UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl) UvA-DARE (Digital Academic Repository) Functional disability in elderly men van den Brink, C.L. Link to publication Citation for published version (APA): van den Brink, C. L. (2005). Functional disability in elderly men. General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. Download date: 14 Aug 2019
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Functional disability in elderly men
van den Brink, C.L.
Link to publication
Citation for published version (APA):van den Brink, C. L. (2005). Functional disability in elderly men.
General rightsIt is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulationsIf you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, statingyour reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Askthe Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,The Netherlands. You will be contacted as soon as possible.
claudication,, stroke, cancer, diabetes, and asthma and chronic obstructive pulmonary disease.
Diagnosess were obtained from a questionnaire'47' and verified using clinical examination and
writtenn information from the subjects' general practitioners and hospitals.
Statistica ll analyse s
Too investigate whether the three countries differed in self-reported disability and performance-
basedd limitation, the differences between countries in self-reported disability scores and
performancee scores were tested using analysis of variance, adjusted for age.
Too determine whether possible differences in self-reported disability between countries could be
attributedd to differences in performance score, the association between country and self-reported
disabilityy was adjusted for performance score. Therefore, polytomous logistic regression models
weree used with country as independent variable and the four-level self-reported disability score (0,
1,2,3)) as dependent variable. To gain insight into the effect of different covariates on self-reported
23 3
ChapterChapter 2
disability,, three models were used. In the first model, the association between country and self-
reportedd disability was calculated and adjusted for age only. In the second model, SES, household
composition,, and the prevalence of chronic diseases were added as covariates. The third model
wass also adjusted for performance score. The extent to which the variables of this final model
explainedd the variance in self-reported disability was determined by calculating the explained
variancee of the model.
Too compare the strength of the association between self-reported disability and performance-
basedd limitation between countries, stratified analyses were performed. For each country the
associationn between self-reported disability score and performance-based limitation score was
assessedd using a polytomous logistic regression model. Age, SES, household composition, and
thee prevalence of chronic diseases were adjusted for. In each country, the association between
self-reportedd disability and performance-based limitation was presented as odds ratio (OR).
Too check whether disability per domain of self-reported disability was associated with limitation in
specificc performance tests, the mean performance score between men with and without disability
inn that domain were compared per disability domain and per EPESE-test. Linear regression
modelss were used to calculate and compare these means, adjusted for age.
Statisticall analyses were performed using SAS, version 8.2 (SAS Inc., Cary, NC). All tests were
two-tailedd and a p-value < 0.05 was considered to be statistically significant.
Result s s
Sociodemographicc characteristics and prevalence figures of chronic conditions in the participants
aree shown in table 2.1. Men in Italy were older (77.0 vs 75.4 and 75.4 years), belonged more
frequentlyy to the low SES group (82.4% vs 40.8 and 44.6%) and had a higher prevalence of
chronicc diseases (73% vs 46 and 50%) compared with men in Finland and the Netherlands. The
percentagee of men living alone was highest in Finland (23% versus 17% in the Netherlands and
12%% in Italy).
Self-reportedd disability and performance-based limitation in different countries
Thee prevalence of self-reported disability on a certain item or domain differed between countries
(tablee 2.2). The overall disability in the IADL domain did not differ much between countries, but the
prevalencee of self-reported disability on 'do light housework' was about four times higher in Italy
thann in Finland and the Netherlands. Prevalence of self-reported disability on the mobility items
andd on most items in the BADL domain was highest in the Netherlands. Italian men had the
highestt overall prevalence of disability in the BADL domain (table 2.2).
24 4
Performance-basedPerformance-based limitation and self-reported disability
Regardingg the overall disability score of all countries, more than 80 percent of the men were not
disabledd or disabled in lADLs only (scores 0 and 1; table 2.2). The mean self-reported disability
score,, which was based on the hierarchy levels, was worse in Italy (0.72) and the Netherlands
(0.70)) than in Finland (0.54).
Thee performance-based limitation score was worse in Italy than in Finland and the Netherlands
(4.800 vs 4.04 and 3.74; table 2.3). Finnish men scored worst on standing balance (0.29 vs 0.22
andd 0.20) and scored best on chair stand (1.37 vs 1.56 and 2.00) and shoulder rotation (0.47
versuss 0.95 and 0.94). Dutch men scored best on walking speed (1.05 vs 1.92 and 1.67).
Tablee 2.1 Description of demographic characteristics and chronic conditions.
Finland d
(nn = 340)
Netherlands s (nn = 481)
Italy y
(nn = 340)
Meann age
Socioeconomicc status
%% low
%% middle
%% high
Householdd composition
%% living alone
Prevalencee of chronic diseases, %
75.4* * 75.4* * 77.0 0
40.8* *
58.9™ ™
0.3* *
2 3" "
46* *
44.6* *
40.2'* *
15.3** *
17* *
50* *
82.4*1 1
16.7'1 1
0.9* *
12" "
73* *
** Significantly different from Finland (p < 0.05); tested by analysis of variance (ANOVA) or chi-square test, tt Significantly different from the Netherlands (p < 0.05); tested by ANOVA or chi-square test, ii Significantly different from Italy (p < 0.05); tested by ANOVA or chi-square test.
Associatio nn betwee n self-reporte d disabilit y and countr y adjuste d for performance-base d
limitatio n n
Regressionn analyses on the association between country and self-reported disability (table 2.4)
showedd that, adjusted for age, Dutch and Italian men reported 50% more disabilities than Finnish
men.. This is in accordance with results from table 2.2. After adjustment for SES, household
composition,, and chronic diseases, Dutch men still reported 50% more disabilities than Finnish
men,, whereas the difference between Italian and Finnish men was reduced to 20%. Also after
furtherr adjustment for performance score, Dutch men reported most disabilities (OR: 1.66; 95%
confidencee interval (CI): 1.23, 2.25) and no difference was observed between Finland and Italy.
Thee variables of this final model explained 15% of the variance in self-reported disability.
25 5
ChapterChapter 2
Tabl ee 2.2 Self-reported disability per domain and per item and distribution of disability scores (%
participants)) by country, adjusted for age.
lADLs s
Preparee own meal
Doo light housework
Doo heavy housework
Mobility y
Movee outdoors
Usee stairs
Walkk at least 400 meters
Carryy a heavy object for 100 meters
BADLs s
Walkk between rooms
Usee the toilet
Washh and bath oneself
Dresss and undress
Gett in and out of bed
Feedd oneself
Meann disability score (0-3)*
Scoree 0
Scoree 1
Scoree 2
Scoree 3
Finland d
(nn = 340)
41.2(2.6)* *
17.11 (2.2)*
9.2{1.8)s s
36.88 (2.6)s
8.8(1.8)* *
1.11 (0.9)*
1.0(0.9)* *
2.0(1.2)* *
8.44 (1.7)*
3.2(1.2)§ §
0.44 (0.7)*
1.0(0.7) )
2.2(1.0) )
0.44 (0.8)*§
0.77 (0.8)*
0.44 (0.8)*
0.544 (0.04)s*
58.88 (2.6)*5
31.9(2.5) )
6.11 (1.5)*
3.22 (1.2)5
Netherlands s
(nn = 481)
prevalencee standard error
46.99 (2.2)
25.6(1.8)* *
7.9(1.5)s s
38.33 (2.2)s
15.4(1.5)* *
3.88 (0.7)*
4.66 (0.8)*
6.9(1.0)* *
12.9(1.4)* *
4.6(1.0) )
2.66 (0.6)*
2.66 (0.6)
4.00 (0.9)
3.11 (0.7)*
3.33 (0.6)*
2.99 (0.6)*
0.700 (0.04)*
50.66 (2.2)*
33.0(2.1) )
11.7(1.3)* *
4.6(1.0) )
Italy y
(nn = 340)
48.88 (2.6)T
21.3(2.2) )
33.3(1.8)** *
47.33 (2.6)**
12.3(1.8) )
2.77 (0.9)
2.55 (0.9)
4.5(1.2) )
10.6(1.7) )
7.0(1.2)* *
1.4(0.7) )
1.8(0.7) )
5.11 (1.0)
2.99 (0.8)*
1.4(0.8) )
2.33 (0.8)
0.722 (0.04)*
50.88 (2.6)*
34.00 (2.5)
8.2(1.6) )
7.0(1.2)* *
** 0 = not disabled; 1 = disabled in IADL only; 2 = disabled in IADL and mobility; 3 = disabled in IADL, mobility,, and BADL.
tt Significantly different from Finland (p < 0.05). tt Significantly different from the Netherlands (p < 0.05). §§ Significantly different from Italy (p < 0.05).
Associatio nn betwee n self-reporte d disabilit y and performance-base d limitatio n
Figuree 2.1 shows the association between self-reported disability and performance-based
limitation.. In all countries, performance score was positively associated with self-reported disability
score.. Regression analyses showed that the strength of this association did not differ between the
countries,, because the ORs of performance score on self-reported disability were comparable. For
Finlandd the OR was 1.35 (95% CI: 1.21, 1.50), for the Netherlands 1.23 (95% CI: 1.13, 1.34), and
forr Italy 1.30 (95% CI: 1.18, 1.43). The summary OR for all three countries together was 1.28 (95%
CI:: 1.21, 1.35).
