Content uploaded by Takashi Naruse
Author content
All content in this area was uploaded by Takashi Naruse on Apr 28, 2021
Content may be subject to copyright.
www.biosciencetrends.com
BioScience Trends. 2015; 9(4):270-274.270
Diseases that precede disability among latter-stage elderly
individuals in Japan
Takashi Naruse*, Mahiro Sakai, Hiroshige Matsumoto, Satoko Nagata
Department of Community Health Nursing, Graduate School of Medicine, The University of Tokyo, Tokyo, Japan.
*Address correspondence to:
Dr. Takashi Naruse, Department of Community Health nursing,
Graduate School of Medicine, the University of Tokyo, 7-3-1,
Hongo, Bunkyo-ku, Tokyo 113-0033, Japan.
E-mail: takanaruse-tky@umin.ac.jp
1. Introduction
Understanding the cause of disability and factors that
precede it among elderly individuals is important to
ensure both prevention of disability and cost saving in
care for disabled persons. In Japan, disabled elderly
individuals are supported by the long-term care insurance
system (1), which covers long-term care services
including home visits and institutional care service.
In March 2014, about 53% of disabled persons who
received long-term care services were mildly disabled
(care levels 1-3 in the Japanese long-term care insurance
system), while 47% were severely disabled (care level 4
or 5).
Cerebrovascular disorder (CVD), including stroke,
has been reported as the most common disease among
disabled persons. Adamson and colleagues found that
the effect of CVD on disability was greater than that of
other chronic diseases among UK adults (2). In the US,
stroke is a leading cause of disability (3). Further, in a
nationwide self-reported survey, over 33.0% of disabled
people recognized that the direct cause of their disability
was CVD (4).
However, in Newman and Brach's (5) review, arthritis
was reported as the greatest cause of disability among
elderly individuals because it limits their everyday
activities. While the impact of arthritis on disability
severity might be smaller than that of CVD, from a
public health perspective, the benefit of preventing mild
disability among elderly individuals might be comparable
Summary Understanding causes of disability among elderly individuals is an important public health
issue, particularly because of the increasing rate of disabled elderly individuals and the
social costs in a rapidly aging society. Accordingly, we aimed to describe the diseases that
precede disability and investigate the types of diseases that are related to severe disability
among Japanese elderly individuals aged over 75 years. Using claim data from the latter-
stage elderly healthcare system and long-term care insurance system, we identified 76,265
elderly individuals over 75 years old who did not qualify as disabled on April 1, 2011.
Among them, 3,715 elderly individuals who had been newly qualified as disabled between
April 1, 2011 and March 31, 2012 were selected. Disease codes from the medical claim data
in the 6 months prior to disability were collected. All descriptions were developed separately
for six groups divided by gender and disability level (low, middle, and high). The results
of the ordinal logistic analysis including sex and age revealed that men tended to have
significantly higher levels of disability (β = 0.417, p < 0.001) than women. Cerebrovascular
disorder (CVD) was the most common disease in almost all age and disability level groups.
In low-level disability groups, cancer in men (12.8%) and arthropathy and fracture in
women (11.9% and 13.5%, respectively) were as common as cerebrovascular disorder
(12.2% and 9.7%, in men and women, respectively). Stroke was the most common disease
for all genders and disability levels. The diseases preceding low-level disability differed by
gender. This study demonstrated the need to consider arthropathy and fracture as well as
CVD in order to prevent disability.
Keywords: Disability, disease, potential risk factors for disability, aging, long-term care
DOI: 10.5582/bst.2015.01059
Brief Report
www.biosciencetrends.com
BioScience Trends. 2015; 9(4):270-274.
to that related to preventing CVD.
Objective data collection instruments are preferred
in measuring disability; however, the majority of studies
rely on self-report measures of disability (5). This might
lead to problems of reliability in measuring disability
and detecting disabled elderly individuals. In Japan,
claim data from the long-term care insurance system
can provide objective information on who is disabled
among elderly individuals, when s/he became disabled,
and how severely s/he is disabled. Further, other claim
data from the healthcare system (latter-stage elderly
healthcare system) can provide objective information
about disease among disabled elderly individuals. If we
merge the two types of claim data, we can determine the
diagnosed disease that preceded disability. This study
aimed to describe the kinds of diseases that preceded
disability and investigate the types of diseases that are
related to severe disability among Japanese elderly
individuals. Because of data availability, we targeted
individuals aged at least 75 years.
