing staff in
secure psychiatric settings: reliability and validity of St Andrew's Nutrition Screening Instrument.
J Psychiatr Ment Health Nurs, 19, 722-728.
Rowell, D.S. & Jackson, T.J. (2011) Additional costs of inpatient malnutrition, Victoria, Australia, 2003-
2004. Eur J Health Econ, 12, 353-361.
Rubenstein, L.Z., Harker, J., Guigoz, Y. & Vellas, B. (1999) Comprehensive geriatric assessment (CGA)
and the MNA: an overview of CGA, nutritional assessment, and development of a shortened
version of the MNA. Nestle Nutr Workshop Ser Clin Perform Programme, 1, 101-116.
Rufenacht, U., Ruhlin, M., Wegmann, M., Imoberdorf, R. & Ballmer, P.E. (2010) Nutritional counseling
improves quality of life and nutrient intake in hospitalized undernourished patients. Nutrition, 26,
53-60.
Rush, D. (1993) Evaluating the nutrition screening initiative. Am J Public Health, 83, 944-945.
242
Rypkema, G., Adang, E., Dicke, H., Naber, T., de Swart, B., Disselhorst, L., Goluke-Willemse, G. & Olde
Rikkert, M. (2004) Cost-effectiveness of an interdisciplinary intervention in geriatric inpatients to
prevent malnutrition. J Nutr Health Aging, 8, 122-127.
Sacks, G.S., Dearman, K., Replogle, W.H., Cora, V.L., Meeks, M. & Canada, T. (2000) Use of subjective
global assessment to identify nutrition-associated complications and death in geriatric long-term
care facility residents. J Am Coll Nutr, 19, 570-577.
Sanada, H., Yokokawa, H., Yoneda, M., Yatabe, J., Sasaki Yatabe, M., Williams, S.M., Felder, R.A. &
Jose, P.A. (2012) High body mass index is an important risk factor for the development of type 2
diabetes. Intern Med, 51, 1821-1826.
Sayer, A.A., Robinson, S.M., Patel, H.P., Shavlakadze, T., Cooper, C. & Grounds, M.D. (2013) New
horizons in the pathogenesis, diagnosis and management of sarcopenia. Age Ageing, 42, 145-150.
Scanlan, F., Dunne, J. & Toyne, K. (1994) No more cause for neglect. Introducing a nutritional
assessment tool and action plan. Prof Nurse, 9, 382, 384-385.
Schindler, K., Pernicka, E., Laviano, A., Howard, P., Schutz, T., Bauer, P., Grecu, I., Jonkers, C.,
Kondrup, J., Ljungqvist, O., Mouhieddine, M., Pichard, C., Singer, P., Schneider, S., Schuh, C. &
Hiesmayr, M. (2010) How nutritional risk is assessed and managed in European hospitals: a
survey of 21,007 patients findings from the 2007-2008 cross-sectional nutritionDay survey. Clin
Nutr, 29, 552-559.
Schneider, S.M. & Hebuterne, X. (2000) Use of nutritional scores to predict clinical outcomes in chronic
diseases. Nutr Rev, 58, 31-38.
Scolapio, J.S., Bowen, J., Stoner, G. & Tarrosa, V. (2000) Substrate oxidation in patients with cirrhosis:
comparison with other nutritional markers. JPEN J Parenter Enteral Nutr, 24, 150-153.
Scott, A. & Hamilton, K. (1998) Nutritional screening: an audit. Nurs Stand, 12, 46-47.
Sermet-Gaudelus, I., Poisson-Salomon, A.S., Colomb, V., Brusset, M.C., Mosser, F., Berrier, F. & Ricour,
C. (2000) Simple pediatric nutritional risk score to identify children at risk of malnutrition. Am J
Clin Nutr, 72, 64-70.
Shenkin, A. (2006) Serum prealbumin: Is it a marker of nutritional status or of risk of malnutrition? Clin
Chem, 52, 2177-2179.
Shirley, S., Davis, L.L. & Carlson, B.W. (2008) The relationship between body mass index/body
composition and survival in patients with heart failure. J Am Acad Nurse Pract, 20, 326-332.
Singh, H., Watt, K., Veitch, R., Cantor, M. & Duerksen, D.R. (2006) Malnutrition is prevalent in
hospitalized medical patients: are housestaff identifying the malnourished patient? Nutrition, 22,
350-354.
Skipper, A., Ferguson, M., Thompson, K., Castellanos, V.H. & Porcari, J. (2012) Nutrition screening
tools: an analysis of the evidence. JPEN J Parenter Enteral Nutr, 36, 292-298.
Smedley, F., Bowling, T., James, M., Stokes, E., Goodger, C., O'Connor, O., Oldale, C., Jones, P. & Silk,
D. (2004) Randomized clinical trial of the effects of preoperative and postoperative oral
nutritional supplements on clinical course and cost of care. Br J Surg, 91, 983-990.
Soeters, P.B., Reijven, P.L., van Bokhorst-de van der Schueren, M.A., Schols, J.M., Halfens, R.J.,
Meijers, J.M. & van Gemert, W.G. (2008) A rational approach to nutritional assessment. Clin
Nutr, 27, 706-716.
243
Somanchi, M., Tao, X. & Mullin, G.E. (2011) The facilitated early enteral and dietary management
effectiveness trial in hospitalized patients with malnutrition. JPEN J Parenter Enteral Nutr, 35,
209-216.
Starke, J., Schneider, H., Alteheld, B., Stehle, P. & Meier, R. (2011) Short-term individual nutritional care
as part of routine clinical setting improves outcome and quality of life in malnourished medical
patients. Clin Nutr, 30, 194-201.
Steenson, J., Vivanti, A. & Isenring, E. (2013) Inter-rater reliability of the Subjective Global Assessment:
a systematic literature review. Nutrition, 29, 350-352.
Steiber, A., Leon, J.B., Secker, D., McCarthy, M., McCann, L., Serra, M., Sehgal, A.R. & Kalantar-
Zadeh, K. (2007) Multicenter study of the validity and reliability of subjective global assessment
in the hemodialysis population. J Ren Nutr, 17, 336-342.
Stephenson, G.R., Moretti, E.W., El-Moalem, H., Clavien, P.A. & Tuttle-Newhall, J.E. (2001)
Malnutrition in liver transplant patients: preoperative subjective global assessment is predictive
of outcome after liver transplantation. Transplantation, 72, 666-670.
Stratton, R.J., Ek, A.C., Engfer, M., Moore, Z., Rigby, P., Wolfe, R. & Elia, M. (2005) Enteral nutritional
support in prevention and treatment of pressure ulcers: a systematic review and meta-analysis.
Ageing Res Rev, 4, 422-450.
Stratton, R.J. & Elia, M. (2000) Are oral nutritional supplements of benefit to patients in the community?
Findings from a systematic review. Curr Opin Clin Nutr Metab Care, 3, 311-315.
Stratton, R.J., Hackston, A., Longmore, D., Dixon, R., Price, S., Stroud, M., King, C. & Elia, M. (2004)
Malnutrition in hospital outpatients and inpatients: prevalence, concurrent validity and ease of
use of the 'malnutrition universal screening tool' ('MUST') for adults. Br J Nutr, 92, 799-808.
Stratton, R.J., King, C.L., Stroud, M.A., Jackson, A.A. & Elia, M. (2006) 'Malnutrition Universal
Screening Tool' predicts mortality and length of hospital stay in acutely ill elderly. Br J Nutr, 95,
325-330.
Sullivan, D.H., Sun, S. & Walls, R.C. (1999) Protein-energy undernutrition among elderly hospitalized
patients: a prospective study. JAMA, 281, 2013-2019.
Sullivan, D.H. & Walls, R.C. (1998) Protein-energy undernutrition and the risk of mortality within six
years of hospital discharge. J Am Coll Nutr, 17, 571-578.
Sung, K.C. & Ryu, S.H. (2004) Insulin resistance, body mass index, waist circumference are independent
risk factor for high blood pressure. Clin Exp Hypertens, 26, 547-556.
Tangvik, R.J., Guttormsen, A.B., Tell, G.S. & Ranhoff, A.H. (2012) Implementation of nutritional
guidelines in a university hospital monitored by repeated point prevalence surveys. Eur J Clin
Nutr, 66, 388-393.
Tappenden, K.A., Quatrara, B., Parkhurst, M.L., Malone, A.M., Fanjiang, G. & Ziegler, T.R. (2013)
Critical role of nutrition in improving quality of care: An interdisciplinary call to action to
address adult hospital malnutrition. J Acad Nutr Diet, 113, 1219-1237.
Theou, O., Stathokostas, L., Roland, K.P., Jakobi, J.M., Patterson, C., Vandervoort, A.A. & Jones, G.R.
(2011) The effectiveness of exercise interventions for the management of frailty: a systematic
review. J Aging Res, 2011, 569194.
Thomas, B. (2001) Manual of Dietetic Practice: Undernutrition and principles of nutritional support. 3rd
ed. Blackwell Science Ltd, Oxford.
244
Thomas, D.R., Kamel, H., Azharrudin, M., Ali, A.S., Khan, A., Javaid, U. & Morley, J.E. (2005) The
relationship of functional status, nutritional assessment, and severity of illness to in-hospital
mortality. J Nutr Health Aging, 9, 169-175.
Thomas, D.R., Zdrowski, C.D., Wilson, M.M., Conright, K.C., Lewis, C., Tariq, S. & Morley, J.E. (2002)
Malnutrition in subacute care. Am J Clin Nutr, 75, 308-313.
