You are on page 1of 5

JAMDA xxx (2016) 1.e1e1.

e5

JAMDA
journal homepage: www.jamda.com

Original Study

Enhanced Oral Care and Health Outcomes Among Nursing Facility


Residents: Analysis Using the National Long-Term Care Database in
Japan
Sachiko Ono DDS, MPH a, *, Miho Ishimaru DDS, MPH a, Hayato Yamana MD, MPH a,
Kojiro Morita RN, MPH a, Yosuke Ono MD b, Hiroki Matsui PT, MPH a,
Hideo Yasunaga MD, PhD a
a
Department of Clinical Epidemiology and Health Economics, School of Public Health, University of Tokyo, Tokyo, Japan
b
Department of General Medicine, National Defense Medical College, Tokorozawa, Saitama, Japan

a b s t r a c t

Keywords: Background and objective: Although oral care may have salutary effects among frail elderly people, access
Oral care to dental care is often limited in long-term care facilities. In 2009, the Japanese long-term care insurance
long-term care system introduced an additional reimbursement scheme for enhanced oral care supervised by dentists in
nursing home
nursing facilities. The aim of this study was to examine whether enhanced oral care provided by trained
nursing facility staff members is sufcient to improve health outcomes among nursing facility residents.
Design, setting, and participants: This was a quasi-experimental study using a nationwide long-term care
database. Using facility-level propensity score matching, we identied 170,874 residents in 742 facilities
that provided enhanced oral care and 167,546 residents in 742 control facilities that provided only
standard care from 2009 to 2012. We used a resident-level difference-in-differences approach to analyze
the impact of enhanced oral care on health outcomes among nursing facility residents.
Results: After controlling for resident characteristics and background time trends, no signicant differ-
ences were found between residents admitted to the facilities with and without enhanced oral care in
the incidence of critical illness, transfer to a hospital, mortality, or costs. Yearly change in the odds of
discharge to home was signicantly increased for residents with enhanced oral care (odds ratio 1.07;
95% condence interval: 1.02e1.12; P .008).
Conclusion: The results suggest that enhanced oral care provided by trained nursing facility staff mem-
bers may improve the general condition of elderly residents in nursing facilities and promote their
discharge to home.
2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.

Existing studies have suggested that oral health is linked to general However, in long-term care facilities, the implementation of
health and well-being throughout the life course.1 Enhanced oral care enhanced oral care has been challenging because of limited access to
is a multicomponent intervention to improve oral hygiene and feeding dental services and a lack of staff training.1 In the United States,
function. Attention has been paid to enhanced oral care because of its approximately 75% of institutionalized elderly people had poor oral
feasibility and potential salutary effects on systemic conditions. The hygiene.1 Another study revealed that 90.2% of facility staff members
salutary effects of enhanced oral care include the prevention of recognized the importance of oral care, but one-third hesitated to
pneumonia,2e7 glycemic control,8 and nutritional improvement provide it.13
among frail elderly people.9e12 One possible solution is collaborating with dentists to provide the
necessary training and education to enable existing facility staff
members to deliver daily oral care.14 In 2009, the Japanese universal
This work was supported by the Ministry of Health,Labour and Welfare, Japan. public insurance system for long-term care introduced an additional
The authors declare no conicts of interest. reimbursement scheme for enhanced oral care (ie, the maintenance of
* Address correspondence to Sachiko Ono, DDS, MPH, Department of Clinical oral hygiene and oral functional training) managed by dentists in
Epidemiology and Health Economics, School of Public Health, University of Tokyo,
nursing facilities. In this management system, dentists provide oral care
7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
E-mail address: sachikoono-tky@umin.ac.jp (S. Ono). planning and technical instruction to facility staff members, who are

http://dx.doi.org/10.1016/j.jamda.2016.11.024
1525-8610/ 2016 AMDA e The Society for Post-Acute and Long-Term Care Medicine.
1.e2 S. Ono et al. / JAMDA xxx (2016) 1.e1e1.e5

Table 1
Characteristics of Nursing Facilities With and Without Enhanced Oral Care in the Unmatched and Propensity ScoreeMatched Populations