26 6
Performance-basedPerformance-based limitation and self-reported disability
Too explore the association further, the mean performance scores per disability domain and per
EPESEE test were compared between men with and without disability in that domain (results not
shown).. In all countries, men with iADL disabilities had worse performance of both the lower
(walkingg speed test, chair stand test) and the upper extremities (shoulder test) than men without
IADLL disabilities. Also men with mobility disabilities had worse performance in the walking speed
andd chair stand test. In all three countries, men with BADL disabilities had worse scores on the
chairr stand test. Balance test scores were related to disability for all domains in Finland and Italy
only. .
Tablee 2.3 Performance-based limitation scores by country, adjusted forr age.
Standingg balance8
Walkingg speed5
Chairr stand*
Shoulderr rotation5
Totall performance-based score"
Finland d
(nn = 340)
0.299 (0.03)*
1.92(0.05)* *
1.37(0.06)** *
0.477 (0.05)**
4.044 (0.12)*
Netherlands s
(nn = 481)
Meann Standard Error
0.222 (0.02)
1.05(0.05)** *
1.56(0.05)** *
0.955 (0.04)*
3.74(0.10)* *
Italy y
(nn = 340)
0.200 (0.03)
1.67(0.05)** *
2.000 (0.06)**
0.944 (0.05)*
4.80(0.12)** *
** Significantly different from Finland (p < 0.05). tt Significantly different from the Netherlands (p < 0.05). $$ Significantly different from Italy (p < 0.05). §§ Score varied from 0 (best) to 4 (worst). ||| Score varied from 0 (best) to 16 (worst).
Tablee 2.4 Effect of country on self-reported disability.
Modell 1 Modell 2* Model 3*
Oddss Ratio (95% Confidence Interval)
Finlandd (reference)
Netherlands s
Italy y
1.0 0
1.52(1.15,2.00) )
1.50(1.12,2.03) )
1.0 0
1.55(1.15,2.08) )
1.23(0.88,, 1.73)
1.0 0
1.66(1.23,2.25) )
1.08(0.77,, 1.53)
** Adjusted for age. tt Adjusted for age, socioeconomic status (SES), household composition, chronic diseases. tt Adjusted for age, SES, household composition, chronic diseases, performance-based limitation score.
27 7
ChapterChapter 2
OO Finland AA Netherlands DD Italy
0 11 2 3 4 5 6 7 8 9 10 11
performance-basedd score
Figur ee 2.1 Mean self-reported disability score (0-3) by performance-based limitation score per country,
adjustedd for age.
Discussio n n
Inn this study, self-reported disability, performance-based limitation, and the association between
themthem were compared in elderly men from Finland, the Netherlands and Italy. Overall, Dutch and
Italiann men reported more disabilities than Finnish men. Italian men scored worst on the
performance-basedd tests. After adjustment for performance-based score, Dutch men still reported
moree disabilities than Finnish men, whereas the difference between Italian and Finnish men
reduced.. Performance-based limitation score was positively associated with self-reported disability
score.. The strength of this association did not differ between countries.
Furtherr analyses of self-reported disability showed that, in Italy, these poor scores were partly
explainedd by a higher prevalence of lower SES, chronic diseases and performance scores. In the
Netherlands,, the prevalence of these factors was not high and did not explain the higher self-
reportedd scores. There are evidently other factors that influence the self-report of disability, which
iss confirmed by the finding that the model with all mentioned factors explained only 15% of the
variancee in self-reported disability.
Accordingg to the Nagi Scheme, sociocultural and physical environment also influence self-reported
disability.. All these factors influence self-reported disability in a complex way. Cross-cultural
28 8
Performance-basedPerformance-based limitation and self-reported disability
variationn in IADL disability could be caused by differences in role expectations.' ' It could be
explainedd by men of this generation not being accustomed to performing domestic activities at all
andd therefore reporting a need for help. According to one study,(48> IADL questions are based on
normativee roles and activities and might therefore not be applicable to determine health status or
physicall functioning. Another study(19> called these disabilities 'situational disabilities', which are
onlyy partly due to health problems. Mobility and BADL questions focus more on basic physical
ff unctions.(48) Also in these domains, Dutch men reported more disabilities than men from Finland
andd Italy, adjusted for performance score. Possibly men in the Netherlands have more assisting
devicess (social and physical)(39) in daily functioning at their disposal. Another possible difference
betweenn the cultures might be the interpretation of the meaning of the response scales due to
differencess in the cultural and linguistic meaning of 'good health or functioning'.(9;10;49"51) The
perceptionperception of 'objective' health problems differs between cultures.(49) Although all had mentioned
factorss that might have contributed to the differences found in self-reported disability between
countries,, an unequivocal explanation could not be given for these results. Although the authors
foundd no earlier studies comparing self-reported disability adjusted for performance between
differentt cultures, differences between cultures were also found for self-reports on health(49;5153)
andd vision-related functional capacity/54' after adjustment for objective measurements.
Despitee differences in the self-report of disability between countries, the strength of the association
betweenn self-reported disability and performance-based limitation did not differ between Finland,
thee Netherlands and Italy. Studies from the United States'111340' reported a stronger association
thann this study. This study investigated the association between overall limitation and disability in
differentt domains, whereas the U.S. studies used tests that were more specific to disability items
(e.g.,, restricted to lower extremity functioning). European studies are not comparable because of
differentt measurement and scoring methods.18124142' This is the first study that investigated this
associationn in different countries, using the same methodology. The authors in this study are aware
that,, despite the standardisation, some small methodological differences might still have been
present.. Overall, the self-reported disability and functional limitation appeared to be associated
consistentlyy in different countries, and this association can therefore be generalised.
Thee significant associations between performance of the lower (walking speed and chair stand
test)) and upper (shoulder test) extremities and disability in the IADL domain in the present study
weree in accordance with findings from other studies in elderly from Italy and the United States/40411
Otherr investigators also showed associations between lower and upper extremity functioning and
disabilitiess in the BADL domain.(13:40) This association with BADLs was not significant for all
performancee tests in the three countries, which the small number of men with BADL disabilities
mightt explain.
29 9
ChapterChapter 2
Somee methodological remarks must be made. Selection bias in the study populations might have
influencedd the results of the present study. In the Netherlands and Italy, non respondents had more
severee disabilities than respondents.(55:56) In the present study, there was also a selection bias
becausee of missing values. Men removed from this study because of missing values were older,
reportedd more disabilities and had more functional limitations. Both the bias due to non response
andd that to missing values have led to an underestimation of the disability and limitation levels in
thee populations. The number of men removed because of missing values was larger in Finland
thann in the Netherlands and Italy, which was primarily due to missing values on the performance
tests,, but the number of nonrespondents was lowest in Finland. The total percentage of the
numberr of nonrespondents and of men removed because of missing values did not differ between
thee countries. Because men excluded in this study had poor disability and performance scores, the
strengthh of the association between these scores might have been underestimated in this study.
AA few limitations of the present study must be mentioned. The results of the present study were
restrictedd to men aged 70 and older from three European countries around 1990. Whether the
observedd differences between men in different countries also hold for women is not known.
Unfortunately,, data on women was not available. Women o verre port disabilities and men
underreportt them, which complicates comparison of measurements of health based on self-report
betweenn men and women.(57) Also the role expectations, especially for the IADL domain, are
differentt between men and women, which will lead to different results. Furthermore, the differences
betweenn countries might change over time, because of changing roles, particularly on household
taskss for men. Only a restricted number of chronic conditions were considered in this study.
Informationn on musculoskeletal and neurological diseases was not available, which might also
havee influenced the results. Also poor cognitive functioning and clinical depression might play a
rolee in the self-report of disability. However, the prevalence of these conditions was low among the
participants. .
Ann important question is how to interpret the results. Performance tests assess basic objective
functionall limitation, whereas self-reported items reflect dependency and need for care.<7;42:58> If
comparingg the health of populations is based on an indicator for objective functioning, it can be
concludedd from the present study that men in Italy had the worst health. When health was
assessedd on the basis of a self-report of disability as an indicator for dependency and need for
care,, the Dutch men were the unhealthiest of all the populations studied.
Overall,, the results of the present study showed that the self-report of disability differed between
Finland,, the Netherlands, and Italy. These differences may be due to sociocultural and physical
environmentall factors. Self-reported disability was consistently associated with performance-based
limitationn in these countries.
30 0
3 3 Effec tt of widowhoo d on disabilit y onse t
PublishedPublished as: van den Brink C.L., Tijhuis M., van den Bos G.A.M., Giampaoli S., Nissinen A., Kromhout D.
Effectt of widowhood on disability onset among elderly men from three European countries.