Thomas, J.M., Isenring, E. & Kellett, E. (2007) Nutritional status and length of stay in patients admitted to
an Acute Assessment Unit. J Hum Nutr Diet, 20, 320-328.
Thompson, J.D. (1988) DRG prepayment: its purpose and performance. Bull N Y Acad Med, 64, 28-51.
Thorsdottir, I., Eriksen, B. & Eysteinsdottir, S. (1999) Nutritional status at submission for dietetic services
and screening for malnutrition at admission to hospital. Clin Nutr, 18, 15-21.
Toumi, T. & Lawson, S. (2011) Nutritional Screening in a District Hospital in the Greater Manchester
Area. WebmedCentral Surgery. Accessed on 31 Dec 2011, Available from:
http://www.webmedcentral.com, 2, WMC002188.
Tsai, A.C., Chang, T.L., Chen, J.T. & Yang, T.W. (2009) Population-specific modifications of the short-
form Mini Nutritional Assessment and Malnutrition Universal Screening Tool for elderly
Taiwanese. Int J Nurs Stud, 46, 1431-1438.
Tsai, A.C., Wang, J.Y., Chang, T.L. & Li, T.Y. (2013) A comparison of the full Mini Nutritional
Assessment, short-form Mini Nutritional Assessment, and Subjective Global Assessment to
predict the risk of protein-energy malnutrition in patients on peritoneal dialysis: A cross-sectional
study. Int J Nurs Stud, 50, 83-89.
Turnbull, P.J. & Sinclair, A.J. (2002) Evaluation of nutritional status and its relationship with functional
status in older citizens with diabetes mellitus using the mini nutritional assessment (MNA) tool--
a preliminary investigation. J Nutr Health Aging, 6, 185-189.
van Bokhorst-de van der Schueren, M.A., Klinkenberg, M. & Thijs, A. (2005) Profile of the malnourished
patient. Eur J Clin Nutr, 59, 1129-1135.
van Venrooij, L.M., de Vos, R., Borgmeijer-Hoelen, M.M., Kruizenga, H.M., Jonkers-Schuitema, C.F. &
de Mol, B.A.M.J. (2007) Quick and easy undernutrition screening tools to detect disease-related
undernutrition in- and outpatient settings: A systematic review of sensitivity and specificity. E
Spen Eur E J Clin Nutr Metab, 2, 21-37.
Vanderwee, K., Clays, E., Bocquaert, I., Gobert, M., Folens, B. & Defloor, T. (2010) Malnutrition and
associated factors in elderly hospital patients: a Belgian cross-sectional, multi-centre study. Clin
Nutr, 29, 469-476.
Vandewoude, M.F., Alish, C.J., Sauer, A.C. & Hegazi, R.A. (2012) Malnutrition-sarcopenia syndrome: is
this the future of nutrition screening and assessment for older adults? J Aging Res, 2012, 651570.
Vanis, N. & Mesihovic, R. (2008) Application of nutritional screening tests for determining prevalence of
hospital malnutrition. Med Arh, 62, 211-214.
VanWormer, J.J., Martinez, A.M., Benson, G.A., Crain, A.L., Martinson, B.C., Cosentino, D.L. & Pronk,
N.P. (2009) Telephone counseling and home telemonitoring: the Weigh by Day Trial. Am J
Health Behav, 33, 445-454.
Velasco, C., Garcia, E., Rodriguez, V., Frias, L., Garriga, R., Alvarez, J., Garcia-Peris, P. & Leon, M.
(2011) Comparison of four nutritional screening tools to detect nutritional risk in hospitalized
patients: a multicentre study. Eur J Clin Nutr, 65, 269-274.
245
Vermeeren, M.A., Wouters, E.F., Geraerts-Keeris, A.J. & Schols, A.M. (2004) Nutritional support in
patients with chronic obstructive pulmonary disease during hospitalization for an acute
exacerbation; a randomized controlled feasibility trial. Clin Nutr, 23, 1184-1192.
Veterans Affairs Total Parenteral Nutrition Cooperative Study Group (1991) Perioperative total parenteral
nutrition in surgical patients. N Engl J Med, 325, 525-532.
Visser, R., Dekker, F.W., Boeschoten, E.W., Stevens, P. & Krediet, R.T. (1999) Reliability of the 7-point
subjective global assessment scale in assessing nutritional status of dialysis patients. Adv Perit
Dial, 15, 222-225.
Visvanathan, R., Penhall, R. & Chapman, I. (2004) Nutritional screening of older people in a sub-acute
care facility in Australia and its relation to discharge outcomes. Age Ageing, 33, 260-265.
Voyce, C.J. & Seager, H.E. (2009) Incidence and frequency of screening for nutritional risk in hospital
inpatients in East Cheshire NHS Trust (Abstract). J Hum Nutr Diet, 22, 273.
Waitzberg, D.L., Caiaffa, W.T. & Correia, M.I. (2001) Hospital malnutrition: the Brazilian national
survey (IBRANUTRI): a study of 4000 patients. Nutrition, 17, 573-580.
Ware, J.E., Jr. & Sherbourne, C.D. (1992) The MOS 36-item short-form health survey (SF-36). I.
Conceptual framework and item selection. Med Care, 30, 473-483.
Watterson, C., Fraser, A., Banks, M., Isenring, E., Miller, M., Silvester, C. & al., e. (2009) Dietitians
Association of Australia. Evidence based practice guidelines for the nutritional management of
malnutrition in adult patients across the continuum of care. Nutr Diet, 66, S1-S34.
Weekes, C.E., Elia, M. & Emery, P.W. (2004) The development, validation and reliability of a nutrition
screening tool based on the recommendations of the British Association for Parenteral and
Enteral Nutrition (BAPEN). Clin Nutr, 23, 1104-1112.
Weekes, C.E., Spiro, A., Baldwin, C., Whelan, K., Thomas, J.E., Parkin, D. & Emery, P.W. (2009) A
review of the evidence for the impact of improving nutritional care on nutritional and clinical
outcomes and cost. J Hum Nutr Diet, 22, 324-335.
White, J.V., Dwyer, J.T., Posner, B.M., Ham, R.J., Lipschitz, D.A. & Wellman, N.S. (1992) Nutrition
screening initiative: development and implementation of the public awareness checklist and
screening tools. J Am Diet Assoc, 92, 163-167.
White, J.V., Guenter, P., Jensen, G., Malone, A. & Schofield, M. (2012) Consensus statement: Academy
of Nutrition and Dietetics and American Society for Parenteral and Enteral Nutrition:
characteristics recommended for the identification and documentation of adult malnutrition
(undernutrition). JPEN J Parenter Enteral Nutr, 36, 275-283.
Wikby, K., Ek, A.C. & Christensson, L. (2008) The two-step Mini Nutritional Assessment procedure in
community resident homes. J Clin Nurs, 17, 1211-1218.
Wilhelm, M.C., de Paredes, E.S., Pope, T. & Wanebo, H.J. (1986) The changing mammogram. A primary
indication for needle localization biopsy. Arch Surg, 121, 1311-1314.
Wolinsky, F.D., Coe, R.M., Chavez, M.N., Prendergast, J.M. & Miller, D.K. (1986) Further assessment of
the reliability and validity of a Nutritional Risk Index: analysis of a three-wave panel study of
elderly adults. Health Serv Res, 20, 977-990.
Wolinsky, F.D., Prendergast, J.M., Miller, D.K., Coe, R.M. & Chavez, M.N. (1985) A preliminary
validation of a nutritional risk measure for the elderly. Am J Prev Med, 1, 53-59.
246
Wong, S. & Gandy, J. (2008) An audit to evaluate the effect of staff training on the use of Malnutrition
Universal Screening Tool (Abstract). J Hum Nutr Diet, 21, 405-406.
Wong, S.S., Derry, F., Sherrington, K. & Gonzales, G. (2009) An audit to evaluate the use of nutrition
screening tool in the National Spinal Injury Centre in Stoke Mandeville Hospital,
Buckinghamshire Hospitals NHS Trust (Abstract). Proceedings of the Nutrition Society, 68
(OCE1), E53.
World Health Organisation (1995) Physical status: The use and interpretation of anthropometry. Report of
a WHO Expert Committee. WHO Technical Report Series 854. Geneva.
Wyszynski, D.F., Perman, M. & Crivelli, A. (2003) Prevalence of hospital malnutrition in Argentina:
preliminary results of a population-based study. Nutrition, 19, 115-119.
Young, A.M., Kidston, S., Banks, M.D., Mudge, A.M. & Isenring, E.A. (2013) Malnutrition screening
tools: Comparison against two validated nutrition assessment methods in older medical
inpatients. Nutrition, 29, 101-106.
247
Appendix A
Source: Lim SL, et al. Asia Pac J Clin Nutr 2009; 18:395-403.
248
Appendix B
Source: Detsky AS, et al. J Parenter Enteral Nutr 1987; 11:8-13.
249
Appendix C
7-Point Subjective Global Assessment
R A T I N GS
(circle one rating for each category)
Weight loss ____ kg in the past 6 months
Ratings Weight loss
7 <1kg
6 <3%
5 3-<5%
4 5-<7%
3 7-<10% 7 6 5 4 3 2 1
2 10-<15%
1 15%
If weight trend, add 1 point, if weight trend within 1 month, minus 1 point
Dietary Intake (past 2 weeks)