Unmatched Propensity ScoreeMatched

Facilities Without Facilities With Standardized Facilities Without Facilities With Standardized
Enhanced Oral Care Enhanced Oral Care Difference Enhanced Oral Care Enhanced Oral Care Difference

Number of institutions 2003 746 742 742


Type of facility
Public 87 (4.3) 19 (2.5) 0.099 26 (3.5) 19 (2.6) 0.052
Private 1759 (87.8) 669 (89.7) 0.060 655 (88.3) 665 (89.6) 0.041
Unknown 157 (7.8) 58 (7.8) <0.001 61 (8.2) 58 (7.8) 0.015
Physical restraint used 8 (0.4) 2 (0.3) 0.023 0 (0.0) 2 (0.3) 0.074
Private room 124 (6.2) 42 (5.6) 0.024 39 (5.3) 42 (5.7) 0.018
Understaffed facility 11 (0.5) 4 (0.5) 0.002 2 (0.3) 4 (0.5) 0.042
Municipality quintile for average income
Very high 39 (1.9) 27 (3.6) 0.104 26 (3.5) 27 (3.6) 0.005
High 176 (8.8) 124 (16.6) 0.236 118 (15.9) 120 (16.2) 0.008
Medium 51 (2.5) 28 (3.8) 0.074 31 (4.2) 28 (3.8) 0.020
Low 209 (10.4) 84 (11.3) 0.029 101 (13.6) 84 (11.3) 0.070
Very low 1528 (76.3) 483 (64.7) 0.256 466 (62.8) 483 (65.1) 0.048

Data are presented as n (%).

then responsible for administering direct oral care. However, it remains discharged to their homes after functional recovery. Japanese nursing
uncertain whether this model of indirect enhanced oral care improves facilities also provide terminal care. Each of these facilities is staffed
health outcomes among nursing facility residents. A single-center study with at least 1 full-time equivalent physician, who provides daily
conducted in 2013 reported a decreased incidence of pneumonia medical care for residents with chronic conditions and unexpected
among residents in a nursing facility after the introduction of the new acute illnesses.
reimbursement scheme for enhanced oral care.15 However, the causal
relationship between enhanced oral care and improved health out- Introduction of Enhanced Oral Care in Nursing Facilities
comes remains unclear because this previous study compared pre- and
post-intervention outcomes without a control group. In 2009, the Japanese long-term care insurance system introduced
We conducted a quasi-experimental study using a nationwide an additional reimbursement scheme for enhanced oral care, which
database of long-term care insurance claims to analyze the impact of includes oral functional training and the maintenance of oral hygiene
enhanced oral care on health outcomes among elderly nursing facility in nursing facilities. This scheme aims to preserve or improve oral
residents. The outcomes included incidence of critical conditions, function, with dentists providing oral care planning for nursing fa-
transfer to a hospital, mortality, discharge to home, and costs of long- cilities, as well as technical instruction for facility staff members, who
term care. then administer daily direct oral care to the individual residents in
their facilities. Dentists are also required to educate facility staff
Material and Methods members about the assessment of oral health, the proper materials for
oral care, risk management in oral care, and other facility-specic is-
Long-Term Care and Nursing Facilities in Japan sues related to oral health. The details of the provided instruction and
education are decided at the discretion of individual dentists. The
In 2000, Japan introduced a publicly provided long-term care in- reimbursement cost of this new model of providing oral care is
surance. Under this system, nursing facilities are used by elderly approximately one-quarter of that for direct oral care provided by
people with dysfunctional symptoms who are expected to be dentists.

Table 2
Resident Characteristics at Facilities With and Without Enhanced Oral Care in the Unmatched and Propensity ScoreeMatched Populations

Unmatched Propensity ScoreeMatched

Facilities Without Facilities With Standardized Facilities Without Facilities With Standardized
Enhanced Oral Care Enhanced Oral Care Difference Enhanced Oral Care Enhanced Oral Care Difference