JournalJournal of the American Geriatrics Society 2004; 52:353-358
ChapterChapter 3
Abstrac t t Objectiv ee To investigate in different countries the effects of becoming widowed, duration of
widowhood,, and household composition of widowed men on disability onset in different disability
domains. .
Method ss Longitudinal data of three cohorts from Finland, the Netherlands, and Italy was collected
aroundd 1990, 1995, and 2000. Complete information was available for 736 men, aged 70 and older
att baseline. Disability was measured using a standardised questionnaire on activities of daily living
(ADLs).. Three domains were assessed: instrumental ADLs (lADLs), mobility, and basic ADLs
(BADLs).. Duration of widowhood was divided into less than 5 years and more than 5 years and
householdd composition into living alone and living with family or in an institution.
Result ss Men who became widowed developed more IADL (odds ratio (OR): 2.15; 95% confidence
intervall (CI): 1.22, 3.81) and mobility (OR: 1.84; 95% CI: 1.15, 2.96) disabilities than men who were
stilll married. Men who had been widowed for less than 5 years developed more IADL disabilities
thann those who had been widowed for 5 years or more (OR: 2.27; 95% CI: 1.14, 4.54). Widowed
menn living alone showed fewer disabilities in mobility (OR: 0.25; 95% CI: 0.09, 0.73) and BADLs
(OR:: 0.02; 95% CI: 0.001-0.33) than those living with others. The effects on disability onset did not
differr between countries.
Conclusio nn Widowhood in elderly men is a risk factor for dependency in lADLs and mobility. The
growthh in the number of widowers may lead to higher demands on family care and professional
care. .
32 2
Widowhoodd and disability
Introductio n n Populationn ageing and the accompanying increase in the number of persons having problems
performingg activities of daily living (ADLs) independently will lead to a continuous rise in healthcare
needs/59:60)) An important growing group of elderly, expected to be vulnerable to these disabilities,
aree men who lose their partner. For example, in the Netherlands, the number of widowed men in
20011 increased more than 8 per cent over 1991,(61> which is related to the ageing of the population
andd the relatively greater increase in life expectancy in men than in women. To achieve better
insightt into future needs for care, it is of interest to study the effect of widowhood on disability
onset. .
Inn general, a distinction can be made between functional and situational disability/19* Functional
disabilityy is primarily caused by health problems and can affect all domains of disability in ADLs,
i.e.. instrumental ADLs (lADLs), mobility and basic ADLs (BADLs). Widowhood might lead to
functionall disability because of health problems after widowhood.*17162"64' Situational disability
concernss non-health factors and occurs when persons have never learned to perform certain
tasks,, such as household tasks for men. For these tasks, situational disability might appear in
widowers,, because these men were accustomed to receiving instrumental support from their
spouse.. For all disability domains, a higher risk of disability for widowed men was hypothesised
thann for men who did not become widowed.
Twoo factors might be associated with disability onset among widowed men: duration of widowhood
andd household composition. Over time a widower adapts his behaviour and changes his
standards/65'' so men who had been widowed longer were hypothesised to have a lower risk of
disabilityy than men who had become widowed recently. For household composition (living alone or
withh others), the hypothesis was that disability, especially in lADLs, would be more prevalent in
widowedd men living alone than in those living with others, because care tasks of the lost partner
cann be taken over by others for the widowed men not living alone.
Culturall factors might further influence the effect of widowhood on disability onset and the
associationn with household composition/161 Because the experience of widowhood and the way
livingg arrangements are valued may vary with culture/6667' the hypothesis was that the effects of
widowhood,, the duration of widowhood, and the association between household composition and
disabilityy domains would be greater in cultures promoting family interdependency than in those
promotingg autonomy.
Thee aim of the present study was to investigate in elderly men the influence of becoming widowed
andd the duration of widowhood on disability onset in different domains in Finland, the Netherlands,
andd Italy. Moreover, the association between household composition and disability onset was
33 3
ChapterChapter 3
studiedd in widowed men in the three countries. The study variables were measured in a
standardisedd way in all countries, providing the opportunity to evaluate cross-cultural differences.
Method s s
Stud yy populatio n
Thee present study had a longitudinal design and used data of the Finland, Italy, and Netherlands
Elderlyy (FINE) Study, collected around 1990, 1995 and 2000. The FINE Study began in 1985 as a
continuationn of the Seven Countries Study,(36) and is focused on elderly men born between 1900
andd 1920. Detailed information about the FINE Study and its populations has been reported
elsewhere.(43) )
Thee present study included 736 participants: 225 from Finland, 294 from the Netherlands, and 217
fromm Italy. Information was available on marital status and disability. Men who were divorced (2%),
hadd never been married (6%) or were already disabled in all domains in 1990 and 1995 (3%) were
excluded. .
Dataa was collected using a questionnaire and was checked for missing values and inconsistencies
byy staff, according to the international protocol used in surveys of the Seven Countries Study.(36)
Disabilit yy (1990, 1995 and 2000)
Disabilityy was defined as dependency in ADLs and was measured for 14 items, which were
groupedd into three domains: IADL (preparing meals, doing light and heavy housework), mobility
(movingg outdoors, using stairs, walkingg 400 meters, carrying a heavy object for 100 meters), BADL
(walkingg indoors, getting in and out of bed, using the toilet, washing and bathing, dressing and
undressing,, feeding oneself). The item 'cut toenails' was left out of the analysis, because it consists
off aspects of both the mobility and BADL domain (conceptual ambiguity).<8;45) Participants were
classifiedd as being disabled in a certain activity if they reported a need for help, or were not able to
performm that activity. For each domain, disability was dichotomised based on disability on at least
onee item of that domain.
Duratio nn of widowhoo d and househol d compositio n (1995 and 2000)
Durationn of widowhood at the end of a period was dichotomised as less than 5 years (men who
becamee widowed during a certain period) and 5 or more years (men who were widowed at both the
beginningg and the end of the period (in 1990 and 1995 or in 1995 and 2000)).
Householdd composition was dichotomised as living alone or living with other adults (living with
grownn children, family, others, and living in an institution).
34 4
WidowhoodWidowhood and disability
Countrie s s
Inn Finland, participants came from llomantsi, a hilly area in eastern Finland, and Pöytya and Mellila
inn southwestern Finland, which are flat, rural areas. In the Netherlands participants came from
Zutphen,, a commercial town in the eastern part of this flat, lowland country. In Italy, participants
camee from two rural villages: Montegiorgio, located in the hills, and Crevalcore, located in a flat
valley. .
Regardingg household composition and family structure, Italy differs from Finland and the
Netherlandss because the proportion of elderly living with their children is higher in Southern Europe
(Italy)) than in Northern Europe (Finland and the Netherlands)/6869*
Sociodemographi cc characteristic s
Sociodemographicc characteristics included were age and socioeconomic status (SES). The
professionn that was held during the major part of working life was selected as SES. Three groups
weree distinguished: high SES (professionals, high-level managers, and high-level teachers), middle
SESS (middle-level managers, middle-level teachers, and (small) business owners), and low SES
(nonmanuall and manual workers).
Statistica ll analyse s
Thee follow-up period was divided into two periods: from 1990 to 1995, and from 1995 to 2000. Of
thee 736 participants, 315 had complete information only for Period 1, 21 only for Period 2, and 400
forr both periods. Each period a subject participated in accounted for one observation. Combining
thee two periods resulted in 1,136 observations of 736 participants.
Forr all analyses, data from the two study periods were used. Although the outcome variable
disabilityy was measured at two points in time (repeated measurements), no repeated measurement
effectss were included in the logistic regression models because, for each person, the event
(disability)) can occur only once because, for each period, the analyses started with men without
disabilities. .
Too determine the effect of becoming widowed on disability onset, only men who were not widowed
andd had no disabilities at the beginning of the period were included in the analyses. At the end of
thee period, disability was compared between widowed and non-widowed men. Logistic regression
analysess were performed for each domain with 'disability at the end of the period' in that domain as
thee dependent variable and 'widowhood at the end of the period' as the independent variable.
Too determine the influence of duration of widowhood on disability onset, a logistic regression
modell was performed for each disability domain with disability as the dependent variable and
35 5
ChapterChapter 3
durationn of widowhood as the independent variable. These analyses were performed on men
withoutt disability at the beginning of the period who were widowed at the end of a period.
Too determine the association between household composition and disability in men who became
widowed,, only the incident cases of widowhood (men who became widowed between the
beginningg and the end of a period) were included in the analyses. The analyses were again
restrictedd to men without disability at the beginning of the period. At the end of the period, disability
wass compared between widowed men living alone and those living with others. A logistic
regressionn model was used with household composition as the independent variable and disability
ass the dependent variable. Exploratory analyses were done within the group of widowed men living
withh others by calculating mean disability scores for widowers living with family and for those living
inn an institution. Because of the relatively small number of men living in institutions (8%), this
differentiationn was not possible in the logistic regression analyses.