7) Good (Full share of usual meal)
6) Good (> - 1 share of usual meal)
5) Borderline ( share of usual meal) but increasing
4) Borderline ( share of usual meal), no change or decreasing 7 6 5 4 3 2 1
3) Poor (< share of usual meal) but increasing
2) Poor (< share of usual meal) no change/decreasing
1) Starvation (< of usual meal)
Gastrointestinal symptoms (that persisted for > 2 weeks)

Nausea: _____ Vomiting: ______ Diarrhoea: _______
7) No symptom
6) Very few intermittent symptoms (1x per day) 7 6 5 4 3 2 1
5) Some symptoms (2-3x per day) - improving
4) Some symptoms (2-3x per day) no change
3) Some symptoms (2-3x per day) getting worse
1-2) Some/all symptoms (> 3x per day)
Functional status (nutrition related)

6-7) Full functional capacity
3-5) Mild to moderate loss of stamina 7 6 5 4 3 2 1
1-2) Severe loss of functional ability (bedridden)
Disease state affecting nutritional requirements

6-7) Little or no increase in metabolic demand (no or low stress)
3-5) Mild to moderate increase in metabolic demand (moderate stress) 7 6 5 4 3 2 1
1-2) Drastic increase in metabolic demand (high stress)
Muscle wastage: 6-7) Little or no depletion in all areas

(at least 3 areas) 3-5) Mild to moderate depletion 7 6 5 4 3 2 1
1-2) Severe depletion
Fat stores 6-7) Little or no depletion in all areas

3-5) Mild to moderate depletion 7 6 5 4 3 2 1
Oedema: 6-7) Little or no oedema

(nutrition related) 3-5) Mild to moderate oedema 7 6 5 4 3 2 1
1-2) Severe oedema
Nutritional Status: Well Nourished / Mildly to Moderately Malnourished / Severely Malnourished

Overall SGA Rating: 7 6 5 4 3 2 1
(circle one)
250
Appendix D
Euro Quality of Life5 Domain and Visual Analogue Scale
Mobility
I have no problems in walking about
I have some problems in walking about
I am confined to bed
Self-Care
I have no problems with self-care
I have some problems washing or dressing myself
I am unable to wash or dress myself
Usual Activities (e.g. work, study, housework, family or

leisure activities)
I have no problems with performing my usual activities
I have some problems with performing my usual activities
I am unable to perform my usual activities
Pain/Discomfort
I have no pain or discomfort
I have moderate pain or discomfort
I have extreme pain or discomfort
Anxiety/Depression
I am not anxious or depressed
I am moderately anxious or depressed
I am extremely anxious or depressed
2003 EuroQol Group. EQ-5D is a trade mark of the EuroQol Group
251
Best
imaginable
health state
100
To help people say how good or bad a health

state is, we have drawn a scale (rather like a
thermometer) on which the best state you can 9 0
imagine is marked 100 and the worst state you
can imagine is marked 0.
We would like you to indicate on this scale 8 0

how good or bad your own health is today, in
your opinion. Please do this by drawing a line
from the BLACK BOX below to whichever 7 0
point on the scale indicates how good or bad
your health state is today.
6 0
5 0
4 0
Your own
3 0
health state
today
2 0
1 0
0
Worst
2003 EuroQol Group. EQ-5D is a trade mark of the EuroQol Group imaginable
health state
252
Appendix E
Paper on 7-point Subjective Global Assessment
253
Paper submitted to Clinical Nutrition (not part of PhD research programme)
7-point Subjective Global Assessment is more time sensitive than conventional
Subjective Global Assessment in detecting nutritional changes
Su Lin Lim1,2, Xiang Hui Lin1, Lynne Daniels2