2. Methods
2.1. Definition of disability
We referred to the qualification of the long-term care
insurance system for service use as a disability index.
It ranged in terms of support level from care level 1
to 5, which indicate increasing difficulty and need for
assistance with activities of daily living. Care level 1
indicates "having difficulty in independent walking or
daily activity, but not chair bound," care levels 2 and 3
mean "almost chair bound, but not bedridden," and care
levels 4 and 5 indicate "almost or completely bedridden."
For ease of understanding, care levels 1, 2/3, and 4/5
were categorized as low, middle, and high disability,
respectively.
The certification was based on the clients' or
their families' proposal to local governments, and
the certification level was determined based on two
perspectives: the standardized statistical algorithm for
estimating the amount of care requirements and a local
committee of specialists (i.e. physicians, public health
nurses, social workers, and so on).
Elderly individuals were considered "newly disabled"
when they had not been certified for more than 2 months,
and then certification was detected in the next month
based on receipt data. In the Japanese certification
system, disabled individuals rarely distinguish their
certification in 1 month due to any systematic errors. We
consulted with two specialists on a local committee, and
this algorithm was considered valid to detect disabled
elderly individuals.
2.2. Design and data collection
We used claim data from the latter-stage elderly
healthcare system and the long-term care insurance
system in Fukui prefecture, Japan. The data from the
latter-stage elderly healthcare system was provided
by the Fukui Latter-Stage Elderly Healthcare System
Association, and the data from the long-term care
insurance system was provided by the Fukui National
Health Insurance Organization. Two types of claim data
were managed by the different organizations; however,
we can merge them and identify the same individuals
with a common ID. Using the merged data, we can
determine each person's medical and long-term care
insurance service consumption volume, disability level,
disease code, and region for each month.
This study was conducted under a large collaborative
study called the Fukui Gerontology Study of Fukui
prefecture and Institute of Gerontology, The University
of Tokyo. In this study, Fukui prefecture, the Fukui
National Health Insurance Organization, the Fukui
Latter-Stage Elderly Healthcare System Association, and
the University of Tokyo collaborated from April 2011
to March 2015. All data were provided for researchers
in the form of anonymous electric data. The Ethics
Committee of the Graduate School of Medicine at the
University of Tokyo approved this study.
2.3. Subjects
First, elderly individuals aged 75 years or older who
were insured by the Fukui latter-stage elderly healthcare
system from April 1, 2011 through March 31, 2012
were detected. On April 1, 102,450 elderly individuals
were observed. Between April 1 and March 31, 1,627
elderly individuals (1.6%) dropped out after moving
or for other unknown reasons, and 6,411 (6.4%) had
deceased. We excluded 1,627 individuals who dropped
out, leaving a total of 100,823 elderly individuals.
Among them, 24,558 were already disabled with low-
(5,920; 24.1%), middle- (10,166; 41.4%), and high-
level disabilities (8,392; 34.1%), and 80 (0.3%) had
been certified as low- or middle-level disability in
February or March, but not in April 2011. Thus, there
were 76,265 elderly individuals that remained non-
disabled on April 1, 2011. We followed up their claim
data and detected newly disabled persons between April
2011 and March 2012 (n = 3,715), whose data were
then analyzed.
2.4. Definition of the disease preceding disability
For newly disabled elderly individuals, we observed the
disease code in the latter-stage elderly healthcare system
claim data over the 6 months preceding disability.
The latter-stage elderly healthcare system claim data
includes five separate types of medical consumption
data for elderly individuals: admission to hospital,
outpatient, dentistry, pharmacy, and home-visiting
nursing. We included all types of data and investigated
271
www.biosciencetrends.com
BioScience Trends. 2015; 9(4):270-274.272
were quite different from the cross-sectional prevalence
of disability level in April 2011: 24.1%, 41.4%, and
34.1% for low-, middle-, and high-level disability,
respectively. This difference suggests that about half of
latter-stage elderly individuals have low-level disabilities
initially, but then ultimately progress to higher stages of
disability.
