International Journal of Nursing Studies 51 (2014) 562571

Contents lists available at ScienceDirect

International Journal of Nursing Studies

journal homepage: www.elsevier.com/ijns

Validity, reliability and utility of the Irish Nursing Minimum Data Set for General Nursing in investigating the effectiveness of nursing interventions in a general nursing setting: A repeated measures design
Roisin Morris a,*, Anne Matthews b, Anne P. Scott b
a b

Health Service Executive, Dr Steevens Hospital, Dublin 8, Ireland School of Nursing and Human Sciences, Dublin City University, Glasnevin, Dublin 9, Ireland

A R T I C L E I N F O

A B S T R A C T

Article history: Received 16 April 2012 Received in revised form 26 July 2013 Accepted 27 July 2013 Keywords: Nursing effectiveness Nursing Minimum Data Set Path analysis Patient outcomes Reliability Validity

Background: Internationally, nursing professionals are coming under increasing pressure to highlight the contribution they make to health care and patient outcomes. Despite this, difculties exist in the provision of quality information aimed at describing nursing work in sufcient detail. The Irish Minimum Data Set for General Nursing is a new nursing data collection system aimed at highlighting the contribution of nursing to patient care. Objectives: The objectives of this study were to investigate the construct validity and internal reliability of the Irish Nursing Minimum Data Set for General Nursing and to assess its usefulness in measuring the mediating effects of nursing interventions on patient well-being for a group of short stay medical and surgical patients. Design: This was a quantitative study using a repeated measures design. Setting: Participants sampled came from both general surgery and general medicine wards in 6 hospitals throughout the Republic of Ireland. Participants: Nurses took on the role of data collectors. Nurses participating in the study were qualied, registered nurses engaged in direct patient care. Because the unit of analysis for this study was the patient day, patient numbers were considered in estimations of sample size requirements. A total of 337 usable Nursing Minimum Data Set booklets were collected. Methods: The construct validity of the tool was established using exploratory factor analysis with a Promax rotation and Maximum Likelihood extraction. Internal reliability was established using the Cronbachs Alpha coefcient. Path analysis was used to assess the mediating effects of nursing interventions on patient well-being. Results: The results of the exploratory factor analysis and path analysis met the criteria for an appropriate model t. All Cronbach Alpha scores were above .7. Conclusion: The overall ndings of the study inferred that the Irish Nursing Minimum Data for General Nursing possessed construct validity and internal reliability. The study results also inferred the potential of the tool in the investigation of the impact of nursing on patient well-being. As such, this new tool demonstrated potential to be used in the provision of quality information to inform policy in relation to the organisation of nursing care. More research is needed to further establish its use in the assessment of patient outcomes. 2013 Elsevier Ltd. All rights reserved.

* Corresponding author at: Health Service Executive, Dr Steevens Hospital, Dublin 8, Ireland. Tel.: +353 872889923. E-mail address: roisin0212@gmail.com (R. Morris). 0020-7489/$ see front matter 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijnurstu.2013.07.011

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571

563

What is already known about the topic?  Internationally, nurses are increasingly being put under pressure to highlight the contribution they make to health care delivery.  Valid and reliable data are required to provide good quality information to describe nursing care in sufcient detail.  The impact of nursing on patient outcomes has been demonstrated using hospital discharge data bases and cross sectional research design. What this paper adds  This paper establishes the validity and reliability of a new data collection tool for the General Nursing profession.  Unlike previous studies which tend to use large generic, hospital discharge data bases and cross sectional research design, this study highlights the potential of a nursing specic research tool and repeated measures data to demonstrate the impact of nursing on patient outcomes.  This paper highlights the potential of the Irish Nursing Minimum Data Set for General Nursing and similar tools in the assessment of nursing effectiveness and patient outcomes. 1. Background 1.1. The need to describe nursing care in sufcient detail Nursing is central to health care and represents a costly resource that should be managed and used in an organised and efcient manner. Internationally, registered nurses are coming under increasing pressure to highlight the important and specic contribution they make to health care and patients outcomes, particularly within the context of changing models of care delivery. For example, in Ireland there is currently a move towards increased community based health care delivery, thereby decreasing reliance on the acute inpatient care system (HSE, 2009). Nurses working in more integrated community based health care teams consisting GPs, social worker, psychologists, occupational therapists and nurses, among other professionals, will need to make more visible their contribution to patient care. In addition to changes to models of health care delivery, potential task-shifting and the introduction of lower skilled employees to carry out elements of the nursing role is likely to challenge the traditional roles of health care workers, including nurses. In the literature and in practice, difculties exist in articulating and describing nursing work in sufcient detail and shortcomings exist in the provision of quality nursing information (Clark and Lang, 1992; Scott et al., 2006; MacNeela et al., 2006; Maben, 2008; Morris et al., 2010). To highlight the unique contribution that nurses make to patient care and indeed to most effectively manage nursing work, it is essential that information regarding the main tenets of the nursing role be made available to key decision-makers. Until very recently, little scientic evidence existed to identify the central