1
Dietetics Department, National University Hospital, Singapore,
2
School of Exercise and Nutrition Sciences, Queensland University of Technology,
Australia
ABSTRACT
Background & aims: It is important for nutritional intervention in malnourished
patients to be guided by accurate evaluation and detection of small changes in the
patients nutritional status over time. However, the current Subjective Global
Assessment (SGA) is not able to detect changes in a short period of time. The aim of
the study was to determine whether 7-point SGA is more time sensitive to nutritional
changes than the conventional SGA.
Methods: In this prospective study, 67 adult inpatients assessed as malnourished
using both the 7-point SGA and conventional SGA were recruited. Each patient
received nutrition intervention and was followed up by a dietitian post-discharge.
Patients were reassessed using both tools at 1, 3 and 5 months from baseline
assessment.
Results: It took significantly shorter time to see a one-point change using 7-point
SGA compared to conventional SGA (median: 1 month vs. 3 months, p = 0.002). The
likelihood of at least a one-point change is 6 times greater in 7-point SGA compared
to conventional SGA (odds ratio = 6.23, 95% CI 2.73-14.2, p<0.001). Changes were
observed in 75% of patients using 7-point SGA vs. 42% using SGA. Fifty-six percent
of patients who had no change in SGA score had changes detected using 7-point
254
SGA. The level of agreement between two blinded assessors for 7-point SGA was
83% (k=0.726, p<0.001).
Conclusions: The 7-point SGA is more time sensitive in its response to nutritional
changes than conventional SGA. It can be used to guide nutritional intervention for
patients.
Keywords:
7-point Subjective Global Assessment; Nutrition status; Malnutrition; Nutritional
changes; Intervention
255
1. Introduction
Malnutrition is prevalent in hospitals and leads to adverse outcomes.1-3 Studies
have shown that patient outcomes can be improved with nutrition support.4,5
Nutritional intervention must be guided by accurate evaluation and detection of small
changes in the patients nutritional status over time. Subjective Global Assessment
(SGA) is a well validated tool widely used to assess nutritional status of patients.6-8 It
involves assigning a rating of A, B or C for each of eight components: weight and
dietary intake changes, gastrointestinal symptoms, functional capacity, metabolic
stress from disease, muscle wasting, fat depletion and nutrition-related edema.6 The
final rating of SGA is a subjective summation of the eight components to classify
patients into three categories; A: well nourished, B: moderately malnourished and C:
severely malnourished.6 Despite widespread use of SGA for nutritional assessment,
very few studies have used this tool to assess changes in nutritional status over time.9
Interventional studies using SGA usually showed no significant change between the
pre and post results.10 An important factor may be the considerable time lag before
changes in nutritional status are detected using SGA score. Given the well-established
association between malnutrition and increased risk of morbidity and mortality,1-3
monitoring changes in nutritional status is vital, especially in patients who have
already been assessed as malnourished or those at risk of further nutritional
deterioration.
The 7-point SGA was developed for use on 680 patients commencing
peritoneal dialysis in the CANUSA Study.11 Similar to SGA, the overall rating of 7-
point SGA is subjective, but the scale is expanded to 7 points where a rating of 1-2
indicates an individual is severely malnourished, 3-5 moderately malnourished and 6-
7 well nourished. Therefore, the categories of nutrition status as assessed by the 7-
256
point scale are consistent with conventional SGA, i.e. well nourished, moderately
malnourished or severely malnourished. The CANUSA study showed that a one unit
lower in the 7-point SGA score was prospectively associated with a 25% increase in
the relative risk of death.11
Since the CANUSA Study, there has been increased use of 7-point SGA,
however this has been limited to renal patients.12-14 No studies have reported on the
use of 7-point SGA in other patient groups. Some authors have speculated that 7-point
SGA may be more sensitive than the conventional SGA in identifying small changes
in renal patients nutritional status;14,15 however this is yet to be confirmed in studies.
Given the broad nature of a 3-point rating in the conventional SGA, a substantial
improvement in nutritional status may be required before patient transitions from a
B (moderately malnourished) to an A (well nourished) rating. In contrast, when
using 7-point SGA a moderately malnourished patient may improve from a rating of 3
to 4. In this instance, the patient is still classified as moderately malnourished, but
smaller changes in nutritional status may be detected. Valid improvements in score
within a broad category would suggest improved nutritional status, and conversely
any deterioration in status can be detected and addressed quickly. To date, no studies
have been published to support this opinion.
The aim of the study was to determine if 7-point SGA is more time sensitive
in its response to nutritional changes than conventional SGA across different patient
diagnostic groups.
257
2. Methods
2.1. Screening and Study Participants
All patients were screened for risk of malnutrition on admission using 3-
Minute Nutrition Screening16,17 by the ward nurses as per hospital protocol. Any
patient identified as at risk of malnutrition was referred to the hospital dietitian, who
confirmed the diagnosis of malnutrition using SGA6 and provided individualized
nutrition intervention and counseling on the ward. Consecutive malnourished adult
patients aged 21 years of age were recruited for the study. Psychiatry patients,
maternity patients, patients on palliative care and patients discharged to a nursing
home or community hospital were excluded from the study. The National Healthcare
Group Domain Specific Review Board approved the study protocol. Informed written
consent was obtained from each participant.
2.2. Baseline assessments
For the purpose of this study, nutritional status was re-assessed using
conventional SGA and 7-point SGA by a study dietitian no more than four days
before the patient was discharged from hospital, and this was considered as baseline
for tracking the nutritional status of patients post-discharge. For better standardization
among assessors, 7-point SGA (Figure 1) was modified from the one used in the
CANUSA Study11 to include a selection of ratings within each component. At the
same sitting, patients body weight, handgrip strength18 and assessment of quality of
life using the Euro Quality of Life - Visual Analogue Scale (EQ-VAS)19 were
measured.
258
Inter-rater agreement between the two study dietitians was conducted on 37
patients using 7-point SGA. The first dietitian assessed each patient using 7-point
SGA, followed by a second dietitian who repeated the 7-point SGA assessment and
was blinded to the results of the first dietitian.
2.3. Follow-up assessments
A total of 67 patients were recruited for this study. Each was provided with
follow-up appointments at an outpatient clinic 1 month, 3 months and 5 months post
discharge from hospital. During these follow-up visits, patients were reassessed using
7-point SGA and conventional SGA. All patients were given individualized nutrition
intervention and counseling as appropriate by the study dietitian. Patients who failed
to turn up for scheduled outpatient appointments were home-visited by the study
dietitian within one week of the missed appointments. At the fifth month follow-up,
assessment using 7-point SGA and conventional SGA was carried out by a second
dietitian who was blinded to the results of the previous ratings. Patients body weight,
handgrip strength and assessment of quality of life using the EQ-VAS were also
measured. The study workflow is presented in Figure 2.
2.4. Statistical analyses
All statistical analyses were performed using the Statistical Package for the
Social Sciences for Windows (version 19.0, SPSS Inc., Chicago, IL, USA) with
statistical significance set at p < 0.05. Pearsons chi-square test was used to compare
the likelihood of detecting a change between 7-point SGA and conventional SGA
presenting the results as odds ratio at 95% confidence intervals (CI). Wilcoxon Signed
Ranks test was performed to determine the time to see a minimum one-point change
in both 7-point SGA and conventional SGA and reported as median. Spearmans rho
259
was used to determine the correlation between changes in both SGAs and changes in
body weight, handgrip strength and EQ-VAS. The inter-rater agreement between the
two assessors using 7-point SGA was reported as % agreement and Kappa statistics.
260
Figure 1: 7-point Subjective Global Assessment (7-point SGA)
R A T I N GS
(circle one rating for each
category)
Weight loss ____ kg in the past 6 months
Ratings Weight loss
7 <1kg
6 <3%
5 3-<5%
4 5-<7%
3 7-<10% 7 6 5 4 3 2
1 2 10-<15%
1 15%
If weight trend, add 1 point, if weight trend within 1 month, minus 1 point
Dietary Intake (past 2 weeks)

7) Good (Full share of usual meal)
6) Good (> - 1 share of usual meal)
5) Borderline ( share of usual meal) but increasing
4) Borderline ( share of usual meal), no change or decreasing 7 6 5 4 3 2 1
3) Poor (< share of usual meal) but increasing
2) Poor (< share of usual meal) no change/decreasing
1) Starvation (< of usual meal)
Gastrointestinal symptoms (that persisted for > 2 weeks)

Nausea: _____ Vomiting: ______ Diarrhoea: _______
7) No symptom
6) Very few intermittent symptoms (1x per day) 7 6 5 4 3 2 1
5) Some symptoms (2-3x per day) - improving
4) Some symptoms (2-3x per day) no change
3) Some symptoms (2-3x per day) getting worse
1-2) Some/all symptoms (> 3x per day)
Functional status (nutrition related)

6-7) Full functional capacity
3-5) Mild to moderate loss of stamina 7 6 5 4 3 2 1
1-2) Severe loss of functional ability (bedridden)
Disease state affecting nutritional requirements

6-7) Little or no increase in metabolic demand (no or low stress)
3-5) Mild to moderate increase in metabolic demand (moderate stress) 7 6 5 4 3 2 1
1-2) Drastic increase in metabolic demand (high stress)
Muscle wastage: 6-7) Little or no depletion in all areas

(at least 3 areas) 3-5) Mild to moderate depletion 7 6 5 4 3 2 1
Fat stores 6-7) Little or no depletion in all areas

3-5) Mild to moderate depletion 7 6 5 4 3 2 1
Oedema: 6-7) Little or no oedema

(nutrition related) 3-5) Mild to moderate oedema 7 6 5 4 3 2 1
1-2) Severe oedema
Nutritional Status: Well Nourished / Mildly to Moderately Malnourished / Severely Malnourished

Overall SGA Rating: 7 6 5 4 3 2 1
(circle one)
261
Figure 2: Study workflow
Screening and Recruitment