For men, there were 617 (39.8%) individuals with
low-level disability and 327 (21.1%) with high-level
disability. For women, 1,088 (50.2%) individuals had
low-level disability, and 354 (16.3%) had high-level
disability. The results of the ordinal logistic analysis
including sex and age revealed that men tended to have
significantly higher levels of disability (β = 0.417, p <
0.001) than women. These results can be understood
from two perspectives. First, men tend to be disabled
more severely initially, or second, men tend to avoid
filing their qualification for long-term care insurance
until their disability worsens. The former is supported
by the disease distribution presented in Table 2. Ordinal
logistic analysis including age and disease showed
that there were no sex differences in the relationship
between disease and disability severity. In both men and
women, cancer, CVD, fracture, pneumonia significantly
related to higher levels of disability, and arthropathy
and dementia significantly related to lower levels
of disability. In men, CVD and cancer are the most
common diseases at all disease levels (12.2%, 15.9%,
and 27.8% for CVD and 12.8%, 18.8%, 16.8% for
cancer for low, middle, and high levels, respectively).
On the other hand, arthropathy and fracture were more
common in low- and middle-level disabled women than
they were in men (11.9% vs. 7.1% and 10.9% vs. 6.0%
for arthropathy among low- and middle-level disability,
respectively; 8.2% vs. 4.7% and 17.2% vs. 6.3% for
fracture among low- and middle-level disability,
respectively). With regard to diseases that precede
disability, these results imply that men tend to have
a higher number of diseases that are associated with
severe disability, than women. This may be because of
the age-specific rate of stroke is higher in men generally
(6), and thus, men could be considered to have a greater
risk for more severe disability.
the diseases that were coded in the 6 months preceding
disability.
We focused on nine types of diseases as predisposing
factors for disability: (1) cancer (C00-97 in ICD-10);
(2) CVD (I60, 61, 63, 69.0, 69.1, 69.3); (3) arthropathy
(M15-19); (4) fracture (S02, S12, S22, S32, S42, S52,
S62, S72, S82, S92 T02, T08, T10, T12); (5) pneumonia
J12-18); (6) chronic obstructive pulmonary disease
(COPD, J41-44); (7) dementia (F01, F03, G30); (8)
psychiatric disorder (F20-48); and (9) neurological
disorder (G00-29, G31-99). If one or more corresponding
code for each of these nine diseases was detected in 6
months, the person was considered to have the disease.
Other types of individual information, including age at
certified month and sex, were collected from receipt data.
2.5. Analysis
First, a t-test compared the mean age of certified month
between men and women. Second, we separately
summarized percentages of each disease according
to sex and disability level (low, middle, high). To
determine the age-adjusted differences of disability
level among men and women, multivariate ordinal
logistic analysis for disability level (1 = low, 2 =
middle, 3 = high) was conducted. Lastly, in order to
investigate the types of diseases associated with severe
disability, multivariate ordinal logistic regression
analysis for disability level was conducted for men and
women separately, with age and nine types of diseases
included as independent variables. SPSS version 22.0
for Windows was used for all analysis, and p < 0.05
was defined as significant.
3. Results and Discussion
The mean age of certified month was 84.2 years (standard
deviation (SD) = 5.1; range: 75-103). Men were younger
at first certified age (83.5 (SD = 4.8) in men and 84.8
(SD = 5.1) in women, p < 0.001). In Table 1, among all
newly disabled elderly individuals, low-level disability
was present in 1,705 (45.9%) participants, and high-level
disability was present in 681 (18.3%). These percentages
Table 1. Frequency of disease and disability level (n = 3,715)
Items
Total
Cancer
CVD
Arthropathy
Fracture
Pneumonia
COPD
Dementia
Psychiatric
Neurological disorder
Total
3,715 (100.0%)
410 (11.0%)
507 (13.6%)
306 (8.2%)
343 (9.2%)
111 (3.0%)
68 (1.8%)
284 (7.6%)
158 (4.3%)
208 (5.6%)
Values are presented as n (%); CVD, cerebral vascular disorder; COPD, chronic obstructive pulmonary disease.