N 420,814 171,965 167,546 170,874


Female 286,910 (68.2) 117,201 (68.2) 0.001 114,077 (68.1) 116,456 (68.2) 0.001
Age, y 84.50 (8.38) 84.28 (8.47) 0.025 84.27 (8.49) 84.28 (8.47) 0.001
Care need level
5 (most dependent) 98,059 (23.3) 37,910 (22.0) 0.031 38,430 (22.9) 37,666 (22.0) 0.022
4 128,104 (30.4) 53,218 (30.9) 0.011 51,120 (30.5) 52,860 (30.9) 0.009
3 106,626 (25.3) 44,105 (25.6) 0.007 42,606 (25.4) 43,808 (25.6) 0.005
2 61,168 (14.5) 25,729 (15.0) 0.014 24,688 (14.7) 25,588 (15.0) 0.008
1 26,829 (6.4) 10,984 (6.4) <0.001 10,690 (6.4) 10,933 (6.4) <0.001
0 (least dependent) 28 (0.0) 19 (0.0) <0.001 12 (0.0) 19 (0.0) <0.001
Dementia 107,425 (25.5) 55,989 (32.6) 0.155 46,795 (27.9) 55,465 (32.5) 0.099
Nutritional management 382,822 (91.0) 164,742 (95.8) 0.195 153,921 (91.9) 163,658 (95.8) 0.163
Tube feeding 2275 (0.5) 1144 (0.7) 0.016 826 (0.5) 1140 (0.7) 0.023
Dysphagia
Mild 821 (0.2) 702 (0.4) 0.037 307 (0.2) 702 (0.4) 0.037
Severe 12,601 (3.0) 7977 (4.6) 0.084 5207 (3.1) 7863 (4.6) 0.078
Special diet 124,418 (29.6) 58,423 (34.0) 0.095 49,384 (29.5) 58,038 (34.0) 0.097

Data are presented as mean (standard deviation) for age, and n (%) for female, care need level, dementia, nutritional management, tube feeding, dysphagia, and special diet.
S. Ono et al. / JAMDA xxx (2016) 1.e1e1.e5 1.e3

Data Source and Variables

(2818e2827)
(22.9e23.6)

(12.0e12.0)
(1.1e1.1)

(4.4e4.4)
3rd Year We used administrative claim data for long-term care services
1.1 from April 2006 to March 2012. These data were collected by the

4.4

23.6

12.0

2823
Ministry of Health, Labour and Welfare in Japan as a part of the Survey
of Long-term Care Benet Expenditures.16 This national data set

(2793e2802)
(23.1e23.8)

(12.8e12.8)
contains information on all long-term care service use, age, sex, care
(1.3e1.3)

(4.4e4.4)
2nd Year

need level (ranging from 0 to 5, with lower values indicating less


1.3

4.4

23.8

12.8

2798
dependent conditions), the contents of specic services provided,

All models were adjusted for age, sex, care need level, dementia, dysphagia, nutritional management, tube feeding, special diet, and cumulative years of nursing facility stay at each time point.
After Introduction

discharge destination, and death. The data also contain basic infor-
mation on the nursing facilities, including the municipalitys average

(2785e2793)
(23.4e23.6)

(14.3e14.3)
income level (by quintile), room type (ward or private room), use of
(1.1e1.1)

(4.6e4.6)
1st Year

physical restraint, and care personnel stafng. We dened under-


1.1

4.6

23.6

14.3

2789
staffed facilities as those where (1) the number of personnel per 100
residents was less than 1 for physicians, rehabilitation staff members,
(2674e2682)

dietitians, or care managers; (2) the number of pharmacists per 300


(23.6e23.6)

(15.0e15.0)

residents was less than 1; or (3) the number of nurses per 3 residents
(1.5e1.5)

(4.7e4.7)
1st Year

was less than 1.


1.5

4.7

23.6

15.0

2678

Data on the following clinical conditions of the nursing facility


residents were available: dysphagia (none, mild, or severe), dementia,
(2595e2605)

tube feeding, terminal care, nutritional management, and special diet


(23.5e24.2)

(17.0e17.0)
Matched Control Facilities

(1.7e1.7)

(4.7e4.7)

(eg, diabetic meals). We used the following data recorded in the


2nd Year

database as the outcome variables: (1) critical conditions treated in


1.7

4.7

24.2

17.0

2600
Before Introduction

the nursing facility; (2) death; (3) transfer to a hospital; and (4)
discharge to home. Critical conditions included impaired conscious-
(2492e2504)

ness or coma, acute respiratory failure, or acute exacerbation of


(22.4e22.2)

(18.5e18.5)
(1.4e1.4)

(4.0e4.0)
3rd Year

chronic respiratory failure, acute cardiac failure, shock, serious


1.4

4.0

22.2

18.5

2498

metabolic disorder, and other critical conditions (eg, drug poisoning).