Consecutive malnourished patients referred to the dietitians
Inclusion criteria:
21 years old
Malnourished as assessed by Subjective Global Assessment (SGA)
Not Psychiatry, Maternity and Palliative Care patients
Not patients discharged to nursing home or community hospital
Baseline Assessments
(n= 67)
7-point SGA and SGA conducted by the 1st dietitian
Body weight, handgrip strength, Quality of Life Visual Analogue Scale (EQ-VAS)
(all measurements were done at no more than 4 days before discharge from hospital)
Re-assessments
7-point SGA and SGA at 1-month, 3-month and 5-month post baseline
measurements
Body weight, handgrip strength and QoL VAS at 5-month post baseline
measurements
(All 5 month assessments were conducted by a 2nd dietitian, blinded to the
th
results of the previous measurements).
262
3. Results
The demographic profiles of the study subjects are described in Table 1. It took
significantly shorter time to see a one-point change using 7-point SGA compared to
conventional SGA (median: 1 month vs. 3 months, p = 0.002). The likelihood of at least a
one point change is 6 times greater using 7-point SGA compared to conventional SGA (Odds
Ratio: 6.23, 95% CI: 2.73-14.2, p <0.001).
Table 2 shows the frequency of the overall change in SGA score using 7-point SGA
and conventional SGA. Of the 39 patients that had no change in their score using
conventional SGA, 22 patients (56%) had a change in their score within the same nutritional
status category using 7-point SGA.
Table 3 compares the correlation between changes in both the 7-point and
conventional SGA and changes in body weight, handgrip strength and EQ-VAS. There is
moderate positive linear correlation between changes in 7-point SGA and weight change (r=
0.681, p <0.001) and mild positive linear correlation between changes in 7-point SGA and
changes in handgrip strength and EQ-VAS. The correlation between changes in conventional
SGA and weight change is mild (r=0.589, p<0.001) and only weak correlation were found
between changes in conventional SGA and handgrip strength and EQ-VAS.
The level of agreement between two assessors for 7-point SGA was good, at a rate of
83% (k=0.726, p<0.001).
263
Table 1
Demographics of the study subjects at baseline (n= 67)
Variable n (%) Mean SD (Range)