Low
1,705 (45.9%)
138 (8.1%)
180 (10.6%)
174 (10.2%)
118 (6.9%)
31 (1.8%)
28 (1.6%)
207 (12.1%)
92 (5.4%)
103 (6.0%)
Middle
1,329 (35.8%)
179 (13.5%)
159 (12.0%)
102 (7.7%)
142 (10.7%)
51 (3.8%)
28 (2.1%)
63 (4.7%)
49 (3.7%)
76 (5.7%)
High
681 (18.3%)
93 (13.7%)
168 (24.7%)
30 (4.4%)
83 (12.2%)
29 (4.3%)
12 (1.8%)
14 (2.1%)
17 (2.5%)
29 (4.3%)
www.biosciencetrends.com
BioScience Trends. 2015; 9(4):270-274. 273
The latter perspective is supported by previous studies
regarding underuse of services. Elderly individuals and
their caregivers have been reported to underuse long-
term care services until informal care arrangements
become unmanageable (7). In Japan, the most common
primary caregivers are cohabitant spouses (26.2% in
2013) and women (68.7%) (4). Because healthy life
expectancy is longer in women than it is in men (8), men
might be able to depend on their wives for caregiving.
On the other hand, women might not be able to depend
on their husbands. In addition to a shorter healthy life
expectancy, male caregivers require more formal help
(9). The difference in intention to recruit outside support
might relate to differences in disability level among
newly qualified elderly individuals differently by gender.
For the total sample, the most frequent disease was
CVD (13.6% of all newly disabled elderly individuals).
The association between stroke and disability has been
examined in several reviews. In the Global Burden
of Disease Study, Murray and Lopez (10) estimated
that, in developed regions, the leading causes of loss
of disability-adjusted life years include CVD. In the
Japanese nationwide survey "Comprehensive Survey
of Living Conditions," participants living with disabled
persons were asked about the direct cause of disability,
and the most common disease among highly disabled
persons is CVD (about 30–35% of participants) (4).
Because these were the results of caregiver self-reports,
under-reporting or over-reporting is a possibility. Our
findings show the same trend for CVD, supporting
the suggestion that CVD is the most common disease
directly preceding disability in elderly individuals.
Our results further provide evidence for the effect
of arthropathy, fracture, and dementia on incidence of
disability. For women, these diseases accounted for
about 10% of low-level disabled women (Table 2).
This percentage was about the same as CVD. Because
about half of newly disabled elderly individuals had
difficulty with daily living as a result of low-level
disability, prevention of low-level disability could have
a substantial effect on extending healthy life expectancy.
In order to prevent low-level disability, especially among
women, it is important to address arthropathy, fracture,
and dementia in addition to CVD.
Because our analysis was conducted retrospectively,
we could not detect the exact risk of having each
disease on disability. In order to better demonstrate
the importance of disease prevention, prospective
observations are necessary. Further, the data were
obtained from one prefecture, and thus, we cannot apply
the findings to other Japanese regions or other countries.
In conclusion, we found that CVD was the
most common disease in the 6 months preceding
disability among elderly individuals across most age
and disability groups. In low-level disabled women,
arthropathy and fracture were as common as CVD
was. Because the frequency of low-level disability was
three times greater than that of high-level disability in
women, this finding emphasizes the need to consider
arthropathy and fracture as well as CVD in order to
prevent disability.
Acknowledgements
Financial support for this study was provided by the
Health Labor Sciences Research program in 2013
and 2014, and a Grant-in-Aid for Scientific Research
(KAKENHI), in 2013, Japan. Professor Yasushi
Table 2. Frequency of disease and disability level by sex (n = 3,715)
Items
Men
Cancer
CVD
Arthropathy
Fracture
Pneumonia
COPD
Dementia
Psychiatric
Neurological disorder
Women
Cancer
CVD
Arthropathy
Fracture
Pneumonia
COPD
Dementia
Psychiatric
Neurological disorder
Total
1,549 (100.0%)
248 (16.0%)
262 (16.9%)
90 (5.8%)
96 (6.2%)
73 (4.7%)
55 (3.6%)
98 (6.3%)
55 (3.6%)
91 (5.9%)
2,166 (100.0%)
162 (7.5%)
245 (11.3%)
216 (10.0%)
247 (11.4%)
38 (1.8%)
13 (0.6%)
186 (8.6%)
103 (4.8%)
117 (5.4%)
Values are presented as n (%); CVD, cerebral vascular disorder; COPD, chronic obstructive pulmonary disease; *Β, partial regression coeffi cient
in the ordinal regression analysis for disability level (1 = low, 2 = middle, 3 = high), adjusted with age.