components of nursing care in Ireland. This lack of nursing evidence is a problem reected in international health care settings. 1.2. The Nursing Minimum Data Set The Nursing Minimum Data Set is a minimum set of elements of information with uniform denitions and categories concerning the specic dimensions of nursing, which meets the information needs of multiple data users in the health care system (Werley and Lang, 1988). To date, Nursing Minimum Data Sets have been developed in many countries. While taking different forms internationally, the basic aim of the NMDS is to determine what nurses do and to what effect (Werley and Lang, 1988; Gliddon, 1998; Sermeus et al., 1996, 2005; Goossen et al., 2000; Volrathongchai et al., 2003; MacNeela et al., 2006). A valid and reliable NMDS can have many uses. For example, it can be used to describe the nursing care of individuals, families and communities; to demonstrate or project trends regarding nursing care provided; to allocate nursing resources to patients; and to stimulate research (MacNeela et al., 2006). Finally a valid and reliable NMDS can be used to provide data and information about nursing care to inuence practice, administrative and health policy decision making (Werley and Lang, 1988). 1.3. Nursing effectiveness With increased emphasis on integrated health care, the need to illustrate the effectiveness of the nursing input into the multidiscipliniary team is very relevant today. One way of establishing the effectiveness of the nursing role is through the measurement of nursing sensitive patient outcomes. Nursing sensitive patient outcomes are dened as measurable changes in a patients state of health or condition as a result of nursing interventions and for which nurses are responsible (Maas et al., 1996; Van der Bruggen and Groen, 1999). There are two predominant perspectives on the investigation of nursing sensitive patient outcomes that have been investigated in the literature. The rst involves the investigation of outcomes according to a process model of care whereby outcomes are affected not only by the care provided but also by the factors related to the patient, to the interpersonal aspects of care and to the setting or environment in which care is provided (Irvine et al., 1998, p. 58). The second perspective encompasses nursing sensitive patient outcomes which include the unintended effects of inadequate nursing care such as medication errors, patient falls and nosocomial infections, on patient outcomes (McGillis-Hall, 2004). These patient outcomes are frequently examined according to their relationship to varying levels of nursing education and skill mix as well as the nursing environment (e.g. Needleman et al., 2002; Aiken et al., 2008, 2010, 2012; Rafferty et al., 2007). Lankshear et al. (2005) criticised the cross-sectional nature of nursing outcomes research, stating that longitudinal design would serve to reduce error by virtue of the time factors involved. It seems that much of the more contemporary research into nursing related patient outcomes, carried out since the publication of Lankhsears

564

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571

work, still utilises cross sectional data, captured using large generic databases (e.g. Aiken et al., 2010, 2012). 1.4. The Nursing Role Effectiveness Model The Nursing Role Effective Model (NREM) (Irvine et al., 1998; Doran et al., 2002) is a conceptual model that can be used in the analysis of nursing sensitive patient outcomes. The NREM is based on the idea that outcomes are multifaceted and reective of what precedes them. The NREM accounts for the structure, process, and outcomes of care. Structure refers to the attributes of the settings in which care occurs, process relates to what is actually done in giving and receiving care and outcomes relate to the effects of care on the health status of patients and populations (Donabedian, 1966, 1980). The underlying proposition of the model is that structural variables impact on nurses role performance, which impacts on patient outcome achievement (Doran et al., 2002, 2006). In contrast to other approaches to nurse-sensitive patient outcomes research, this approach serves to account, rather than control for the many factors that contribute to patient state and nursing care (Sidani et al., 2004). Furthermore, use of this theory-driven approach to outcome assessment insists that any outcome is responsive to care provided. In this way, it makes elements of nursing care mediators between initial patient state and patient outcomes of care. Such outcomes can relate to patient health, e.g. physical, psychological, social and behavioural well-being (Sidani et al., 2004; Johnson et al., 2000; Doran et al., 2006) and are examined through the illustration of change in patient state over a caring period.

with. The tool also captures patient demographic information such as age group and diagnosis as well as information on the ward and nurse identication. Preliminary piloting and validation of the I-NMDS (GN) established the validity and the inter-rater reliability of the instrument (Scott et al., 2006). Further research to establish its construct validity was recommended. 2.3. Conceptual framework For the purpose of this study the NREM was adapted to incorporate structure, process and outcomes variables relevant to the I-NMDS (GN) as follows: Structure variables included the physical health of the patient on Day 11 (Factor 1 resulting from the test of the scale construct validity), as well as the patients age group. Problems related to the physical health of the patient included for example bleeding, risk of pressure ulcer, infection and nausea. Process variables included interventions related to the promotion of patient physical comfort (again Factor 1 resulting from the test of the scale construct validity) carried out on Days 2 and 3 of the inpatient stay. Interventions included for example pressure area management, facilitating mobility, wound care and controlling infection. Outcomes variables included the physical health status of the patient on Days 2 and 3 of the inpatient stay. 2.4. Sites Hospitals chosen for this study had to have both general surgery and general medicine wards. In order to achieve geographic representation, a total of 9 hospitals representing the four Health Service Executive administrative areas were asked to participate in the study. These sites were selected according to their inpatient numbers, i.e. the largest hospitals in the Health Service Executive administrative areas were invited to take part in the study. Participants were sampled from surgical, general medical and combined surgical and general medical wards. None of these wards were specialised. 2.5. Sample