Too determine whether the effects of widowhood and duration of widowhood on disability onset, and
thee association with household composition differed between countries, interaction terms between
countryy and the different variables were added to the different models. Furthermore, the analyses
weree stratified by country.
Alll the logistic regression analyses were adjusted for age, SES, and country with exception of the
analysess stratified by country, which were only adjusted for age and SES.
Statisticall analyses were performed using SAS version 8.2 (SAS Institute, Inc., Cary, NC). All tests
weree two-tailed, and p < 0.05 was considered statistically significant.
Result s s
Tablee 3.1 shows the characteristics of the study population per period. Mean age around 1990 was
75.00 years for Finland, 74.8 for the Netherlands, and 76.3 for Italy. The prevalence of low SES was
considerablyy higher in Italy (83% in 1991) than in Finland (41% in 1989) and the Netherlands (42%
inn 1990). The prevalence of high SES was highest in the Netherlands (21% versus 1% in the other
countriess around 1990). The percentage of widowed men increased during the follow-up rounds,
fromm about 27% in 1990 to 34% in 2000. The percentage of widowed men living alone was
considerablyy higher in Finland (e.g. 65% in 1999) and the Netherlands (76% in 2000), than in Italy
(33%% in 2000). Furthermore, when living with others, Finnish and Dutch men were more likely to
livee in an institution, whereas Italian men were more likely to live with others (data not shown).
Abstrac t t Objectiv ee To investigate the relationship between duration and intensity of physical activity and
disabilityy 10 years later, and to investigate the possible effect of selective mortality.
Method ss Longitudinal data of 560 men aged 70-89 years, without disability at baseline of the
Finland,, Italy and the Netherlands Elderly (FINE) Study was used. Physical activity in 1990 was
basedd on activities like walking, bicycling and gardening. Disability severity (3 categories) in 1990
andd 2000 was based on instrumental activities, mobility and basic activities of daily living.
Result ss Men in the highest tertile of total physical activity had a lower risk of disability than men in
thee lowest tertile {odds ratio (OR) 0.46; 95% confidence interval (CI): 0.26-0.84). This was due to
durationn of physical activity (OR highest tertile 0.42; 95% CI: 0.23-0.78 compared to the lowest
tertile).. Intensity of physical activity was not associated with disability. Addition of deceased men as
fourthh category led to weaker associations between physical activity and disability (OR highest
tertilee 0.67; 95% CI: 0.44-1.02).
Conclusio nn Even in old age among relatively healthy men, a physically active lifestyle was
inverselyy related to disability. To prevent disability duration of physical activity seems to be more
importantt than intensity.
44 4
PhysicalPhysical activity and disability
Introductio n n Physicall activity in old age seems to be an important determinant of healthy ageing. In addition to
thee effect on postponement of mortality/82"84' physical activity preserves quality of life in the elderly
becausee of its positive longitudinal association with independent functioning ^functioning without
disability)/85"92'' Information on the importance of specific aspects of physical activity, i.e. duration
andd intensity, is however lacking.
Althoughh earlier studies found an association between physical activity and disability, global
assessmentss of physical activity only provide crude information about the role of physical
activity/85*7"90'' Studies that incorporated frequency or total energy expenditure of different activities
foundd a relationship between these aspects of physical activity and disability/869192' However,
durationn and intensity of the activities were not distinguished, so it is not known which role these
aspectss play in the relationship with disability. In the current longitudinal study both duration and
intensityy of physical activity are investigated.
Furthermore,, measures of association in longitudinal studies may be susceptible to bias. Especially
longitudinall studies among elderly people have large losses to follow-up because of death, which
mightt be non-random. Therefore, the effect of selective mortality should be considered when
studyingg associations with health outcome as dependent variable. In an earlier study with
functionall decline as health outcome, the associations with the determinants (social relations)
becamee stronger when death was included in the outcome measure/93'
Thee aim of the present study was to elaborate on earlier studies by investigating the relationship
betweenn physical activity and incident disability, taking into account duration and intensity of
physicall activity, and severity of disability. Furthermore, we investigated whether inclusion of
peoplee who died during the follow-up time resulted in a change in the risk of incident disability.
Dataa from a prospective study (10-years follow-up) in three European countries is presented.
Europeann longitudinal studies about physical activity and disability are lacking; the available
studiess were carried out in North America or Asia/85'
Method s s
Studyy population
Thee present study has a longitudinal design and used data collected around 1990 and 2000 for the
Finland,, Italy and the Netherlands Elderly (FINE) Study. The FINE Study started in 1985 as the
extensionn of the Seven Countries Study/36' consisting of men bom between 1900 and 1920. More
detailss about the FINE Study and its populations have been reported elsewhere/43'
45 5
ChapterChapter 4
Baselinee measurements for this study were earned out in 1990, and data collected in 2000 was
usedd for the 10-years follow-up. Around 1990, 1416 men were examined (response rate Finland
90%,, the Netherlands 78%, Italy 79%). The present study was focused on a subgroup of 560
subjects.. Men with any disability at baseline were excluded (n=780) in order to avoid the possibility
off lack of physical activity caused by disabilities. Of the survivors in 2000, 62 did not participate in
thatt examination year and were excluded. Furthermore, 10 men were removed because of
incompletee information on physical activity in 1990 and 4 because of missing values on disability in
2000.. Of the remaining 560 subjects, 183 came from Finland, 220 from the Netherlands, and 157
fromm Italy. The analyses among survivors in 2000 were restricted to 286 participants.
Dataa collection followed the international protocol used in surveys of the Seven Countries Study.(36)
Inn 1985 in Finland, the research was approved by the Ethical Committee of the Kuopio University
Hospitall and the Dutch part of the study by the Medical Ethical Committee of the University of
Leiden.. In Italy the research was approved by the Ethical Committees at local level. Subjects gave
theirr written informed consent to participate.
Disability y
Disabilityy was measured by a standardised questionnaire about daily routine activities, which was
describedd by Hoeymans et al.<8> Three domains, consisting of 13 items, were assessed:
•• instrumental activities of daily living: preparing one's own meal, doing light, and doing heavy
housework; ;
•• mobility: moving outdoors, using stairs, walking 400 meters, carrying a heavy object for 100
meters; ;
•• basic activities of daily living: walking indoors, getting in and out of bed, using toilet, washing
andd bathing, dressing and undressing, feeding oneself.
Disabilityy in a domain was defined as needing help on at least one item of that domain.
Disabilityy severity was based on the hierarchical order of the three disability domains, described by
Hoeymanss et al.(8) The two most severe groups (disability in instrumental activities and mobility,
andd disability in all domains) were taken together because of the small numbers. The following
classificationn was used:
0.. no disability;
1.. mild disability (instrumental activities);
2.. severe disability (instrumental activities and mobility, and no or any basic activities)
Too investigate the possible effect of selective mortality, further analyses were performed including
deceasedd men as fourth category.
46 6
PhysicalPhysical activity and disability
Physicall activity (In the year 1990)
Physicall activity in 1990 was measured by a standardised self-administered questionnaire,
speciallyy designed for retired men, which has been described in detail by Caspersen et al.(94) This
questionnairee is considered reliable and valid for measuring physical activity in elderly men, having
demonstratedd a 4 month test-retest correlation of 0.93 and having been validated by the doubly
labelledd water method (correlation with total energy expenditure was 0.61) in Dutch elderly.(95> The
coree questionnaire consisted of questions about six activities: the frequency and duration of
walkingg and bicycling during the previous week, the average amount of time spent weekly on
gardeningg and hobbies in both summer and winter, and the average amount of time spent monthly
onn sports and odd jobs. Because of the rural areas where the participants of Finland and Italy live,
questionss on the average amount of time spent weekly on farming in both summer and winter were
addedd to the questionnaire in these countries. Estimated times were converted to minutes per
weekk for each type of activity and summed to obtain total weekly duration of physical activity. For
hobbiess and sports, only activities that demanded a certain amount of physical effort (>2.0
kcal/kg-hour)) were included. For example, activities like playing chess or doing puzzles were not
consideredd as physical activities.
Alll activities were given an intensity code based on Caspersen et al.<94> The codes were expressed
ass kcal/kg-hour and reflected multiples of resting oxygen consumption. For walking and bicycling
thiss code was based on an additional question about the pace of the performed activity, divided
intoo three categories: calm, normal, fast. For gardening and farming the intensity code was based
onn an additional question about the strenuousness of the work, also divided into three categories.
Fromm all activities together, a mean intensity index was constructed by multiplying the intensity of
eachh activity by the time spent on that activity, summing this for all activities and dividing by total
timee spent on physical activity.
AA variable for total physical activity was constructed by multiplying duration with intensity of the
activities. .
Confoundingg factors (in the year 1990)
Possiblee confounding factors comprised of other lifestyle factors. Information on cigarette smoking
statuss (never, ever, current) was collected by questionnaire. Smoking was dichotomised as current
versuss non-smoking (never, ever).