We also assessed individual monthly long-term care costs. Data on the
Risk-Adjusted Outcomes Before and After the Introduction of Enhanced Oral Care Compared With Matched Control Facilities

type of facility (private or public) were obtained from the Survey of


(2833e2843)
(22.9e22.9)

(14.2e14.2)

Long-term Care Facilities.


(1.0e1.0)

(5.0e5.0)
3rd Year

1.0

5.0

22.9

14.2

2838

Long-Term Care Facility Selection

We identied all nursing facilities operating continuously from


(2817e2826)
(23.1e23.1)

(13.9e13.9)

April 2006 to March 2012. We dened those facilities that introduced


(1.1e1.1)

(4.8e4.8)
2nd Year

enhanced oral care during the period from May 2009 to March 2013 as
1.1

4.8

23.1

13.9

2822

the group offering enhanced oral care. The control group was dened
After Introduction

as those facilities that did not implement enhanced oral care during
(2798e2807)

the study period.


(23.4e23.4)

(14.8e14.8)

We excluded (1) facilities that introduced enhanced oral care


(0.9e1.0)

(5.0e5.0)
1st Year

immediately after the new reimbursement started in April 2009,


0.9

5.0

23.4

14.8

2803

because they may have started providing enhanced oral care before
April 2009; (2) facilities that had no residents with critical conditions
(2677e2685)

in 2006, 2007, or 2008; and (3) facilities that introduced enhanced oral
(23.6e23.6)

(15.5e15.5)
(1.3e1.3)

(4.7e4.7)

care but discontinued this care during the study period.


1st Year

1.3

4.7

23.6

15.5

2681

Statistical Analysis
(2600e2609)
Facilities With Enhanced Oral Care

(23.5e23.5)

(17.4e17.4)

To test whether enhanced oral care improved health outcomes, we


(1.3e1.3)

(4.7e4.8)
2nd Year

used a difference-in-differences approach, an econometric method for


1.3

4.7

23.5

17.4

2605

evaluating changes in outcomes occurring after the implementation of


a new policy. This approach isolates the improvement in outcomes
related to the intervention (ie, the introduction of enhanced oral care)
Before Introduction

2497 (2491e2503)
22.4 (22.4e22.4)

19.0 (19.0e19.0)

from background changes in outcomes that occur during the same


1.3 (1.3e1.3)

3.9 (3.9e3.9)

period and are unrelated to the intervention.17,18


To adjust for differences in the characteristics of the nursing fa-
3rd Year

cilities, we performed one-to-one propensity score matching between


the groups with and without enhanced oral care. Propensity scores
were estimated using a logistic regression model with the following
conditions, %

independent variables: adjusted incidence rate of critical conditions in


Risk-Adjusted

costs, mean
Transfer to a

Discharge to
hospital, %

US dollars
Mortality, %

2006, 2007, and 2008; municipality average income quintile; type of


monthly
Individual
home, %
(95% CI)

(95% CI)

(95% CI)

(95% CI)

(95% CI)
Outcome

Critical

facility (private or public); use of physical restraint; room type (ward


Table 3

or private room); and care personnel stafng. A caliper width of 0.2


times the standard deviation was used, and matching was completed
1.e4 S. Ono et al. / JAMDA xxx (2016) 1.e1e1.e5

Table 4 analyses were performed adjusting for clustering within the facilities
Yearly Changes in Critical Conditions, Transfer to Hospital, Mortality, and Discharge using clustered standard errors. Statistical analyses were conducted
to Home Compared Between Matched Groups With and Without Enhanced Oral Care
using Stata/MP version 14.0 (StataCorp, College Station, TX). All P
Through Difference-In-Differences Analyses
values were 2-sided, and P < .05 was considered to be signicant.
Pre-trend (Yearly Post-trend (Yearly
Change) Change)
Results
OR (95% CI) P OR (95% CI) P