2. Methods 2.1. Study design This study was quantitative, employing a repeated measures design. 2.2. Research tool The Irish Nursing Minimum Data Set or General Nursing (I-NMDS GN) is based on a number of nurse-informed studies designed to establish the contribution that general nurses in Ireland make to health care (e.g. Hyde et al., 2005; Butler et al., 2006; MacNeela et al., 2006; Scott et al., 2006). These studies uncovered what the research team believe to be the core elements of general nursing practice in Ireland and include core patient presenting problems and core nursing interventions, including coordination and organisation of care activities. The tool included 33 interventions (including physical, psychological and social interventions and coordination of care activities) and 41 problem variables (including physical, psychological and social interventions). All of these variables are measured on a 5 point Likert scale according to the level of the intervention carried out and the level, or severity, of the problem the patient presented

A convenience sample of qualied, registered nurses engaged in direct patient care was recruited to the study. Because the unit of analysis for this study was the patient day, patient numbers were considered in estimations of sample size requirements. The preferred sample size (of completed I-NMDS booklets, which represented the nursing care and patient problem state over the inpatient stay) for study was based on analysis requirements and was estimated to be 400. This sample size represented an approximate ratio of 10:1 cases to variables per scale. A minimum of 200 cases was deemed acceptable (Hair et al., 2005). Data collected represented nurse perceptions of the care episode.

Note that Day is referred to as D within this paper.

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571

565

2.6. Procedure This study was carried out in conjunction with, albeit independently of, the larger European Union funded RN4CAST: nurse forecasting for Europe study (www.rn4cast.eu). Ethical approval was granted from the academic institution of the researchers and all participating hospitals. Staff nurses and clinical nurse managers involved in direct patient care were recruited to the study to collect data. Before the study commenced, a researcher went out to each site to provide participant with study information and answers to potential questions on the research as a whole. Nurses were assured that the research was both voluntary and condential and that data would be kept in a secure locked area, accessible only to study researchers. The work described here was carried out in accordance with The Code of Ethics of the World Medical Association (Declaration of Helsinki) for experiments/ research involving humans. 2.7. Data collection The nurse data collectors were asked to select patients within their care who they knew would be admitted for a period of 7 days or fewer. This inpatient length of stay period was considered appropriate to capture data relating to short stay patients and was informed/ supported by nursing staff and management engaged in the study. Data collectors were asked to complete the I-NMDS (GN) for their chosen patients from the day they were admitted to the unit to the day of discharge. Researchers stressed that data should be collected for as many patients as possible without compromising nursing work. Whenever possible, the researcher was on site to answer questions relating to data collection. Telephone calls were also made to sites to offer support. When the nurse was not available to complete the entire I-NMDS (GN) booklet for his/her patient (due to work shift and leave arrangements), the nurse who took over the care of the patient completed the tool on his/her behalf. Data collection ran for approximately 46 weeks in each site. 2.8. Analysis All data were examined for skewness, kurtosis, outliers and missing data. A total of 346 I-NMDS tools were collected across the study sites. Of these, 9 were considered to be incomplete, i.e. they were not adequately completed with over 80% of the data missing. Following data cleaning, the sample was reduced to 337 patients, i.e. 337 I-NMDS (GN) booklets. Demographic data were analysed using descriptive statistical tests. Exploratory factor analysis (EFA) with Maximum Likelihood estimation and Promax rotation techniques was carried out. Two separate analyses were performed for the respective interventions and problems scales. Day 2 data was used for this purpose as data from the day of admission was considered less appropriate and Day 2 data consisted the most cases thereafter (n = 297). The data were checked for signicance of correlations, singularity and sampling

adequacy using the KaiserMeyerOlkin Measure of Sampling Adequacy (KMO) value and Bartletts test of sphericity. To determine the number of factors to extract from the data Kaisers criterion and the scree plot were used. Factor loading cut off points of .35 and .4 were used to interpret the ndings of the respective scales (Hair et al., 2005). The internal reliability of each factor was tested using Cronbachs alpha coefcient. The relationships between the structural, process and outcome variables were tested via cross lagged path analysis using the AMOS statistics programme. Parameters were estimated using maximum likelihood extraction. Fit indices used included the Normed X2 goodness of t score, the comparative t index (CFI) and the root mean square error of approximation (RMSEA). In order to determine a good t, the Normed X2 score should be at or below 3:1 (X2:df), the CFI should be greater than .9 and the RMSEA should be below .1 or more preferably .08 (Hu and Bentler, 1999; Hair et al., 2005). Data for all patients with an admission of 4 days or more were used to conduct a path analysis to determine the impact of nursing interventions on patient problems over the rst 3 days of their admission. It was decided to examine the early stage of the patient stay as it was presented most data for analysis and it was anticipated that the effects of nursing interventions at the initial stages of the patient stay would impact most on problem alleviation and improvement in physical well-being. According to Hair et al. (2005), over 5 subjects per parameter are required for adequate path analysis. As there were 28 parameters within the model, a sample size of 140 or more participants was desirable. As the sample size for the 3 day or more admission period was 178, it was considered adequate for path analysis.