Heightt and weight were measured while the participant stood in light clothing without shoes. Body
masss index was calculated by dividing weight (kg) by the square of height <m2). Men were
categorisedd as being obese (body mass index £ 30) or being non-obese. Although body mass
indexx < 18.5 is also a risk factor for disability, this group was not distinguished or excluded, since
onlyy four men belonged to this group and exclusion of these four men did not lead to other results.
47 7
ChapterChapter 4
Alcoholl consumption was obtained from questions about wine, spirits, and beer. These were
addedd to obtain total alcohol consumption. Alcohol consumption was dichotomised as non drinkers
versuss drinkers. Non drinkers did not drink alcohol at all. Information on beer was not available for
Italiann participants. Forty-two of the Finnish participants did not have any information on alcohol
consumptionn and were removed from the analyses in which alcohol was added as independent
variable. .
Statisticall analyses
Baselinee characteristics were compared between countries using analysis of variance for
continuouss variables and chi-square test for categorical variables.
Too investigate the relationship between total physical activity and incident disability 10 years later,
aa polytomous logistic regression model was constructed with the three levels of disability (no, mild,
severe)) as dependent variable and tertiles of total physical activity as independent dummy
variables.. The lowest fertile was the reference group. These analyses were repeated after addition
off deceased men as fourth category. The analyses were also performed adjusted for smoking,
obesity,, and alcohol consumption.
Too investigate whether duration and intensity of physical activity contributed separately to the
associationn between physical activity and incident disability, both variables were put into one model
ass separate independent variables. Because the association between duration and intensity was
nott strong (correlation coefficient = 0.20), collinearity between these factors is not likely to have
influencedd the results. To investigate whether smoking, obesity, and alcohol consumption
confoundedd the relationship between physical activity and disability, the models were also tested
adjustedd for these factors. Again, analyses were repeated with inclusion of men who died between
19900 and 2000.
Thee analyses were carried out for the men of all countries together. All analyses were adjusted for
agee and country, because these characteristics were associated with both physical activity and
disability.. Although chronic diseases were also associated with physical activity and disability, the
analysess were not adjusted for these diseases, because they are assumed to be reflected in the
disabilityy measurement.
Statisticall analyses were performed using SAS, version 8.2. All tests were two-tailed and a p-value
off < 0.05 was considered statistically significant.
48 8
PhysicalPhysical activity and disability
Result s s
Thee baseline characteristics of the participants are shown in table 4.1. Men in Italy were
statisticallyy significantly older (about 1.5 to 2 years) than men in Finland and the Netherlands. Men
inn Italy spent statistically significantly more time on physical activity (1120 minutes per week) than
menn in Finland (939) and the Netherlands (694). Mean intensity of the activities was also
statisticallyy significantly higher in Italy. In all three countries, walking contributed considerably (18-
34%)) to duration of physical activity. In the Netherlands, bicycling and gardening were also
importantt and in Italy gardening. Furthermore, in both Finland and Italy, farming contributed about
25%% to total physical activity.
Afterr 10 years of follow-up, Italian men had a statistically significantly lower mortality rate (35%
versuss 54%) and more often lived without disabilities (2000: 24% versus 14-19%) (table 4.1).
Menn who were excluded because of disabilities at baseline were 2.5 years older and their amount
off physical activity was 45% lower compared to men without disabilities at baseline. Furthermore,
100 years later, among the excluded men there were 25% more deceased men and the prevalence
off men without disabilities was 14% lower (data not shown).
Totall physica l activit y and inciden t disabilit y
Totall physical activity was related to disability, adjusted for age and country (table 4.2). Compared
too the lowest tertile of total physical activity, men from the middle (odds ratio (OR): 0.56; 95%
confidencee interval (CI): 0.32, 0.99) and highest tertile (OR: 0.50; 95% CI: 0.29, 0.88) had a lower
riskrisk of disability. Addition of deceased men resulted in slightly weaker associations between
physicall activity and disability.
Thee odds ratios of the middle and highest tertile of total physical activity did hardly differ, which
waswas the case for all models.
Adjustmentt for smoking, obesity, and alcohol consumption resulted in a slightly weaker association
betweenn physical activity and disability in the analyses in which deceased men were included. The
associationss however remained statistically significant.
Duratio nn and intensit y of physica l activit y and inciden t disabilit y
Thee cut-off points for the tertiles of duration of physical activity were 486 and 960 minutes per week
withh a median value of 270 minutes per week for the lowest tertile, 690 for the middle and 1432 for
thee highest tertile. Duration was statistically significantly associated with functional decline (table
4.3).. Men in the middle and highest tertile of duration of physical activity had about 50% lower risk
off disability (OR: 0.51; 95% CI: 0.29, 0.89 and OR: 0.45; 95% CI: 0.25, 0.81 respectively) than men
inn the lowest tertile. After adjustment for smoking, obesity, and alcohol consumption, the risk ratios
49 9
ChapterChapter 4
remainedd roughly the same. To explore the possible effect of selective mortality deceased men
weree added. The associations between physical activity and disability became weaker and became
borderlinee significant after adjustment for other lifestyle factors. The odds ratios for the middle and
highestt tertile became 0.68 (95% CI: 0.45,1.02) and 0.67 (95% CI: 0.44, 1.02) respectively.
Thee associations between intensity of physical activity and disability, independent of duration, were
nott statistically significant.
Tablee 4.1 Baseline characteristics (1990) and disability and mortality in 2000 in men free of disability at
baseline. .
Agee (years)
Physicall activity
Meann duration (min/week)
Meann intensity (kcal/kg/hour), adjusted for duration
Typee (% of total time)
walking g
bicycling g
gardening g
farming g
sports s
oddd jobs
hobbies s
Otherr lifestyle factors (%)
non-smokers s
non-obesee (body mass index < 30 kg/m2)
non-drinkers s
Disabilityy 2000 (%)
noo disability
mildd disability*
severee disability*
Deceasedd 1990-2000 (%)
Finland d
n=183 3
74.77 (4.0)
9399 (823)*
3.77 (0.7)*
34 4
11 1
12 2
25 5
1 1
9 9
8 8
87 7
86 6
13 3
14 4
16 6
16 6
54 4
Netherlands s
n=220 0
74.22 (4.1)
694(546) )
3.77 (0.7)
21 1
31 1
21 1
5 5
13 3
9 9
80 0
93 3
26 6
19 9
12 2
15 5
54 4
Italy y
n=157 7
76.11 (3.4)
1120(948) )
3.99 (0.7)
18 8
13 3
37 7
23 3
1 1
7 7
1 1
83 3
90 0
18 8
24 4
22 2
19 9
35 5
** Numbers in parentheses, standard deviation. tt Disability in instrumental activities of daily living. tt Disability in instrumental activities and mobility and no or any basic activities of daily living.
50 0
PhysicalPhysical activity and disability
Tablee 4.2 Relationship between total physical activity and disability.
Disability y Disabilityy including
deceasedd men
Totall physical activity 1990
lowestt tertile*
middlee tertile
highestt tertile
lowestt tertile*
middlee tertile
highestt tertile
oddss ratio
1.00 0
0.56 6
0.50 0
modelsmodels ac
1.00 0
0.55 5
0.46 6
95%% CI oddss ratio' 95%% CI
modelsmodels adjusted for age and country
1.00 0
0.32-0.999 0.58 0.38-0.88
0.29-0.888 0.52 0.34-0.80
modelsmodels adjusted for smoking, obesity, alcohol, age, and country
1.00 0
0.30-0.999 0.63 0.40-0.98
0.26-0.844 0.60 0.38-0.94
Abbreviations:: CI, confidence interval. ** polytomous logistic regression analysis: 3 categories of disability severity. tt polytomous logistic regression analysis: 3 categories of disability severity and 1 category of deceased men. %% reference category.
Tablee 4.3 Relationship between duration and intensity of physical activity and disability.
PhysicalPhysical activity 1990
Duration n
lowestt tertile*
middlee tertile
highestt tertile
Intensity y
Duration n
lowestt tertile*
middlee tertile
highestt tertile
Intensity y
Disability y
oddss ratio'
1.00 0
0.51 1
0.45 5
1.13 3
models models
1.00 0
0.50 0
0.42 2
1.22 2
95%% CI
Disabilityy including
deceasedd men
oddss ratioT
modelsmodels adjusted forage and country
0.29-0.89 9
0.25-0.81 1
0.81-1.58 8
1.00 0
0.59 9
0.59 9
0.89 9
adjustedadjusted for smoking, obesity, alcohol, age,
0.28-0.91 1
0.23-0.78 8
0.85-1.75 5
1.00 0
0.68 8
0.67 7
0.91 1
95%% CI
0.40-0.87 7
0.39-0.88 8
0.70-1.12 2
andand country
0.45-1.02 2
0.44-1.02 2
0.72-1.16 6
Abbreviations:: CI. confidence interval. ** polytomous logistic regression analyses: 3 categories of disability severity. tt polytomous logistic regression analyses: 3 categories of disability severity and 1 category of deceased men. %% reference category.