Critical conditions 0.98 (0.91e1.05) .516 1.03 (0.94e1.13) .548 We identied 3243 nursing facilities in operation throughout the
Transfer to a hospital 1.00 (0.98e1.02) .999 0.99 (0.96e1.01) .362 study period. Of these facilities, 913, 179, and 110 introduced enhanced
Mortality 1.02 (0.95e1.10) .517 1.02 (0.94e1.12) .570 oral care in the scal years (AprileMarch) 2009, 2010, and 2011,
Discharge to home 1.01 (0.98e1.04) .507 1.07 (1.02e1.12) .008
respectively. We excluded 411 facilities that introduced enhanced oral
All models were adjusted for patient characteristics, the year that enhanced oral care care in April 2009 and 31 facilities that discontinued enhanced oral
was introduced in each facility, and the years before and after enhanced oral care was
care during the study period. We also excluded 22 facilities that had
introduced in each facility.
no residents with critical conditions in any of the scal years 2006,
2007, or 2008. Propensity score matching within 2749 facilities (746
without replacement. We used standardized differences to compare facilities with enhanced oral care and 2003 facilities without
facility and resident characteristics before and after matching. A enhanced oral care) yielded 742 facility pairs.
standardized difference of less than 0.1 was considered to be indica- Table 1 shows the characteristics of the nursing facilities before
tive of balance. After performing propensity score matching, we and after the propensity score matching. Nursing facilities with
described the adjusted outcomes in the rst, second, and third years enhanced oral care had more residents and were located in munici-
before and after the introduction of enhanced oral care. We used these palities with higher average income levels. After propensity score
relative time points instead of scal years to describe the time trends matching was performed, all of the variables were balanced. In the
of these outcomes, because the timing of the introduction of enhanced resident-level comparison, residents in nursing facilities with
oral care varied among the facilities. For each control facility without enhanced oral care were more likely to have dementia and to receive
enhanced oral care, the index year for the relative time points was set nutritional management than were those in facilities without
as the year during which enhanced oral care was introduced in this enhanced oral care (Table 2). These differences persisted after facility-
facilitys matched counterpart. level propensity score matching was performed.
For the difference-in-differences analysis, resident-level logistic Although there were yearly trends toward higher long-term care
regression models were used to assess the relationships between the costs, improved mortality, and decreased discharge to home in the fa-
outcomes and the provision of enhanced oral care. Residents who cilities with enhanced oral care, similar trends were also observed in
were admitted to the nursing facility before April 2006 and those those facilities without enhanced oral care (Table 3). In the difference-in-
receiving terminal care were excluded from the analyses. Long-term differences analyses, yearly change in the odds of discharge to home was
care costs were assessed by tting a linear regression model. The signicantly higher in the facilities with enhanced oral care compared
following covariates were included in the models: age, sex, care need with those without this care (odds ratio [OR] 1.07, 95% condence
level, dysphagia, dementia, tube feeding, nutritional management, interval [CI]: 1.02e1.12, P .008), whereas yearly changes in critical
special diet, and cumulative years of stay at the nursing facility at each conditions, transfer to a hospital, and mortality did not differ signicantly
time point. The year during which enhanced oral care was introduced between these groups (Table 4). The difference of change in long-term
in each facility and the years before and after the introduction of care costs between the groups also did not reach statistical signicance
enhanced oral care in each facility were included in the models as (4.9 US dollars, 95% CI: 0.8, 10.5; P .092) (Table 5). In all of the an-
independent variables to adjust for background time trends. Interac- alyses, the interaction terms (enhanced oral care multiplied by the years
tion terms between enhanced oral care and the years after enhanced before enhanced oral care was introduced in each facility) were insig-
oral care was introduced in each facility were included to assess the nicant, indicating that the parallel trends assumption was met.
associations of enhanced oral care with the outcomes and long-term
care costs. The parallel trends assumption (ie, that the trends in out- Discussion
comes between the groups with and without enhanced oral care are
the same before the introduction of enhanced oral care) was assessed To our knowledge, this is the largest study examining the effect of
by examining the signicance of the interaction terms between enhanced oral care managed by dentists on health outcomes among
enhanced oral care and the years before enhanced oral care was nursing facility residents. The difference-in-differences analyses
introduced in each facility. All of the difference-in-differences showed a signicant increase in discharge to home in the facilities
with enhanced oral care compared with those without enhanced oral
care, after adjusting for multiple factors including background time
trends. Differences in yearly changes in critical conditions, mortality,
Table 5
Yearly Change in Individual Long-term Care Costs per Month Compared Between
transfer to a hospital, and long-term care costs did not differ signi-
Matched Groups With and Without Enhanced Oral Care Through Difference-In- cantly between the facilities with and without enhanced oral care. Our
Differences Analysis results showed that enhanced oral care managed by dentists, even
Pre-trend (Yearly Post-trend (Yearly
without providing direct professional care, was associated with
Change) Change) discharge to home.
Previous studies have indicated that in addition to improving oral
Coefcient P Coefcient P
(95% CI) (95% CI) health, enhanced oral care could provide several health benets,
including preventing pneumonia and malnutrition by improving
Individual long-term 3.7 (9.4 to 1.9) .197 4.9 (0.8 to 10.5) .092
care costs per month feeding function and oral hygiene in frail elderly people.1e7,10,19
(US dollars) However, the oral care regimens used in previous studies with posi-
The model was adjusted for patient characteristics, the year that enhanced oral care
tive ndings were too resource-consuming to implement as standard
was introduced in each facility, and the years before and after enhanced oral care care in long-term care facilities, where access to dental services is
was introduced in each facility. often limited. Our study indicated that indirect intervention by
S. Ono et al. / JAMDA xxx (2016) 1.e1e1.e5 1.e5