3. Results 3.1. Demographic results A total of 6 hospitals took part in the study, representing the four Health Service Executive administrative areas. Approximately 29% of booklets were completed at Hospital 5, 21% at Hospital 1, 15% at Hospital 2, 14% at Hospital 3 and 10% at both the Hospitals 4 and 6. Within the booklets 337 days of data were collected for Day 1 of the study, 300 days for Day 2, 178 for Day 3, 115 days for Day 4, 90 days for Day 5, 72 days for Day 6 and 7 for Day 7. If we consider day 1 returns as 100% of patients taking part in the study, then it can be said that 89% stayed for 2 days, 53% stayed for 3 days, 34% stayed for 4 days, 27% stayed for 5 days, 21% stayed for 6 days and 2% stayed for 7 days. The majority of booklets were completed for male patients, i.e. 59% while 41% of were for females. Approximately 15% of patients were under the age of 40 years while the majority were over the age of 50 (i.e. approximately 75%) (see Table 1). Fifty ve percent were patients in surgical wards, 43% were in medical wards while 2% were in combined general medical wards. Despite these gures, only 90 patients (27%) underwent a surgical procedure during their hospital stay.

566 Table 1 Patient age group. Age group 1619 2029 3039 4049 5059 6069 7079 80+

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571 Table 2 I-NMDS interventions pattern matrix. Percent 2.7 5.4 6.8 9.8 20.2 22.0 17.9 14.9 Interventions D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 pressure area management toileting facilitating mobility feeding attending to hygiene wound care inserting, monitoring, caring for medical devices controlling Infection controlling pain managing mood Providing informal psychological support managing anxiety monitoring psychological condition encouraging adherence to treatment developing/maintaining trust responding to altered thought/ cognition dealing with the persons information needs teaching skills/health promotion facilitating links between the family and MDT facilitating external activities supporting care delivery liaising with MDT members and other nurses supporting the family focused discussion with other nurses monitoring, assessing, evaluating physical condition 1 .94 .91 .86 .77 .69 .63 .50 .40 .31 .92 .87 .82 .80 .76 .75 .54 .49 .46 .95 .80 .79 .62 .55 .48 .86 .81 .49 .34 .31 2 3 4

A total of 37 patients stayed only one day in the ward, 119 had a 2 day admission, 64 had an admission of 3 days, 24 had an admission of 4 days, 19 had an admission of 5 days, 68 had an admission of 6 days and 6 had an admission period of 7 days. We extrapolated that 11% of patients in the study were admitted to a participating ward for one day only, 35% were admitted for 2 days only, 19% were admitted for 3 days, 7% were admitted for 4 days, 5% were admitted for 5 days, approximately 20% were admitted for 6 days and 2% were admitted for 7 days. The diagnosis related data for the sample was coded according to the ICD-10 diagnostic classication system. Diagnostic information relating to 101 cases was missing and incomplete. Of the remaining 236 cases, 16% were diagnosed with diseases of the digestive system, 15% were diagnosed with diseases of the circulatory system, 15% were diagnosed with neoplasms, 14% were diagnosed with diseases of the respiratory system, 11% were diagnosed with diseases of the genitourinary system and 29% were diagnosed with other diseases including, diseases of the musculoskeletal system and connective tissue, diseases of the nervous system and diseases of the endocrine, nutritional and endocrine systems. Missing diagnostic related data may have been due to nurses not having diagnostic codes to hand when completing the tool. Approximately, 73% of participants were discharged to their homes after their hospital stay. Approximately, 16% of patients in the study were discharged to another ward or unit within the hospital while only 2% of participants were discharged to another hospital. Nine percent of participants were discharged to a nursing home, one participant was discharged to prison, one participant was discharged to a shelter and one to a residential home. 3.2. Construct validity results For conceptual and statistical reasons, a four factor model for the interventions scale was accepted (see the pattern matrix in Table 2). This four factor model explained 63% of the variance in the data and was found to be consistent with the widely accepted biopsychosocial model of health care (Engel, 1980). The Normed X2 goodness of t score for this model was an acceptable 2.8 while the Root Mean Square Error of Approximation was a borderline, but acceptable .09 (RMSEA). Generally a Normed X2 score of 3 or less is associated with good tting models while an RMSEA score below .1 is considered acceptable with better tting models producing RMSEA scores below .08 (Hair et al., 2005).

D2 administering med/uid/blood D2 documenting the patients care

The variables Admitting and initial assessment and Planning discharge were not included in this analysis as they were not considered relevant to Day 2 of the patients stay, although one could argue that planning discharge can occur at any time of the inpatient stay. The variable Repositioning the patient was eliminated from analysis when it was found to produce a Haywood case. The variables Advocating and Responding to emergency situations were also eliminated as they loaded across 2 factors of the scale. Finally, while the variable Controlling pain was also found to cross load, it was deemed relevant to clinical practice and was therefore retained. Within the problems scale analysis, a factor loading cut off of .4 was used to determine the nal factor structure of the scale. For conceptual and statistical reasons, a four factor model for the problems scale was accepted (see the pattern matrix in Table 3). This four factor model explained 67% of the variance in the data and was also found to be consistent with the widely accepted biopsychosocial model of health care (Engel, 1980). The Normed X2 goodness of t score for this model was an acceptable 2.7 while the Root Mean Square Error of Approximation was a borderline, but acceptable .09. All indicator variables were eliminated from the analysis, having served their purpose of loading with appropriate variables on sub-scales related to, e.g. physical, psychological and social well being. Indicator variables