Age (years) 63.9+14.5 (27-87)
Gender
Male 30 (45)
Female 37 (55)
Ethnicity
Chinese 52 (78)
Malay 8 (12)
Indian 4 (6)
Others 3 (4)
1
Baseline Nutritional Status
Moderately malnourished 62 (93)
Severely malnourished 5 (7)
Specialty
General Surgery 19 (28)
General Medicine 15 (22)
Cardiology 10 (15)
Respiratory 5 (8)
Gastroenterology 4 (6)
Oncology 4 (6)
Endocrinology 3 (4)
Geriatrics 3 (4)
Orthopaedic 2 (3)
Nephrology 2 (3)
n= number; SD = standard deviation
1
Severity of malnutrition as defined by Subjective Global Assessment
Table 2
Change in overall nutrition assessment score between baseline and the fifth month using 7-
point SGA and conventional SGA (n=67)
264
Change in score between baseline and 5th 7-point SGA Conventional SGA
month n (%) n (%)
0-point (no change) 17 (25) 39 (58)
1-point 30 (45) 28 (42)
2-point 15 (22) 0 (0)
3-point 5 (8) NA
Total patients with a change in score 50 (75) 28 (42)
SGA = Subjective Global Assessment
NA = Not applicable
265
Table 3
Correlation between changes in 7-point SGA and conventional SGA and changes in body
weight, handgrip strength and quality of life measures using Euro-Quality of Life - Visual
Analogue Scale (EQ-VAS) in 5 months
Changes observed between n Changes 7-point SGA Changes in
baseline and 5th month conventional SGA
Correlation p Correlation p
Weight change 67 0.681a <0.001* 0.589b 0.001*
Changes in handgrip 59 0.346b 0.007* 0.210c 0.111
strength
Changes in quality of life 55 0.369b 0.006* 0.124c 0.366
(EQ-VAS)
SGA = Subjective Global Assessment
*significant p value, aModerate correlation, bMild correlation, cWeak correlation
266
4. Discussion
The current study has shown for the first time that 7-point SGA is able to detect
response to nutritional intervention faster than conventional SGA. It took significantly shorter
time to see a one-point change using 7-point SGA in comparison to conventional SGA. Our
study shows that even though there appears to be no change in classification using
conventional SGA, changes in score within the traditional categories using 7-point SGA were
observed in 56% of these patients. This is important as repeated measures of nutritional
assessment over time yields valuable information that might help guide the nature of the
dietary advice or intervention given. For example, if a patient was decreasing in nutritional
status over time, more aggressive nutritional therapy could be provided. From this current
study, we are able to show that 7-point SGA is a useful nutrition assessment tool in detecting
nutritional changes over relatively short periods of time when compared to conventional
SGA.
It is ironic that although SGA has been a widely validated and well accepted tool to
determine the nutritional status of patients,7,8 it has not been used in many studies to report
changes in nutritional outcomes. Even studies that use SGA initially do not report outcomes
using this tool.5 Instead, changes in body weight are most commonly cited in studies that
span over three months to determine changes in nutritional status of patients.4,20,21 This is
probably due to the limitation of conventional SGA, where it is often not able to detect
change in nutritional status in a shorter period of time even when weight change is present.
However, there are limitations to using weight change to monitor nutritional status, as
changes in body weight may be confounded by the alteration in body composition and fluid
retention commonly associated with illness.22,23 In addition, weight measurements pose
challenges in patients who are bed bound or old and frail. In an audit of 526 hospital
admissions, only 67% of the population had information on weight.24 Even in the clinical
267
research context, there are difficulties in obtaining complete weight and height data.25,26 In
clinical practice, unavailable weight record can be as high as 74-85%.25-27
Malnutrition has been shown to have numerous detrimental effects on health and
quality of life.1-3 To ensure appropriate nutrition care is provided, an in-depth assessment of a
patients nutritional status is needed, and SGA has been developed for this purpose.6
However, once nutrition intervention is implemented, tracking changes in nutritional status is
required to evaluate the effectiveness of the chosen intervention, and to prompt changes in
the treatment plan as required. The benefit of 7-point SGA is that it can potentially detect
comparatively small changes within the broader categories of nutritional status. A study by
Campbell et al. (2007) on patients with chronic kidney disease showed a difference in body
composition between the rating points of 7-point SGA (3, 4 and 5) within the same category
of nutrition status (SGA B).14 Using total body potassium, a gold-standard measure for body
cell mass, a linear increase in mean body cell mass from ratings 3 to 5 in 7-point SGA was
detected. This suggests that nutritional changes took place even though patients would still
have been considered moderately malnourished (rating B) within the broad categories of
conventional SGA.14
The detailed response options in 7-point SGA enable standardized scoring and
objectivity of the assessors within each item in the 7-point SGA. This may partly explain the
good inter-rater reliability of 7-point SGA between dietitians, despite the tool having seven
ratings of nutritional status. Previous studies on SGA have shown inter-rater reliability of
79% and 81%.7,8 The ambiguity in the conventional SGA is addressed in 7-point SGA,
whereby the expanded items in each component are specified clearly (Figure 1), thus
facilitating greater standardization between assessors. The clarity of 7-point SGA is enhanced
by clear instruction that functional status should be nutrition related and not the consequence
268
of a debilitating medical condition such as stroke, and that at least 3 muscle areas need to be
examined.
Similar to conventional SGA, the final rating in 7-point SGA is based on the
subjective weighting of the components to classify patients into 3 categories: well nourished,
moderately malnourished and severely malnourished. Hence, 7-point SGA can always be
converted to conventional SGA rating (but not vice-versa). With this, the prognostic validity
of 7-point SGA remains the same as conventional SGA, which has been shown to have good
prognostic value for a range of clinical outcomes.1,3
The current study shows that changes in 7-point SGA better correlates with changes
in body weight, handgrip strength and quality of life (QoL) measures than conventional SGA.
Body weight, handgrip strength and QoL are commonly used as outcome measures for
nutrition intervention in malnourished patients.4,5 However, as each of these parameters on its
own cannot be used as a sole marker of malnutrition, the ability of 7-point SGA to diagnose
malnutrition as well as to monitor the nutritional progress of patients is notable.
There are several strengths in this study. This is the first study to show that 7-point
SGA can be used to detect nutritional changes faster than conventional SGA. This facilitates
earlier evaluation of the impact of any nutrition intervention, and provides critical guidance to
the healthcare professional in making decisions regarding medical nutrition therapy. Another
strength of this study is the use of a blinded assessor method to test the inter-rater reliability
of 7-point SGA.
In addition, this study was carried out across a broad range of medical conditions and
age groups. In contrast, 7-point scale SGA introduced in the CANUSA study11 has only been
studied in renal patients.12-14 The inclusion of a broad range of medical conditions is
advantageous as patients usually present with multiple comorbidities. Furthermore, it is not
practical to switch from one tool to another for different medical conditions. The aim is to
269
minimize confusion among staff, standardize practice and conserve resources in training staff
when they transfer from one ward to another. This study also showed that it is feasible to use
7-point SGA on patients from all adult age groups. In acute care hospitals, it is not practical
to have different nutrition assessment tools for different adult age groups. For example, Mini
Nutritional Assessment (MNA) has been developed for use on the elderly.28 This tool has
been validated and is frequently used in nursing homes and long term care facilities, and in
these settings has served its purpose well.28-30 Having such a targeted tool may be more
sensitive to the needs of the specific population. However in the hospital setting where there
is a wide age range of patients, it is not practical to carry out MNA on the elderly and use a
separate tool for younger adult patients.
As the same dietitian assessed 7-point SGA during the baseline measurement and at 1
month and 3 months post discharged, observer bias is a limitation of this study. To overcome
this limitation, the fifth month assessment was carried out by a second dietitian blinded to the
previous ratings in the sequential measures of 7-point SGA. Another limitation of our study is
that confounders such as age may have influenced the assessment of nutritional status using
either tool. However, this has been minimized as the trained study dietitians undergo yearly
competency assessments on the use of 7-point SGA and conventional SGA.
270
5. Conclusions
The 7-point SGA is more time sensitive in its response to nutritional changes than
conventional SGA. It can be used in a broad range of medical conditions and adult age
groups to assess and monitor the progress of nutritional status in patients. More importantly,
it can be used to guide nutritional intervention for patients.
Acknowledgements
We are very grateful to Linda McCann for providing advice and initial training on 7-point
Subjective Global Assessment.
Statement of authorship
SLL conceptualized, designed and conducted the study, interpreted the data and wrote
the manuscript. XHL conducted the study, analyzed and interpreted the data and wrote the
manuscript. LD provided significant advice on the design of the study and assisted in writing
the manuscript. All authors have made substantial contributions and approved the final
manuscript.
Conflict of interest
None of the authors had any conflict of interest related to the authorship of the
submitted paper.
Statement of funding
The study was supported by a grant from the Healthcare Quality Initiative and
Innovation Fund (HQI2F).
271
References:
1. Lim SL, Ong KC, Chan YH, Loke WC, Ferguson M, Daniels L. Malnutrition
and its impact on cost of hospitalization, length of stay, readmission and 3-
year mortality. Clin Nutr 2012;31(3):345-50.
2. Norman K, Pichard C, Lochs H, Pirlich M. Prognostic impact of disease-
related malnutrition. Clin Nutr 2008;27(1):5-15.
3. Correia MI, Waitzberg DL. The impact of malnutrition on morbidity,
mortality, length of hospital stay and costs evaluated through a multivariate
model analysis. Clin Nutr 2003;22(3):235-9.
4. Ha L, Hauge T, Spenning AB, Iversen PO. Individual, nutritional support
prevents undernutrition, increases muscle strength and improves QoL among
elderly at nutritional risk hospitalized for acute stroke: a randomized,
controlled trial. Clin Nutr 2010;29(5):567-73.
5. Rufenacht U, Ruhlin M, Wegmann M, Imoberdorf R, Ballmer PE. Nutritional
counseling improves quality of life and nutrient intake in hospitalized
undernourished patients. Nutrition 2010;26(1):53-60.
6. Detsky AS, McLaughlin JR, Baker JP, Johnston N, Whittaker S, Mendelson
RA, et al. What is subjective global assessment of nutritional status? JPEN J
Parenter Enteral Nutr 1987;11(1):8-13.
7. Baker JP, Detsky AS, Wesson DE, Wolman SL, Stewart S, Whitewell J, et al.
Nutritional assessment: a comparison of clinical judgement and objective
measurements. N Engl J Med 1982;306(16):969-72.
8. Hirsch S, de Obaldia N, Petermann M, Rojo P, Barrientos C, Iturriaga H, et al.
Subjective global assessment of nutritional status: further validation. Nutrition
1991;7(1):35-7.
272
9. Campbell KL, Ash S, Davies PS, Bauer JD. Randomized controlled trial of
nutritional counseling on body composition and dietary intake in severe CKD.
Am J Kidney Dis 2008;51(5):748-58.
10. Steiber AL, Handu DJ, Cataline DR, Deighton TR, Weatherspoon LJ. The
impact of nutrition intervention on a reliable morbidity and mortality
indicator: the hemodialysis-prognostic nutrition index. J Ren Nutr
2003;13(3):186-90.
11. Churchill DN, Taylor DW, Keshaviah PR. Adequacy of dialysis and nutrition
in continuous peritoneal dialysis: association with clinical outcomes. Canada-
USA (CANUSA) Peritoneal Dialysis Study Group. J Am Soc Nephrol
1996;7(2):198-207.
12. Steiber A, Leon JB, Secker D, McCarthy M, McCann L, Serra M, et al.
Multicenter study of the validity and reliability of subjective global assessment
in the hemodialysis population. J Ren Nutr 2007;17(5):336-42.
13. Visser R, Dekker FW, Boeschoten EW, Stevens P, Krediet RT. Reliability of
the 7-point subjective global assessment scale in assessing nutritional status of
dialysis patients. Adv Perit Dial 1999;15:222-5.
14. Campbell KL, Ash S, Bauer JD, Davies PS. Evaluation of nutrition assessment
tools compared with body cell mass for the assessment of malnutrition in
chronic kidney disease. J Ren Nutr 2007;17(3):189-95.
15. Jones CH, Wolfenden RC, Wells LM. Is subjective global assessment a
reliable measure of nutritional status in hemodialysis? J Ren Nutr
2004;14(1):26-30.
273
16. Lim SL, Tong CY, Ang E, Lee EJC, Loke WC, Chen Y, et al. Development
and validation of 3-Minute Nutrition Screening (3-MinNS) tool for acute
hospital patients in Singapore. Asia Pac J Clin Nutr 2009;18(3):395-403.
17. Lim SL, Ang E, Foo YL, Ng LY, Tong CY, Ferguson M, et al. Validity and
reliability of nutrition screening administered by nurses. Nutr Clin Pract
2013;28:729-35.
18. Casanova JS. Grip strength. In: Fess EE, Ed. Clinical assessment
recommendations. 2nd edn. Chicago: American Society of Hand Therapists,
1992:41-5.
19. Cheung K, Oemar M, Oppe M, Rabin R. EQ-5D User Guide Basic
Information on how to use EQ-5D. Rotterdam. EuroQOL Group, 2009.
(Accessed January 9, 2012, http://www.euroqol.org/.)
20. Beattie AH, Prach AT, Baxter JP, Pennington CR. A randomised controlled
trial evaluating the use of enteral nutritional supplements postoperatively in
malnourished surgical patients. Gut 2000;46(6):813-8.
21. Ravasco P, Monteiro-Grillo I, Marques Vidal P, Camilo ME. Impact of
nutrition on outcome: a prospective randomized controlled trial in patients
with head and neck cancer undergoing radiotherapy. Head Neck
2005;27(8):659-68.
22. Nightingale JM, Walsh N, Bullock ME, Wicks AC. Three simple methods of
detecting malnutrition on medical wards. J R Soc Med 1996;89(3):144-8.
23. Shirley S, Davis LL, Carlson BW. The relationship between body mass
index/body composition and survival in patients with heart failure. J Am Acad
Nurse Pract 2008;20(6):326-32.
274
24. Campbell SE, Avenell A, Walker AE. Assessment of nutritional status in
hospital inpatients. QJM 2002;95(2):83-7.
25. Wyszynski DF, Perman M, Crivelli A. Prevalence of hospital malnutrition in
Argentina: preliminary results of a population-based study. Nutrition
2003;19(2):115-9.
26. Correia MI, Campos AC. Prevalence of hospital malnutrition in Latin
America: the multicenter ELAN study. Nutrition 2003;19(10):823-5.
27. Waitzberg DL, Caiaffa WT, Correia MI. Hospital malnutrition: the Brazilian
national survey (IBRANUTRI): a study of 4000 patients. Nutrition 2001;17(7-
8):573-80.
28. Guigoz Y. The Mini Nutritional Assessment (MNA) review of the literature--
What does it tell us? J Nutr Health Aging 2006;10(6):466-85.
29. Bleda MJ, Bolibar I, Pares R, Salva A. Reliability of the mini nutritional
assessment (MNA) in institutionalized elderly people. J Nutr Health Aging
2002;6(2):134-7.
30. Donini LM, Savina C, Rosano A, De Felice MR, Tassi L, De Bernardini L, et
al. MNA predictive value in the follow-up of geriatric patients. J Nutr Health
Aging 2003;7(5):282-93.
275
Appendix F
Standard questions and advice given by the dietetics

assistant via telephone follow-up as part of
Ambulatory Nutrition Support Service
Before calling the patient, the dietetics assistant needs to read the
nutritional documentation by the dietitians and understand what have
been advised previously.
Standard Questions:
1. Diet history (Breakfast, lunch, dinner, supper, snacks)