dentistsda less resource-consuming solutiondmay be effective in establishing a sustainable long-term care system in any country with
promoting health benets for nursing facility residents in terms of an aging population. Oral care training systems for long-term care
promoting discharge to home. facility staff members may be a feasible measure to improve outcomes
In the present study, we were unable to specify the underlying for residents.
mechanisms explaining the increased proportion of nursing facility
residents who were discharged to home in the group with enhanced References
oral care. However, one possible explanation for the increased discharge
to home is that enhanced oral care, especially functional training, may 1. Grifn SO, Jones JA, Brunson D, et al. Burden of oral disease among older adults
have enabled oral intake and subsequently prevented malnutrition.9,20 and implications for public health priorities. Am J Public Health 2012;102:
411e418.
Previous studies have shown that approximately half of elderly people 2. Marik PE, Kaplan D. Aspiration pneumonia and dysphagia in the elderly. Chest
had dysphagia and were at increased risk for malnutrition,9,12 which 2003;124:328e336.
caused a loss of muscle mass and function and resulted in further 3. El-Solh AA. Association between pneumonia and oral care in nursing home
residents. Lung 2011;189:173e180.
frailty.11 Malnutrition has also been shown to affect the masticatory and
4. Yoneyama T, Yoshida M, Ohrui T, et al. Oral care reduces pneumonia in older
swallowing muscles, and untreated dysphagia has therefore been found patients in nursing homes. J Am Geriatr Soc 2002;50:430.
to be self-reinforcing.11 In the present study, the implementation of 5. Pace CC, McCullough GH. The association between oral microorganisms and
aspiration pneumonia in the institutionalized elderly: Review and recom-
enhanced oral care may have broken this vicious cycle and improved
mendations. Dysphagia 2010;25:307e322.
systemic conditions in individuals with dysphagia. 6. Juthani-Mehta M, De Rekeneire N, Allore H, et al. Modiable risk factors for
However, the associations between enhanced oral care and critical pneumonia requiring hospitalization of community-dwelling older adults: The
conditions, transfer to a hospital, and mortality were insignicant. Health, Aging, and Body Composition Study. J Am Geriatr Soc 2013;61:
1111e1118.
Pneumonia is a common infectious disease with high mortality and 7. Sjgren P, Nilsson E, Forsell M, et al. A systematic review of the preventive
morbidity in nursing homes.21,22 Past work has shown oral care to be effect of oral hygiene on pneumonia and respiratory tract infection in elderly
effective in preventing aspiration pneumonia by decreasing oral mi- people in hospitals and nursing homes: Effect estimates and methodological
quality of randomized controlled trials. J Am Geriatr Soc 2008;56:2124e2130.
croorganisms and improving swallowing function.2e8 Nonetheless, 8. Simpson TC, Needleman I, Wild SH, et al. Treatment of periodontal disease for
the insignicant associations between enhanced oral care and these glycaemic control in people with diabetes. Cochrane Database Syst Rev;
outcomes suggest that the indirect enhanced oral care in the present 2010:CD004714.
9. Wirth R, Dziewas R, Beck AM, et al. Oropharyngeal dysphagia in older per-