51 1
ChapterChapter 4
Discussio n n Thiss study shows that among relatively healthy men aged 70-89 years, total physical activity was
relatedd to decreased risk of disability 10 years later. Especially duration of physical activity was
associatedd with disability, whereas intensity was not. Addition of deceased men in our analyses
resultedd in weaker associations between physical activity and disability.
Forr the interpretation of our data, some methodological remarks on study design and selection bias
mustt be made. In order to create the most optimal design for the research question, men with
disabilityy at baseline were excluded. Although according to the definition of disability men who did
nott need help were included, these men might have had difficulties in performing activities of daily
livingg that contributed to lower physical activity. A large proportion of the population was excluded
inn the present study. However, the lower level of physical activity in this excluded group and the
higherr prevalence of disability and mortality 10 years later, point into the same direction as the
resultss of the men included in the analyses.
Selectionn bias caused by excluding men with missing values is also of concern. Men who were
removedd because of missing values on disability spent less time on physical activity and men
removedd because of missing values on physical activity had more disabilities than men included in
thee study. Also non respondents were known to have more disabilities/5556* It is however not known
whetherr the association between physical activity and disability is different in this group compared
too the men included in the present study.
Thee study population came from three countries, which differed in disability status and level of
physicall activity. Because the number of participants per country was too small to allow
comparisonss between countries, all participants were pooled. Since differences in circumstances
betweenn the countries, for example weather and physical environment (hills), could have affected
thee relationship between physical activity and disability, country was adjusted for, which resulted
onlyy in a small decrease of the associations between physical activity and disability.
Forr the present analyses elderly men with no disabilities at baseline were selected and were
thereforee relatively healthy. In addition, the amount of physical activity in these men was relatively
high,, even in men of the lowest activity group, who spent on average around 40 minutes per day
onn activities such as walking, bicycling, and gardening. Being active for 40 minutes per day is
consideredd good for health according to the Dutch physical activity guideline for elderly people that
recommendss 30 minutes per day of activities like walking and bicycling/96' However, the present
studyy suggests that spending around 100 minutes per day is even better.
Thee positive effect of physical activity on disability was found for both the middle and highest tertile
andd these odds ratios did not differ. This suggests a ceiling effect of physical activity. Earlier
52 2
PhysicalPhysical activity and disability
studiess on physical activity and disability from the United States and Hawaii did not find such a
ceilingg effect: the risk of disability decreased further between the middle and the most active
group.(86:88;91)) Studies on physical activity are however difficult to compare. One of the mentioned
studiess used tertiles of total energy expenditure, another used tertiles of frequency of certain
activities,, while the third used walking distances for determining the level of physical activity. In
general,, the dose-response curve of the relationship between physical activity and health benefit
showss that at higher levels of physical activity the effect levels off.<97) It is possible that in the
mentionedd studies from the United States and Hawaii the levels of physical activity were too low to
reachh the leveling off level, whereas in our study the physical activity level was much higher. In
general,, physical activity levels are known to be higher in European countries than in the United
Statess (United States: 38.3% inactive adults(98) versus 12% in the Netherlands<99)).
Inn addition to duration of physical activity, the effect of intensity was investigated. Earlier studies
investigatingg intensity of physical activity were not focused on disability but on diseases as health
outcomee and were described in a review that showed that the effect of intensity depended on the
individuall disease.(100) For example, for prevention of stroke moderate intensities were
recommended,, whereas for cardiovascular health a threshold of higher intensities seemed to be
better.. Our findings suggest that intensity of physical activity is not important for disability.
However,, our findings could be influenced by a limited measurement quantifying intensity.
Informationn about the intensity of walking, bicycling, gardening and farming was based on self-
report,, and for the other activities standard intensity values were used. Alternative methods would
bee to take into account individual physical fitness to determine relative intensity instead of absolute
intensityy or monitoring intensity directly. However, in another study using standard intensity values
perr activity, this kind of measurement was adequate to distinguish the effect of high and low
intensityy on coronary heart disease.(22) In our study only activities with an intensity above the
thresholdd of 2 kcal/kghour were included, and carrying out activities on a higher level seemed to
havee no additional effect on the prevention of disability. Further research on intensity of physical
activityy in relation to disability is recommended to confirm our results.
Inn the present study the role of selective mortality was also investigated. Although it is generally
assumedd that losses of follow-up due to death lead to underestimation of the association studied,
inclusionn of deceased men in our study led to weaker associations instead of stronger
associations.. Apparently, physical activity is a stronger determinant of disability than of mortality. It
iss possible that this is especially the case in our relatively healthy population with high levels of
physicall activity.
Physicall inactivity is known to occur together with other unfavourable lifestyle factors,'101} which
mightt confound the association between physical activity and disability. Because smoking,(86) body
masss index(102) and alcohol consumption003* appeared to be associated with disability in earlier
53 3
ChapterChapter 4
studiess and cluster with physical activity,'101 > these factors were adjusted for. The results of the
presentt study however showed that these lifestyle related factors hardly contributed to the
observedd association between physical activity and disability.
Althoughh we assumed that disability reflected the presence of chronic conditions at baseline, it is
possiblee that chronic conditions were present, while people were not yet disabled. Additional
analysess showed that chronic conditions such as myocardial infarction, stroke, and angina pectoris
weree associated with physical activity and with disability. However, exclusion of persons with these
chronicc conditions did not change the results, which confirms that the associations between lack of
physicall activity at baseline and higher levels of disability or mortality 10 years later were not
causedd by the presence of chronic conditions at baseline.
Inn order to translate the findings of this study to a public health message we have to consider that
physicall activity at older age might be a proxy for lifetime history of physical activity. If this is the
casee the observed effect of physical activity on disability late in life can be a result of life time
activityy pattern. However, the US Surgeon General's report on physical activity and health
suggestss that people of all ages can benefit from regular exercise.<20) Also an earlier program on
successfull ageing spread the message that it is never too late to begin healthy habits such as
moderatee physical activity.<104)
Inn conclusion, the results of the present study suggest that even in old age among relatively
healthyy men, a physically active lifestyle should be encouraged. Because the amount rather than
thee intensity of physical activity seemed to be important, there are more options for people to select
activitiess that can be incorporated into their daily lives. Spending 100 minutes per day on activities
likee walking, bicycling, and gardening decreases the risk of disability.
54 4
5 5 Disabilit yy and forma l hom e care
SubmittedSubmitted as: van den Brink CL., Tijhuis M., Klazinga N.S., Kromhout D., van den Bos GAM. Use of formal
homee care among elderly men according to need?
ChapterChapter 5
Abstrac t t
Objectiv ee - Due to pressure on professional home care, rationing and allocating care may conflict
withh the principle of equity, i.e. equal use for equal need. The aim of this study was to evaluate
whetherr use of formal home care (home nursing and home help) is according to need among
elderlyy men, using the Andersen model.
Method ss - Cross-sectional data was collected in 2000 by questionnaires, among 160 Dutch men
agedd 80 years and older in Zutphen, the Netherlands. Home nursing and home help were analysed
inn relation to need factors (e.g. chronic diseases, disability), predisposing (e.g. marital status), and
Result ss - Men with severe disability reported higher use of home nursing than men with no
disabilityy (odds ratio (OR): 20.6; 95% confidence interval (CI): 2.2, 188.8). Men who had more than
eightt years education used home nursing more often than men with less education (OR: 5.8; 95%
CI:: 1.0, 32.2). Home help was not associated with disability, but men who were married reported
lowerr use of home help (OR: 0.4; 95% CI: 0.2, 1.0).
Conclusio nn - The association of education with use of home nursing suggests inequity. Use of
homee help was not associated with health-related needs, but with marital status. Support by the
spousee decreases the demand for formal care, so there is no firm evidence for inequity in use of
homee help. Both phenomena require further attention in research and among care providers.
56 6
DisabilityDisability and formal care
Introductio n n Demographicc and epidemiologic transitions have resulted in increasing numbers of elderly people.
Ass a result, growing health care needs are expected in the decades to come, especially in long-
termm care. Because elderly people increasingly express the desire for care at home rather than
institutionalisation,, professional home care in elderly people should be sustained. However, there
iss a lot of pressure on professional home care nowadays, because of scarcity of financial
resources.. Rationing and allocating care can easily come into conflict with the principle of equity.
Thiss principle presupposes equal use for equal need. To evaluate whether use of care is according
too need for care, the model of Andersen can be used.{105) This model has often been applied to
evaluatee the use of a wide range of health care services, such as physical therapy, hospital
utilisation,, physician visits, home care, and institutional care.(106"114)
Accordingg to the model of Andersen, use of health care depends on three different groups of
determinants:: need factors, predisposing characteristics, and enabling resources. Equity in use of
caree is demonstrated when care use is explained by need or need-related factors. Inequity in use
off care is present if care use is explained by factors enabling or impeding use of health care. Need
factorss represent the most immediate cause for health service use, e.g. chronic diseases,
disability,, and perceived health. Earlier research showed that chronic diseases'1151 and
disability008112"114116117'' were associated with higher use of home care. For perceived health,
contradictoryy results were found. Some authors found a positive association with use of formal
homee care,<112:114;118> while others did not.<111;113;117)
Predisposingg characteristics refer to demographic and social structural characteristics, such as
age,, marital status, or living arrangement. From earlier research it is known that higher age<116) and
beingg widowed or living alone(108:111;112;114;117:118) are associated with higher use of home care.