2. Are you consuming the nutrition supplements prescribed by
your dietitian?
3. How is your appetite? Has there been any recent loss of
appetite?
4. Has there been any loss of weight since you were last seen by
the dietitian?
5. Are there any other issues with regards to your diet?
6. Are there any questions on your diet?
Standard Advice:
1. Add 1-2 teaspoons of oil into food (e.g. porridge, soup)

2. Follow the meal plan given by your dietitian
3. Include egg (whole/whites) for breakfast
4. Spread margarine/peanut butter thickly on breads
5. Take nourishing fluids (e.g. fruit juice, milkshake)
6. Add 1-2 teaspoons of margarine into piping hot rice
7. Change cooking methods (pan fry or deep fry instead of
boiling)
8. Supplement with or 1 can or packet of nutrition supplement if
taking only less than 1/2 share meal. (Nutrition supplement
would usually have been prescribed during the last visit by the
dietitian)
276
Appendix G
Research Grant
277
* The above grant was subsequently topped up to S$153,000
278
Appendix H
NHG Domain-Specific Review Board Approval (Singapore)
279
280
281
282
283
Appendix I
QUT Human Research Ethics Committee Approval
284
285
From: QUT Research Ethics Unit

Sent: Friday, August 17, 2012 9:43 AM
To: Su Lin Lim; Lynne Daniels; Maree Ferguson
Cc: Janette Lamb
Subject: Ethics Application Approval 1200000301
Dear Ms Su Lin Lim
Project Title: Nutrition status and clinical outcomes of patients

receiving nutrition support in a tertiary hospital in Singapore
Ethics Category: Human - Low Risk

Approval Number: 1200000301
Approved Until: 17/08/2015 (subject to receipt of satisfactory
progress reports)
We are pleased to advise that your application has been reviewed by the
Chair, University Human Research Ethics Committee (UHREC), and confirmed as
meeting the requirements of the National Statement on Ethical Conduct in
Human Research (2007).
I can therefore confirm that your application is APPROVED.

If you require a formal approval certificate, please respond via reply email
and one will be issued.
CONDITIONS OF APPROVAL
Please ensure you and all other team members read through and understand all
UHREC conditions of approval prior to commencing any data collection:
> Standard: Please go to
www.research.qut.edu.au/ethics/humans/stdconditions.jsp
> Specific: None apply
Decisions related to low risk ethical review are subject to ratification at

the next available UHREC meeting. You will only be contacted again in
relation to this matter if UHREC raises any additional questions or
concerns.
Whilst the data collection of your project has received QUT ethical
clearance, the decision to commence and authority to commence may be
dependent on factors beyond the remit of the QUT ethics review process. For
example, your research may need ethics clearance from other organisations or
permissions from other organisations to access staff. Therefore the proposed
data collection should not commence until you have satisfied these
requirements.
Please don't hesitate to contact us if you have any queries.
We wish you all the best with your research.
Kind regards
Janette Lamb on behalf of the Chair UHREC

Research Ethics Unit | Office of Research | Level 4 88 Musk Avenue,
Kelvin Grove | Queensland University of Technology
p: +61 7 3138 5123 | e: ethicscontact@qut.edu.au | w:
www.research.qut.edu.au/ethics/
286
From: QUT Research Ethics Unit

Sent: Friday, March 22, 2013 2:49 PM
To: Lynne Daniels; Su Lin Lim
Cc: Janette Lamb
Subject: Ethics Application Approval - 1200000602
Dear Ms Su Lin Lim
Project Title: Development of novel approaches to improve the nutritional

status of malnourished patients discharged from hospital and evaluation of
its effectiveness
Ethics Category: Human - Low Risk

Approval Number: 1200000602
Approved Until: 22/03/2016 (subject to receipt of satisfactory
progress reports)
We are pleased to advise that your application has been reviewed by the
Chair, University Human Research Ethics Committee (UHREC) and confirmed as
meeting the requirements of the National Statement on Ethical Conduct in
Human Research (2007).
I can therefore confirm that your application is APPROVED.

If you require a formal approval certificate please respond via reply email
and one will be issued.
CONDITIONS OF APPROVAL
Please ensure you and all other team members read through and understand all
UHREC conditions of approval prior to commencing any data collection:
> Standard: Please see attached or go to
www.research.qut.edu.au/ethics/humans/stdconditions.jsp
> Specific: None apply

Decisions related to low risk ethical review are subject to ratification at
the next available UHREC meeting. You will only be contacted again in
relation to this matter if UHREC raises any additional questions or concerns.
Whilst the data collection of your project has received QUT ethical
clearance, the decision to commence and authority to commence may be
dependent on factors beyond the remit of the QUT ethics review process. For
example, your research may need ethics clearance from other organisations or
permissions from other organisations to access staff. Therefore the proposed
data collection should not commence until you have satisfied these
requirements.
Please don't hesitate to contact us if you have any queries.
We wish you all the best with your research.
Kind regards
Janette Lamb on behalf of the Chair UHREC Research Ethics Unit | Office of
Research | Level 4 88 Musk Avenue, Kelvin Grove | Queensland University
of Technology
p: +61 7 3138 5123 | e: ethicscontact@qut.edu.au |
w:www.research.qut.edu.au/ethics/
287
Appendix J
Statement of Contributions of Co-Authors
288
Statement of Contribution of Co-Authors for
Thesis by Published Paper
The authors listed below have certified* that:

1. they meet the criteria for authorship in that they have participated in the conception,
execution, or interpretation, of at least that part of the publication in their field of expertise;
2. they take public responsibility for their part of the publication, except for the responsible
author who accepts overall responsibility for the publication;
3. there are no other authors of the publication according to these criteria;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or
publisher of journals or other publications, and (c) the head of the responsible academic
unit, and
5. they agree to the use of the publication in the student's thesis and its publication on the
Australasian Research Online database consistent with any limitations set by publisher
requirements.
Publication title and date of publication or status:

Malnutrition and its impact on cost of hospitalisation, length of stay, readmission and 3-year
mortality. Clinical Nutrition 2012; 31 (3):345-350.
Contributor Statement of contribution*

Lim Su Lin
J ~jL)
~
conceptualized, designed and conducted the study, interpreted
the data and wrote the manuscript.
Signature
5 AUj\!isf ~o13
Date
AlProf. Benjamin Ong provided significant advice on the design of the study and input
into manuscript
Dr Chan Yiong Huak performed statistical analysis and interpreted the data
Dr Loke Wai Chiong provided significant advice on the design of the study and input
into manuscript
Dr Maree Ferguson provided significant advice on the design of the study and input
into manuscript
Prof. Lynne Daniels interpreted the data, provided significant advice on the design of
the study and input into manuscript
Principal Supervisor Confirmation

I have sighted email or other correspondence from all Co-authors confirming their certifying
authorship.
Prof. Lynne Daniels

Name
frto7O/S
Da e
289

unit, and
requirements.
Reply - Malnutrition and its impact on cost of hospitalisation, length of stay, readmission and 3-
year mortality. Clinical Nutrition 2013; 32(3):489-490.