study may be insufcient to affect the incidence of critical illness, sonsdFrom pathophysiology to adequate intervention: A review and summary
including severe pneumonia, among nursing facility residents. of an international expert meeting. Clin Interv Aging 2016;11:189e208.
We acknowledge several limitations to our study. One is the use of 10. Kikutani T, Enomoto R, Tamura F, et al. Effects of oral functional training for
nutritional improvement in Japanese older people requiring long-term care.
administrative claim data lacking information on detailed resident
Gerontology 2006;23:93e98.
characteristics and outcomes. To mitigate this weakness of the data, 11. Clav P, Shaker R. Dysphagia: Current reality and scope of the problem. Nat Rev
we used propensity score matching to adjust for observed facility Gastroenterol Hepatol 2015;12:259e270.
12. Valentini L, Schindler K, Schlaffer R, et al. The rst nutritionDay in nursing
characteristics and the difference-in-differences approach to adjust
homes: Participation may improve malnutrition awareness. Clin Nutr 2009;28:
for unobserved differences between the 2 groups that did not change 109e116.
over time. The data also lack detailed information on oral care, such as 13. Chiba Y, Shimoyama K, Suzuki Y. Recognition and behaviour of caregiver
the proportion of residents who actually received oral care and the managers related to oral care in the community. Gerodontology 2009;26:
112e121.
oral care measures used in each facility. The measures of enhanced 14. Jones JA, Wehler CJ. The Elders Oral Health Summit: Introduction and rec-
oral care may have differed among the facilities, because the oral care ommendations. J Dent Educ 2005;69:957e960.
strategies were decided at the discretion of individual dentists. This 15. Kawada Y, Kawasaki S, Kudou A, et al. The effect of oral care on aspiration
pneumonia. J Japanese Soc Dent Medically Compromised Patient 2013;22:
lack of standardization may have affected the extent of change in the 91e96.
outcomes. Nevertheless, we focused on the net effect of the indirect 16. Ministry of Health, Labour and Welfare. Survey of long-term care benet ex-
intervention by dentists rather than the effect of each oral care mea- penditures, FY2012. Tokyo: Ministry of Health, Labour and Welfare; 2012.
17. Dimick JB, Ryan AM. Methods for evaluating changes in health care policy: The
sure on health outcomes. We were also unable to conrm whether difference-in-differences approach. JAMA 2014;312:2401e2402.
there were any additional therapeutic interventions made by dentists 18. Ryan AM, Burgess JF, Dimick JB. Why we should not be indifferent to specica-
in the facilities with enhanced oral care. tion choices for difference-in-differences. Health Serv Res 2015;50:1211e1235.
19. Watando A, Ebihara S, Ebihara T, et al. Daily oral care and cough reex
sensitivity in elderly nursing home patients. Chest 2004;126:1066e1070.
Conclusions 20. Turner MR, Talbot K. Swallowing and oropharyngeal dysphagia. Clin Med
(Lond) 2014;14:456.
21. El Solh AA. Nursing home-acquired pneumonia. Semin Respir Crit Care Med
In conclusion, enhanced oral care, even without direct intervention
2009;30:16e25.
from dentists, was associated with discharge to home among nursing 22. Nicolle LE, Strausbaugh LJ, Garibaldi RA. Infections and antibiotic resistance in
facility residents. This study has important implications for nursing homes. Clin Microbiol Rev 1996;9:1e17.