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571 Table 3 I-NMDS problems pattern matrix. Patient problems relating to. . . Factor 1 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 D2 bleeding skin integrity risk of pressure ulcer nausea physical mobility medical devices body temperature cardiac rhythm/circulation sleep hygiene breathing infection family coping family knowledge decit social stigma social support social disadvantage delayed discharge appropriateness of care environment Physical comfort pain physical side-effect weakness and fatigue uid balance elimination mood coping/adjustment longstanding anxiety patient knowledge decit regarding illness trust in others level of motivation longstanding mental health problems .95 .89 .79 .73 .71 .66 .66 .65 .61 .58 .58 .51 .88 .87 .87 .86 .83 .80 .59 .98 .74 .73 .65 .43 .42 .87 .77 .77 .70 .67 .66 .63 2 3 4

567

3.4. Path analysis results As per Fig. 1, the following structure process and outcomes variables were highlighted for analysis purposes. The t statistics for this model indicated a relatively good t to the data, i.e. the Normed X2 was 2.8, the CFI was .99 and the RMSEA was .1. While the CFI and the Normed X2 indicated a good model t, the RMSEA indicated a borderline acceptable model t. The regression and standardised regression estimates are outlined in Tables 4 and 5. Signicant, negative regression relationships were observed for the impact of interventions on the day of admission on problems presenting on Day 2 (r = .32, p < .05) and on Day 3 (r = .15, p < .05). This infers that, nursing interventions led to a reduction in the level of the patients presenting problems. A similar observation was made for the impact of interventions on Day 2 on problems presenting on Day3 (r = .24, p < .05). Examination of the comparable standardised regression estimates in Table 5 indicates that nursing interventions on day 1 accounted for the greatest reduction in patient problems over the inpatient stay. 4. Discussion Establishing the validity and reliability of the I-NMDS (GN) was an important development in the current context of Irish and international nursing, given the need for the systematic description of nursing care to increase the visibility of the nursing contribution to care within the multidisciplinary team. Exploratory factor analysis using ML extraction and a PROMAX rotation was favoured for the purpose of establishing the construct validity of the I-NMDS (GN) as it focused on conrmation of the t of the data to the factor model (Fabrigar et al., 1999). The results of the construct validity analysis inferred that the I-NMDS (GN) was valid for use in the clinical setting. The t statistics were within the limitations of what might be considered an acceptable t. Internal reliability results inferred that the variables within each resulting factor were appropriately placed. Research to establish the variables to include in the tool indicated that this factor structure might be appropriate for the tool (Scott et al., 2006). To assess the effectiveness of nursing interventions within the realm of patient care, a path analysis to investigate the mediating effects of physical health oriented nursing interventions on patient outcomes was carried out. The t statistics for the path model used indicated a relatively good t to the data. Nursing interventions were found to signicantly and negatively impact patient problems across the study, an encouraging nding from the perspective of highlighting the impact of nursing work on patient recovery. Signicant, negative regression relationships were observed for the impact of interventions on the day of admission on problems presenting on Day 2 (r = .32, p < .05) and on Day 3 (r = .15, p < .05). This infers that, nursing interventions led to a reduction in the level of the patients presenting problems. A similar observation was made for the impact

were included in the early analysis as a validity cross checking mechanism. Other variables that cross-loaded and deemed acceptable to eliminate from the analysis included Communication, Pre-existing conditions (this data was captured in the background section of the I-NMDS), Current anxiety, Spiritual needs and Nutrition. 3.3. Internal reliability results Factors were named and the internal reliability was estimated. The internal reliability results for the interventions scale factors were as follows: Factor Factor Factor Factor 1 2 3 4 promoting physical Comfort a = .92. psychological support a = .94. coordination of care a = .89. monitoring clinical condition a = .79.

Internal reliability results for the problems scale factors were as follows: Factor Factor Factor Factor 1 2 3 4 physical health status a = .95. patient social circumstance a = .95. physical comfort a = .90. psychological health status a = .94.

568

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571

0, 1

e1
0, 1

Problems on Admission

Interventions on admission

e4

Age group

Problems Day 2

0,

e2

Interventions Day 2

0,

0, 1
Interventions Day 3

e3

e5

Problems Day 3

Fig. 1. Path model.

of interventions on Day 2 on problems presenting on Day3 (r = .24, p < .05). The impact of interventions on problems appeared to reduce over the course of the stay, indicating that the need for nursing care was most acute at the beginning of the patient admission period. As the patient recovered from his/her surgery or medical procedure, problems related to bleeding, wound care, infection, etc. needed less nursing intervention. Almost all surgical and other procedures took place on Day 1 and 2 of the patient stay. As noted, 27% of patients in the surgical wards underwent a surgical procedure during their hospital stay. This may be due to other related but non-surgical procedures being carried out, e.g. for exploratory purposes or in addition to the principal diagnosis (ESRI, 2012). Signicant and same day regression relationships were found between interventions on a given day and the patient problems presenting on that same day, e.g. if interventions on Day X reduced, so too did patient problem levels reduce and vice versa. It is possible to infer that reductions in intervention level were positively related to improvements in the patients problem