Low
617 (39.8%)
79 (12.8%)
75 (12.2%)
44 (7.1%)
29 (4.7%)
19 (3.1%)
20 (3.2%)
60 (9.7%)
28 (4.5%)
37 (6.0%)
1,088 (50.2%)
59 (5.4%)
105 (9.7%)
130 (11.9%)
89 (8.2%)
12 (1.1%)
8 (0.7%)
147 (13.5%)
64 (5.9%)
66 (6.1%)
Middle
605 (39.1%)
114 (18.8%)
96 (15.9%)
36 (6.0%)
38 (6.3%)
36 (6.0%)
23 (3.8%)
32 (5.3%)
19 (3.1%)
38 (6.3%)
724 (33.4%)
65 (10.7%)
63 (10.4%)
66 (10.9%)
104 (17.2%)
15 (2.5%)
5 (0.8%)
31 (5.1%)
30 (5.0%)
38 (6.3%)
High
327 (21.1%)
55 (16.8%)
91 (27.8%)
10 (3.1%)
29 (8.9%)
18 (5.5%)
12 (3.7%)
6 (1.8%)
8 (2.4%)
16 (4.9%)
354 (16.3%)
38 (11.6%)
77 (23.5%)
20 (6.1%)
54 (16.5%)
11 (3.4%)
0 (0.0%)
8 (2.4%)
9 (2.8%)
13 (4.0%)
Β*
0.286
0.766
- 0.512
0.561
0.470
0.159
- 0.951
- 0.410
- 0.106
0.626
0.645
- 0.428
0.588
0.727
- 0.581
- 1.284
- 0.381
- 0.288
p
0.029
< 0.001
0.015
0.004
0.036
0.538
< 0.001
0.125
0.607
< 0.001
< 0.001
0.003
< 0.001
0.017
0.332
< 0.001
0.070
0.130
www.biosciencetrends.com
BioScience Trends. 2015; 9(4):270-274.274
Iwamoto of The University of Tokyo was the chief
of the study team for the claim data analysis, Fukui
Gerontology Study. Associate Professor Ryoko
Morozumi of the University of Toyama and Associate
Professor Michio Yuda of Chukyo University were the
board members of the Fukui Gerontology Study. They
contributed to the claim data collection in this study.
The authors would like to thank the staff of Fukui
prefecture and the Institute of Gerontology at The
University of Tokyo for their assistance in carrying out
this research project.
References
1. Tsutsui T, Muramatsu N. Japan's universal long-term care
system reform of 2005: Containing costs and realizing a
vision. J Am Geriatr Soc. 2007; 55:1458-1463.
2. Adamson J, Beswick A, Ebrahim S. Is stroke the most
common cause of disability? J Stroke Cerebrovasc Dis.
2004; 13:171-177.
3. American Stroke Association. About stroke. http://www.
strokeassociation.org/STROKEORG/AboutStroke/About-
Stroke_UCM_308529_SubHomePage.jsp (accessed
March 30, 2015).
4. Ministry of Health, Labour and Welfare. Comprehensive
Survey of Living Conditions (2013). http://www.mhlw.
go.jp/toukei/saikin/hw/k-tyosa/k-tyosa13/ (accessed
March 30, 2015).
5. Newman AB, Brach JS. Gender gap in longevity and
disability in older persons. Epidemiol Rev. 2001; 23:343-
350.
6. Reeves MJ, Bushnell CD, Howard G, Gargano JW,
Duncan PW, Lynch G, Khatiwoda A, Lisabeth L.
Sex differences in stroke: Epidemiology, clinical
presentation, medical care, and outcomes. Lancet Neurol.
2008; 7:915-926.
7. Kadushin G. Home health care utilization: A review of
the research for social work. Health Soc Work. 2004;
29:219-244.
8. Liu Y, Arai A, Obayashi Y, Kanda K, Boostrom E,
Lee RB, Tamashiro H. Trends of gender gaps in life
expectancy in Japan, 1947-2010: Associations with
gender mortality ratio and a social development index.
Geriatr Gerontol. 2013; 13:92-97.
9. Naruse T, Nagata S, Taguchi A, Murashima S.
Classification tree model identifies home-based service
needs of Japanese long-term care insurance consumers.
Public Health Nurs. 2010; 28:223-232.
10. Murray CJ, Lopez AD. Alternative projections of
mortality and disability by cause 1990–2020: Global
Burden of Disease Study. Lancet. 1997; 349:1498-1504.
(Received April 22, 2015; Revised June 15, 2015;
Accepted June 18, 2015)