Becausee help delivered by the spouse decreases the need for formal care, marital status and living
arrangementss can be considered as need-related predisposing factors.
Enablingg resources include the means and know-how people must have to obtain the services and
makee use of them, indicated by for example income or education. Studies on the association
betweenn income and formal home care use showed inconsistent results/111112114119'
Whetherr those in need of formal home care actually receive appropriate levels of care also
dependss on informal care. In the Andersen model informal care is an enabling or impeding factor.
Thee association between formal and informal care use is not unequivocal. There are two theories <120):: The first suggests an inverse association: informal caregivers are the preferred caregivers and
formall care is only delivered when informal care is not available ('hierarchical compensatory
model').(121>> The second suggests a positive association between formal and informal care, and
statess that the presence of informal care facilitates use of formal care ('bridging hypothesis').022'
57 7
ChapterChapter 5
Mostt studies in which use of home care was evaluated by the Andersen-model, were performed in
specificc patient groups, such as patients with stroke or rheumatoid arthritis.(108;111;117) Studies
directedd at the elderly did not investigate the effect of need factors, predisposing characteristics,
andd enabling resources simultaneously/116118;119:123) or were outdated/112"114'
Inn the present study we focused on elderly men, using more recent data and different groups of
determinantss simultaneously.
Thee aim of the present study is to evaluate whether use of formal home care is according to need
inn elderly men from the Netherlands.
Method s s
Studyy population
Thee present study has a cross-sectional design and used data collected on use of health care in
20000 in the Zutphen Elderly Study,(124) which belongs to the Seven Countries Study.(36) The
Zutphenn Elderly Study is a longitudinal population-based study in elderly men, born between 1900
andd 1920 and living in the town of Zutphen in the Netherlands. Of the 939 men enrolled in 1985,
2355 men were still alive in 2000. Of these survivors, 171 men participated, which corresponds with
aa response rate of 73%. Two participants were excluded because they lived in an institution. The
analysess were restricted to 160 men with complete information on chronic diseases, disability, and
formall care use.
Formall home care
Usee of formal home care was assessed as receiving home nursing or home help (yes/no) in the
previouss year. Home nursing particularly concerns personal care and nursing, while home help
concernss household activities. Both types of care were analysed separately and were
dichotomisedd as users versus non-users.
58 8
DisabilityDisability and formal care
Needd factor s
Needd factors comprised chronic diseases, disability severity and self-rated health. Information on
prevalencee of chronic diseases was collected for the following chronic conditions: myocardial
infarction,, stroke, diabetes, asthma, cancer, rheumatoid arthritis, chronic back complaints, and
arthrosis.. In the analyses one variable was used for the absence or presence of chronic diseases.
Disabilityy severity was based on 13 items of activities of daily living, which were grouped into three
domains: :
•• instrumental activities of daily living: preparing meals, doing light and heavy housework;
•• mobility: moving outdoors, using stairs, walking 400 meters, carrying a heavy object for 100
meters; ;
•• basic activities of daily living: walking indoors, getting in and out of bed, using the toilet,
washingg and bathing, dressing and undressing, feeding oneself.
Disabilityy in a domain was defined as needing help on at least one item of the domain. Disability
severityy was based on the hierarchical order of the three disability domains, which was described
inn earlier studies.(8;70)The following classification was used:
0.. no disability;
1.. mild disability (instrumental activities);
2.. moderate disability (instrumental activities and mobility);
3.. severe disability (all domains).
Twoo and a half percent of the men did not fit the hierarchy. One man who reported disability in
mobility,, but not in instrumental activities, was classified in category 2. Three men who reported to
needd help with instrumental and basic activities, but not with mobility were classified in category 3.
Categoryy 2 and 3 were combined, because of small numbers in those groups.
Globall self-rated health was assessed with a single-item question: 'How would you rate your
overalll health', with four answer categories: 1) healthy, 2) rather healthy, 3) moderately healthy, 4)
nott healthy. The categories 1 and 2, and the categories 3 and 4 were combined to 1) healthy; 2)
unhealthy. .
Predisposin gg factor s
Predisposingg factors included age and marital status. Age was dichotomised with 85 years as cut
offf point. Marital status was dichotomised as married versus not married. Of the 65 men who were
nott married, 4 were never married, 4 were divorced and 57 were widowed. Living arrangement was
nott included additionally, since most men who were married did not live alone (96%), and men who
weree not married most often lived alone (94%).
59 9
ChapterChapter 5
Enablin gg factor s
Enablingg factors encompassed occupation, education, and informal care. The information on
occupationn and education was based on the survey in 1990. Occupation was defined as the
professionn that was held during the major part of the working life. We dichotomised occupation as
highh versus low level. High occupation consisted of professionals and high-level managers and
high-levell teachers. Low occupation consisted of middle-level managers and middle-level teachers,
smalll business owners, nonmanual and manual workers.
Forr educational level the total number of years of education was asked and was categorised into
tertiles:: low: < 8 years; b) moderate: 9-12 years; c) high: > 13 years.
Informall care was defined as receiving assistance of the spouse, family members, neighbours, or
acquaintancess in the previous year. Receiving informal care was dichotomised (yes/no).
Statistica ll analyse s
Too investigate the crude associations between the determinants and use of formal home care,
generall linear models were used to assess the prevalence of formal home care for each factor.
P-valuess were calculated to determine whether differences in use of home care were statistically
significantt between categories.
Multivariatee logistic regression analyses were carried out to investigate the independent
associationss (odds ratios and 95% confidence intervals) of the need, predisposing, and enabling
factors,, in relation to use of home nursing and home help.
Inn addition, we calculated how much of the variation in home care use was explained by the
independentt variables in our models, also by using multivariate logistic regression analyses.
Statisticall analyses were performed using SAS, version 9.1 (SAS Inc., Cary, NC). All tests were
two-tailedd and a p-value < 0.05 was considered to be statistically significant.
Result s s
Menn included in this study were between 79.6 and 99.7 years old with a mean age of 84.5 years.
Tablee 5.1 shows the characteristics of the study population. About 70% of the men reported
chronicc diseases, 30% reported no disability, while 38% had severe disability. Only 17% of the
menn felt not healthy. More than 60% of the men were married. The mean educational level was
11.22 years and 40% had a high occupational level. Thirty percent of the men used formal care:
14%% received home nursing and 23% received home help. More than half of the men received
informall care (58%).
60 0
DisabilityDisability and formal care
Tablee 5.1 Prevalence of need, predisposing, and enabling factors and the association with use of home
nursingg and home help
NeedNeed factors
Chronicc diseases
no o
yes s
Disabilityy severity
noo disability
mildd disability
severee disability
Self-ratedd health
healthy y
unhealthy y
PredisposingPredisposing factors
Age e
80-855 years
85-1000 years
Maritall status
married d
nott married
EnablingEnabling factors
Occupation n
low w
high h
Education n
low w
moderate e
high h
Informall care
no o
yes s
inn elderly men (n=160).
** trend is statistically significant (p < 0.01). tt difference is statistically significant (p < 0.01).
n n
49 9
111 1
48 8
52 2
60 0
131 1
27 7
104 4
56 6
98 8
62 2
89 9
60 0
44 4
59 9
46 6
64 4
90 0
61 1
Homee nursing
8% %
17% %
2%" "
6% %
32% %
13% %
22% %
11% %
21% %
12% %
18% %
12% %
15% %
7% %
17% %
15% %
9% %
17% %
Homee help
12%+ +
27% %
13%' '
23% %
30% %
2 1 % %
30% %
21% %
25% %
16%* *
32% %
24% %
23% %
23% %
32% %
13% %
23% %
22% %
ChapterChapter 5
Determinant ss of use of forma l hom e care
HomeHome nursing
Disabilityy severity was significantly associated with the use of home nursing (table 5.1). While 2%
off men without disability received home nursing, about one third of the men with severe disability
receivedd home nursing. Neither the other need factors, nor predisposing and enabling factors were
significantlyy associated with use of home nursing.
HomeHome help
Thee need factors chronic diseases and disability severity were both associated with the use of
homee help (table 5.1). Use of home help among men with chronic diseases was more than twice
thatt of men without chronic diseases. Thirteen percent of men without disability used formal home
help,, 23% of the men having mild disability used home help, and 30% of men with severe disability
usedd home help. Also predisposing factors were of influence. Use of formal home help among men
whoo were not married was twice that of married men. None of the enabling factors were associated
withh use of formal home help.