presentation as the patient was responding positively to care. This nding needs further exploration to better understand this relationship. The fact that the I-NMDS (GN) was found to be construct valid and internally reliable infers its usability in the future to facilitate nursing policy related decision making in a number of areas. For example, data collected using the I-NMDS (GN) can be easily analysed and graphed to provide information on nursing trends in, e.g. patient populations, diagnosis, nursing interventions and practice across service and geographic boundaries. Such information could be very valuable in facilitating effective health service management in Ireland. The I-NMDS (GN) also has the potential to provide valuable information to inform hospital budgeting, nurse stafng and consequently patient safety in Ireland, a use that is made of the Belgian Nursing Minimum Data Set (BNMDS). In Belgium, the BNMDS is used to provide information on patient needs to inform nurse stafng levels and consequently to ensure patient to nurse ratios are adequate and safe (Sermeus et al., 2005; Van den Heede et al., 2009). Indeed, the

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571 Table 4 Regression estimates. Estimate Interventions on admission Interventions on admission Interventions day 2 Problems day 2 Problems day 2 Problems day 2 Interventions day 3 Interventions day 3 Problems day 3 Problems day 3 Problems day 3 Problems day 3 Problems day 3 *** = <0.001. Problems on admission Age group Interventions on admission Interventions on admission Interventions day 2 Problems on admission Problems day 2 Interventions day 2 Interventions day 3 Problems day 2 Interventions day 2 Interventions on admission Problems on admission .816 .001 .581 .322 .756 .574 .410 .361 .597 .437 .236 .151 .330 Standard error .044 .021 .068 .078 .046 .073 .080 .083 .050 .067 .064 .067 .070 Critical ratio 18.712 .047 8.493 4.150 16.440 7.923 5.097 4.367 11.948 6.565 3.703 2.244 4.700

569

P value *** .962 *** *** *** *** *** *** *** *** *** .025 ***

research team involved in the development of the I-NMDS (GN) drew on the work of the BNMDS researchers in developing the tool. Unlike, the BNMDS, the Irish tool includes patient problem variables to allow for the analysis of patient problem proles as they are perceived by the nurse and that impact on the nurses work. Keeping up with workforce demands and the changing nature of health service provision both internationally and at home is imperative to ensuring a quality nurse education system. Data collection using the I-NMDS (GN) could allow for the study of change in problem severity and related nursing activity across diagnoses, specialties, wards and units, local and regional geographic boundaries and time. This kind of research could provide valuable information to educators and policy makers in manpower planning and skills training for the future development of nursing in Ireland. Further to this, integration of the I-NMDS (GN) into the Electronic Patient Record has potential to greatly facilitate the access to nursing information to facilitate decision making and increase the efciency of nursing care. Because the I-NMDS (GN) specically relates to nursing related patient problems and interventions, it is appropriate to use in the assessment of nursing role effectiveness and can compliment similar research using cross sectional analysis. In many of these studies outcomes tend to be measured using large generic databases and data is used retrospectively (e.g. Aiken et al., 2002, 2003, 2008; Rafferty
Table 5 Standardised regression coefcients.

et al., 2007). As this study ran in conjunction with the EU funded RN4CAST nurse forecasting in Europe study (www.rn4cast.eu), future research will consider comparisons and the merits of using cross sectional and/or repeated measures data in outcomes analysis. Finally, it is acknowledged that the work of the multidisciplinary team plays a coordinated role in patient care and patient recovery and that this study simply focuses on the work of the nurse through the investigation of the nurses perspective of patient care and nurse sensitive patient outcomes. 4.1. Limitations  Problems recorded in I-NMDS (GN) are not systematically screened and recordings are judgement based and therefore subjective. While it is possible that nurse experience may impact on responses relating to patient problem and intervention ratings, preliminary interrater reliability of the tool was established (Scott et al., 2006).  Further validation of the section of the tool relating to patient diagnosis is required to establish why there was a high level of missing data and to address this problem.  At the time of data collection for this study there was some industrial unrest in the nursing profession in Ireland. For example, nurses in some hospitals engaged in an unofcial work to rule. This may have impacted

Estimate Interventions on admission Interventions on admission Interventions day 2 Problems day 2 Problems day 2 Problems day 2 Interventions day 3 Interventions day 3 Problems day 3 Problems day 3 Problems day 3 Problems day 3 Problems day 3 Problems on admission Age group Interventions on admission Interventions on admission Interventions day 2 Problems on admission Problems day 2 Interventions day 2 Interventions day 3 Problems day 2 Interventions day 2 Interventions on admission Problems on admission .821 .002 .541 .292 .734 .523 .429 .368 .590 .452 .237 .142 .311

570

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571

participation rates in the study. Higher participation rates would have been preferred. As with much research of this kind, the Hawthorne Effect may have been at play, therefore impacting on the participant responses to reect what they think they should be doing rather than what they are doing in reality. There is a possibility that participants completed the tool for patients in their care that they perceived as most convenient therefore introducing bias into the study. This should be considered in more detail in any future validation of the tool. Further research need to be carried out before results can be generalised beyond general medical and surgical wards (including mixed general and medical wards) within large acute hospitals in Ireland. The path model engaged data collected for patients with a hospital stay of 3 days, due to, among other things, sample size and data availability. While it might be argued that this was a short length of stay for investigative purposes, the results of the study suggest that the nurses work can be seen to positively impact on patients with a relatively short length of stay. Future research should investigate the results of path analysis on a sample of patients with a longer length of stay to further establish nursing role effectiveness using this methodology. More research into the validity and reliability of the INMDS (GN) is warranted to further establish its clinical utility. More research into the area of nursing role effectiveness using similar models and methodologies as those outlined herein are required to improve current understandings regarding what aspects of nursing care are most crucial to patient recovery. All of this has the potential to more rmly establish the importance of the nursing role within the system of health care delivery both in Ireland and internationally.