Lim Su Lin
~ wrote the reply letter to the editor

Siqnature
5 A ~lAsf.2AJ1J
Date
Prof. Lynne Daniels assisted in writing the reply letter to the editor
authorship.
Prof. Lynne Daniels

Name
290

1. they meet the criteria for authorship in that they have participated in the conception, execution, or
interpretation, of at least that part of the publication in their field of expertise;
2. they take public responsibility for their part of the publication, except for the responsible author who
accepts overall responsibility for the publication;
4. potential conflicts of interest have been disclosed to (a) granting bodies, (b) the editor or publisher of
journals or other publications, and (c) the head of the responsible academic unit, and
5. they agree to the use of the publication in the student's thesis and its publication on the Australasian
Research Online database consistent with any limitations set by publisher requirements.
Development and validation of 3-Minutes Nutrition Screening (3-MinNS) Tool for acute hospital patients in
Singapore. Asia Pacific Joumal of Clinical Nutrition 2009;18(3):395-403.
Lim Su Lin
1~
Signature
conception of the research design and planning of study, data collection,
recruitment of the participants, compilation, analysis and interpretation of
the data, and writing of the manuscript
5 ~'-1JtA rf :to13
Date
Tong Chung Van assisted in the collection of data
Dr Emily Ang involved in the research design and commented on the manuscript
A/Prof. Evan Lee provided significant advice on the design of the study and input into
Jon Choon manuscript
Dr Lake Wai provided significant advice on the design of the study and input into
Chiang manuscript
Chen Yuming assisted in the statistical analysis and interpreted the data
Dr Maree Ferguson provided significant advice on the design of the study, interpretation of
data and input into manuscript
Prof. Lynne Daniels provided significant advice on the design of the study, interpretation of

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.
Prof. Lynne Daniels

Name
sJ1&r.S
ate
291

Validity and reliability of nutrition screening administered by nurses. Nutrition in Clinical Practice 2013
(Accepted)

Um Su Un
conception of the research design and planning of study, data

~ collection, recruitment of the participants, compilation, analysis and
Signature interpretation of the data, and writing of the manuscript
5 AlAJ~Jt ;tot3
Date
Dr Emily Ang involved in the research design and commented on the manuscript
Faa Yet U assisted in the collection of data
Ng Uan Ye assisted in the collection of data
Tong Chung Yan assisted in the collection of data and input into manuscript
Dr Maree Ferguson provided significant advice on the design of the study, interpretation
of data and input into manuscript
Prof. Lynne Daniels provided significant advice on the design of the study, interpretation
of data and input into manuscript

I have sighted email or other correspondence from all Co-authors confirming their certifying authorship.
Prof. Lynne Daniels

Name
~.{(- 81 n
292

Improving the performance of nutrition screening through continuous quality improvement

initiatives. Joint Commission Journal of Quality and Patient Safety 2013. (Peer-reviewed and in
revision)

Um Su Un
t~
Signature
conception of the research design and planning of study, analysis and
interpretation of the data, and writing of the manuscript
5 (\\'\~\J~~~lJ
Date
Ng Sow Chun involved in the research design and commented on the manuscript
Jamie Lye assisted in writing the manuscript

provided significant advice on the design of the study, interpretation of
Dr Loke Wai Chiong
Dr Maree Ferguson provided significant advice on the design of the study, interpretation of
Prof. Lynne Daniels provided significant advice on the design of the study, interpretation of

authorship.
Prof. Lynne Daniels

Name
293

unit, and
requirements.
A pre-post evaluation of an ambulatory nutrition support service for malnourished patients post
hospital discharge: a pilot study. Annals Academy of Medicine Singapore 2013. (In press)

Lim Su Un
conceptualized, designed and conducted the study,

~
Signature interpreted the data and wrote the manuscript.
13 Ptv.&",St 2..J/3
Date
Un Xianghui performed data collection and assisted in writing the
manuscript
Dr Chan Yiong Huak performed statistical analysis and interpreted the data
Dr Maree Ferguson provided significant advice on the design of the study and
assisted in writing the manuscript
Prof. Lynne Daniels provided significant advice on the design of the study and
assisted in writing the manuscript

authorship.
Prof. Lynne Daniels

Name
294

Su Lin Lim Thesis

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Su Lin Lim Thesis

Uploaded by

Copyright:

Available Formats

Malnutrition in hospitalised patients and clinical outcomes: A missed opportunity?

MALNUTRITION IN HOSPITALISED PATIENTS AND

Bachelor of Science in Dietetics (Honours)

submitted for the

School of Exercise and Nutritional Sciences

Queensland University of Technology

In order for malnutrition to be properly addressed, there must be a

The first critical step in managing malnutrition is the identification of malnourished

For a nutrition screening tool to be effective, it must be completed fully and

To complete the process of nutrition screening, any patient identified at screening

The aims of the research programme were to:

quality improvement initiatives in improving the overall performance of

The study programme was conducted in four phases:

Using Subjective Global Assessment (SGA), the prevalence of malnutrition was

In the development of the new nutrition screening tool (3-MinNS), a combination

This research programme is amongst the first to examine the impact of

In conclusion, this research programme has successfully delivered a

Malnutrition, Prevalence, Nutrition Screening, Subjective Global Assessment,

1.2 Nutrition Screening. 60

1.3 Interventions for Patients with Malnutrition. 109

CHAPTER 5: VALIDITY AND RELIABILITY OF 3-MINUTE NUTRITION 156

5.2 Publication: Validity and reliability of nutrition screening administered

CHAPTER 6: IMPROVING THE PERFORMANCE OF NUTRITION

CHAPTER 7: FOLLOW-UP FOR MALNOURISHED POST-

7.2 Publication: A pre-post evaluation of an ambulatory nutrition support

CHAPTER 8: DISCUSSION AND RECOMMENDATIONS...... 200

8.1 Overview of Research Questions and Key Findings 200

Appendix A: 3-Minute Nutrition Screening.. 248

Figure 1.1 Vicious cycle of the development and progression of

3-MinNS 3-Minute Nutrition Screening

NPV Negative Predictive Value

PUBLICATIONS RELATED TO THE RESEARCH

2. Lim SL, Daniels L. Reply - Malnutrition and its impact on cost of

Quantification. Journal of Renal Nutrition 2012; 22(6):558-566. (IF =

9. Lim SL, Lye J. Nutritional Intervention Incorporating Expedited 10g

CONFERENCE ABSTRACTS, POSTERS AND ORAL PRESENTATIONS

and Biomedical Congress; 11-12 Nov 2011: Singapore: Annals of the

6. Lim SL, Ang E, Ng SC, Tong CY, Ferguson M, Daniels L. Reducing

2011 Awarded National Healthcare Quality Improvement Commendation

2010 Awarded a grant (S$153,000) from the Healthcare Quality Improvement

Media Interest Related to the PhD Research

Newspaper interview: More Malnutrition cases detected with NUH

Newspaper interview: NUH sees Success in Tackling Patient with

Newspaper interview: Malnutrition Problems Berita Harian (a Singapore

Health Magazine interview: Malnutrition and Ageing EzyHealth, April

National University Hospital Publication: Beating Malnutrition, LifeLine,

Newspaper interview: Eating Well in Your Autumn Years, TODAY, 20

Agency of Integrated Care Publication: 3-Minute Nutrition Screening

SPEAKER AT VARIOUS PLATFORMS RELATED TO TOPIC OF THESIS

DATE TOPICS AUDIENCE

Nursing Directors from all

4/10/12 3-Minute Nutrition Screening New Nurses in NUH

Signed: Date: 17 January 2014

My gratitude goes also to National University Hospitals Director of Clinical

Finally, I would like to acknowledge National Healthcare Group and subsequently

National University Hospital (NUH), a 987-bed acute tertiary hospital, is one of

Singapore Healthcare Financing System

Singapore emphasises personal responsibility in healthcare financing. In general,

Out-of-pocket payment at the point of consumption Patient have to pay out

inpatient charges when a patient is admitted to hospital. Medisave cannot be

The Singapore government provides subventions to subsidise the healthcare

Private patient (no

Medifund or social assistance is available for patients who cannot afford to

The Singapore Government recognises that intermediate and long-term