Contributors The authors would like to acknowledge the contribution of the Directors of Nursing, clinical nurse managers who enabled this research to take place in the study hospitals and the staff nurses who collected the data. Dr Marcia Kirwan, Ms. Daniela Lehwaldt and Mr. Sean Duffy facilitated data collection for this study. The original INMDS (GN) tool was developed as part of an Irish Health Research Board funded study on nursing decision making in Ireland led by Professor Anne Scott, Presidents Ofce, draig MacNeela, Department Dublin City University, Dr. Pa of Psychology, NUI Galway, Professor Pearl Treacy, School of Nursing and Health Systems, University College Dublin and Dr. Abbey Hyde, School of Nursing and Health Systems, University College Dublin. Author contributions Drs. Roisin Morris and Anne Matthews helped in research design, data collection, accessing the sites, analysis, conceptual underpinnings of paper and paper preparation. Professor Anne Scott contributed in accessing the sites, research design, conceptual underpinnings of paper, paper preparation. Conict of interest: There are no conicts on interest. Funding: This study was carried out with the support of the Irish Nurses and Midwives Organisation (INMO). The INMO had no input into the study design; in the collection, analysis and interpretation of data; in the writing of the study report; and in the decision to submit the paper for publication. Ethical approval: Approved by the university and by all hospitals taking part in the study. References
Aiken, L., Sermeus, W., Van den Heede, K., Sloane, D.M., Busse, R., McKee, M., Bruyneel, L., Rafferty, A.M., Grifths, P., 2012. Patient safety, satisfaction, and quality of hospital care: cross sectional surveys of nurses and patients in 12 countries in Europe and the United States. British Medical Journal early online version 114. Aiken, L., Sloane, D.M., Cimiotti, J.P., Clarke, S.P., Flynn, L., Seago, J.A., Spetz, J., Smith, H.S., 2010. Implications of the California Nurse stafng mandate for other states. Health Services Research 45 (4) 904921. Aiken, L., Clarke, S., Sloane, D., Lake, E., Cheney, T., 2008. Effects of hospital care environment on patient mortality and nurse outcomes. Journal of Nursing Administration 38 (5) 223229. Aiken, L.H., Clarke, S.P., Sloane, D.M., Sochalski, J., Silber, J.H., 2002. Hospital nurse stafng and patient mortality, nurse burnout, and job dissatisfaction. Journal of American Medical Association 288, 19871993. Aiken, L., Clarke, S., Cheung, R., Sloane, D., Silber, H., 2003. Educational levels of hospital nurses and surgical patient mortality. Journal of American Medical Association 12, 16171623. Butler, M., Treacy, M.P., Scott, A., Hyde, A., MacNeela, P., Byrne, A., Drennan, J., Hyde, A., Irving, K., 2006. Towards a nursing minimum data set: making the key elements of nursing visible. Journal of Advanced Nursing 55 (3) 364375. Clark, J., Lang, N.M., 1992. Nursings next advance: an international classication for nursing practice. International Nursing Review 39, 109112. Donabedian, A., 1966. Evaluating the quality of medical care. Milbank Memorial Fund Quarterly 44 (3) 166206. Donabedian, A., 1980. Exploration in Quality Assessment and Monitoring: The Denition of Quality and Approaches to Its Assessment. Health Administration Press, Ann Arbor, MI.

5. Conclusion We consider this study to be both topical and timely. In Ireland, like many countries throughout the world, health systems reorganisation and efciency attainment are demanding that nursing make very clear its contribution to health care delivery and patient recovery. Further to this, policies such as task shifting are threatening the prominent role nursing play within health services internationally and in some instances are leading to the appointment of less qualied and less costly health professionals in the place of nurses. Research suggests that policies like these threaten patient safety and that better educated nursing workforce with good patient to nurse ratios within the clinical setting produce better patient outcomes (e.g. Aiken et al., 2012). The research reported herein compliments such research and adds to our understanding of how nursing impacts patient wellbeing. This study is but a starting point from which other similar research can be implemented to highlight the effectiveness of the nursing role using a general nursing specic NMDS tool, repeated measures data and path analysis.

R. Morris et al. / International Journal of Nursing Studies 51 (2014) 562571 Doran, D., Sidani, S., Keatings, M., Doidge, D., 2002. An empirical test of the nursing role effectiveness model. Journal of Advanced Nursing 38, 2939. Doran, D.M., Harrison, M., Spence-Laschinger, H., Hirdes, J., Rukholm, E., Sidani, S., McGillis-Hall, L., Tourangeau, A., Cranley, L., 2006. The relationship between nursing interventions and outcome achievement in acute care and long-term care. Research in Nursing and Health 29, 6170. Economic and Social Research Institute, 2012. Activity in Acute Public Hospitals, 2011 Annual Report. ESRI, Dublin. Engel, G.L., 1980. The clinical application of the biopsychosocial model. American Journal of Psychiatry 137, 535544. Fabrigar, L.R., Wegener, D.T., MacCallum, R.C., Strahan, E.J., 1999. Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods 43, 272299. Gliddon, T., 1998. The home and community care HACC minimum data set. ACCNS Journal for Community Nurses 3, 14. Goossen, W., Epping, P., Van Den Heuvel, W., Feuth, T., Fredericks, C., Hasman, A., 2000. Development of the nursing minimum data set for the Netherlands NMDSN: identication of categories and items. Journal of Advanced Nursing 313, 536547. Hair, J., Black, B., Babin, B., Anderson, R.E., Tatham, R.L., 2005. Multivariate Data Analysis, 6th ed. Prentice Hall, Upper Saddle, NJ. Hu, L., Bentler, P.M., 1999. Cut off criteria for t indexes in covariance structure analysis: conventional criteria versus new alternatives. Structural Equation Modeling 6, 155. Hyde, A., Scott, P.A., Treacy, M.P., Mac Neela, P., Irving, K., Hanrahan, M., Butler, M., 2005. Modes of rationality in nursing documentation: Biology, biography and the marginal voice of nursing. Nursing Inquiry 12, 6677. Health Service Executive, 2009. Transformation Programme, 20072010. Health Service Executive, Kildare. Irvine, D., Sidani, S., McGillis-Hall, L., 1998. Linking outcomes to nurses roles in health care. Nursing Economics 16, 5864. Johnson, M., Maas, M.L., Moorhead, S., 2000. Nursing Outcomes Classication. Mosby, St. Louis, MO. Lankshear, A., Sheldon, T., Maynard, A., 2005. Nurse stafng and healthcare outcomes: a systematic review of the international research evidence. Advances in Nursing Science 282, 163174. Maas, M.L., Johnson, M., Moorhead, S., 1996. Classifying nursing-sensitive patient outcomes. Journal of Nursing Scholarship 28 (4) 295299. Maben, J., 2008. The art of caring: invisible and subordinated? A response to Juliet Corbin: Is caring a lost art in nursing?. International Journal of Nursing Studies 45, 335338.

571

MacNeela, P., Scott, P.A., Treacy, M., Hyde, A., 2006. Nursing minimum data sets: A conceptual analysis and review. Nursing Inquiry 131, 4451. McGillis-Hall, L., 2004. Nursing staff mix models and outcomes. Journal of Advanced Nursing 44, 217226. Morris, R., MacNeela, P., Scott, A., Treacy, P., Hyde, A., Matthews, A., Morrison, T., Byrne, A., 2010. The Irish nursing minimum data set for mental health a valid and reliable tool for the collection of standardized nursing data. Journal of Clinical Nursing 19, 359367. Needleman, J., Buerhaus, P., Mattke, S., Stewart, M., Zelevinsky, K., 2002. Nurse-stafng levels and quality of care in hospitals. New England Journal of Medicine 346, 14151422. Rafferty, A., Clarke, S., Coles, J., Ball, J., James, P., McKee, M., Aiken, L., 2007. Outcomes of variation in hospital nurse stafng in English hospitals: cross-sectional analysis of survey data and discharge records. International Journal of Nursing Studies 44 (2) 175182. Scott, A., Treacy, M.P., MacNeela, P., Hyde, A., Morris, R., Drennan, J., Byrne, A., Henry, P., Butler, M., Clinton, G., Corbally, M., Irving, K., 2006. Report on the Delphi study of Irish Nurses to Articulate the Core Elements of Nursing Care in Ireland. Dublin City University, Dublin. Sermeus, W., Delesie, L., 1996. RIDIT analysis on ordinal data. Western Journal of Nursing Research 18, 351359. Sermeus, W., Van den Heede, K., Michiels, D., Delesie, L., Thonon, O., Boven, C., Codognotto, J., Gillet, P., 2005. Revising the Belgian nursing minimum dataset: from concept to implementation. International Journal of Medical Informatics 74 (11) 946951. Sidani, S., Doran, D.M., Mitchell, P.H., 2004. A theory-driven approach to evaluating quality of nursing care. Journal of Nursing Scholarship 36 (1) 6065. Van den Heede, K., Michiels, D., Thonon, O., Sermeus, W., 2009. Using nursing interventions classication as a framework to revise the Belgian nursing minimum data set. International Journal of Nursing Terminologies and Classications 20 (3) 122131. Van der Bruggen, H., Groen, M., 1999. Toward an unequivocal denition and classication of patient outcomes. Nursing Diagnosis 10 (3) 93102. Volrathongchai, K., Delaney, C., Phuphaibul, R., 2003. Nursing minimum data set development and implementation in Thailand. Journal of Advanced Nursing 43, 588594. Werley, H., Lang, N., 1988. Identication of the Nursing Minimum Data Set. Springer Publishing, New York.