Multivariat ee model s
HomeHome nursing
Too investigate the independent associations between the determinants and use of formal home
care,, multivariate logistic regression models were used (table 5.2). In accordance with the single
factorr analysis of table 5.1, the results showed that for use of home nursing, severe disability was
thee predominant explaining factor (p = 0.003), after adjustment for the other variables. The other
needd factors chronic diseases and self-rated health, and the predisposing factors did not contribute
too the use of home nursing. Of the enabling factors, higher educational level was associated with
higherr use of home nursing than low educational level. When the two highest tertiles of education
weree combined, the effect was statistically significant (p-value 0.045; odds ratio 5.8).
Additionall analyses showed that 19% of the variation in use of home nursing could be explained by
alll included independent variables together. About 14% of the variation was explained by the need
factorr disability. Educational level explained almost 2%.
HomeHome help
Thee multivariate analyses for use of home help showed different results compared to the univariate
analysess (table 5.2). None of the need or enabling factors were significantly associated with use of
homee help after adjustment for the other factors. The predisposing factor marital status was
62 2
DisabilityDisability and formal care
borderlinee statistically significantly associated with use of formal home help (p=0.058). Men who
weree not married received home help more often than those who were married.
Thee model of need, predisposing, and enabling factors explained 11 % of the variation in use of
homee help. Approximately 3% was explained by marital status.
Tablee 5.2 Multivariate logistic regression analyses (odds ratios and 95% confidence intervals) with use of
formall home care as dependent variable based on the Andersen model.
Homee nursing Homee help
NeedNeed factors
Chronicc diseases
noo (n=43)
yess <n=97)
Disabilityy severity
noo (n=45)
mildd <n=47)
severee (n=48)
Self-ratedd health
healthyy (n=115)
unhealthyy (n=25)
PredisposingPredisposing factors
Age e
80-855 years (n=94)
85-1000 years <n=46)
Maritall status
nott married (n=52)
marriedd (n=88)
EnablingEnabling factors
Occupation n
loww (n=82)
highh (rt=58)
Education n
loww (n=38)
moderatee (n=58)
highh (n=44)
Informall care
noo (n=58)
yess (n=82)
1.0 0
0.7(0.1,3.7) )
1.0 0
1.5(0.1,, 17.7)
20.66 (2.2, 188.8)
1.0 0
0.88 (0.2, 3.3)
1.0 0
1.4(0.4, ,
1.0 0
1.2(0.4, ,
4.7) )
4.3) )
1.0 0
0.5(0.1,2.0) )
1.0 0
5.9(1.0,34.8) )
5.55 (0.8, 40.4)
1.0 0
2.99 (0.7, 12.2)
1.0 0
1.6(0.5,4.9) )
1.0 0
2.55 (0.8, 7.9)
2.66 (0.8, 8.5)
1.0 0
1.3(0.5,3.9) )
1.1 1
0.4 4
1.0 0
(0.4,, 2.5)
1.0 0
(0.2,, ' 0) )
1.0 0
1.5(0.6,3.8) )
1.0 0
1.4(0.5,4.0) )
0.4(0.1,1.6) )
1.0 0
0.8(0.3,1.9) )
'' All independent variables were included in the same model.
63 3
ChapterChapter 5
Discussio n n
Thiss study among elderly men aged 80 years and older from the Netherlands aimed to evaluate
whetherr use of home care was according to need. We evaluated the principle of equal use for
equall need on the basis of the Andersen model, by distinguishing need, predisposing, and
enablingg factors. Disability severity was the predominant explaining factor of use of home nursing.
Inn addition, use of home nursing was associated with education. Men with higher educational level
usedd home nursing more often than men with low educational level. Predisposing and the other
enablingg factors were not statistically significantly associated with the use of home nursing.
Althoughh use of home nursing was associated with need factors, our findings showed that inequity
cannott be ruled out. For use of home help the multivariate analyses showed that none of the need
orr enabling factors were associated. Marital status was borderline statistically significantly
associatedd with use of home help: married men used home help less often than those not married.
Becausee support by the spouse decreases the demand for formal home help, there is no firm
evidencee for inequity in the use of formal home help.
Inn the present study there was selection bias in health status due to non-response. The non-
respondentss in 2000 reported more disabilities in 1995 and were older than the respondents. As a
consequencee the prevalence of disability in this population has been underestimated. Men who
weree excluded because of missing values on disability or formal care use, hardly differed from
thosee included in the analyses. The relatively small number of participants have caused the wide
confidencee intervals and the lack of statistical significance in some analyses.
Ourr data on use of home nursing and home help was restricted to the prevalence of care use. No
informationn was available on the frequency or intensity of care received.
Additionally,, we have to realise that our population came from one town in the eastern part of the
Netherlands.. Our results can probably not be generalised to other populations, because of e.g.
differentt population characteristics, cultural aspects and capacity of home care.
Ourr hierarchical disability scale was the predominant explaining factor of use of home nursing.
Accordingg to an earlier study among elderly people, in which also disability domains were
distinguished,, men with disability in instrumental activities received less home care than those with
disabilityy in basic activities of daily living.(116) Our results were also comparable with studies among
elderlyy people that evaluated use of home care using the Andersen model.<113;114) These studies
showedd that disability was more important than perceived health, and the amount of explained
variancee in these studies was comparable with ours. The amount of the total explained variance
wass not very high, but the relative contribution of disability was considerable.
64 4
DisabilityDisability and formal care
Accordingg to our results, use of home nursing was also associated with a higher educational level.
Thiss is in accordance with another Dutch study that demonstrated that people who are higher
educatedd are more inclined to search help when suffering from disabilities.025' Furthermore, people
withh lower education were admitted to a nursing home earlier than those with higher
education/107125'' These findings suggest inequity in use of home nursing.
Althoughh the distribution of home nursing was associated with need, it is possible that there is
underusee of home nursing at all levels of disability severity. For example, among men with severe
disabilityy in our study, only 32% received home nursing. Although among those with severe
disabilityy who did not receive home nursing, more than 60% received informal care, delivery of
caree by informal caregivers might be insufficient to meet the needs for personal care and nursing.
Inn contrast with home nursing, formal home help was not associated with health-related need
factors,, but was associated with marital status only. Men who were not married reported a higher
usee of formal home help than married men. Although disability severity was univariately associated
withh use of home help, this association disappeared after adjustment for marital status. This
suggestss that use of home help is not associated with health-related disability, but with situational
disability.. While disability among married men can be taken care of by the spouse, those who are
nott married are dependent on formal caregivers. Also earlier studies among elderly people and
peoplee with chronic diseases showed that widowers and people living alone received more home
caree than married people.(108;111;112;114:117;118>
Thee inverse association of being married with use of formal home help supports the 'hierarchical
compensatoryy model', which supposes that professional care is a substitute for informal care when
thiss informal care is not available. Support for this hypothesis was also found in an earlier Dutch
studyy that showed that after adjustment for disability severity, the presence of informal care and a
higherr number of network members were inversely associated with the use of formal home
help.'112'' No firm conclusions can be drawn on the bridging hypothesis, because of small numbers.
However,, the results suggest a positive association of informal care with use of home nursing.
Additionall analyses revealed that especially the presence of non-partners as informal caregivers
seemedd to facilitate the use of home nursing (not shown). These results were in accordance with
thee results of a recent Dutch study that showed a positive association between informal care and
formall care, and that the positive association was strongest when non-partners were involved.020'
Ourr analyses focused on characteristics of participants. However, access to formal care will also
dependd on characteristics of the health care system. In the Netherlands, financing of home care is
att present covered by the General Law on Special Medical Expenses (AWBZ). This law concerns
65 5
ChapterChapter 5
thee insurance of all Dutch citizens for care and support in cases of protracted illness, invalidity, or
geriatricc diseases. In the needs assessment under this Law the presence of a spouse or other
informall caregivers is taken into account.<126) In the present study, use of home help was borderline
significantlyy associated with marital status. However, use of home help was not associated with
health-relatedd need factors. This finding underpins the current reforms of the health care system in
thee Netherlands. In order to control the costs, in 2006 the AWBZ will be restricted to care, such as
homee nursing. Support, such as home help, will disappear from the AWBZ and becomes part of
thee new Law on Social Support (WMO). If family members are able to provide care, citizens will
receivee less or no support. The current government aims to promote individual responsibility, which
mightt be problematic for people without the availability of informal caregivers and those with lower
incomes.. In a few years it should be evaluated whether people who need home help actually
receivee this in the new system.
Inn supporting elderly people to stay home as long as possible, equal access to home care is
essential.. The results of the present study suggest that inequity in the use of home nursing cannot
bee ruled out, because the level of education influenced the use of home nursing. Especially those
whoo are lowly educated require attention. Use of home help was not health-related, but was
associatedd with marital status. Because support by the spouse decreases the demand for formal
homee help, there is no firm evidence for inequity in use of home help. However, the current reforms
off the Dutch health care system might cause changes in use of home help that merit careful
evaluation. .
66 6
6 6 Disabilit yy and depressiv e symptom s
WillWill be published as: van den Brink C.L., Tijhuis M.A.R., Aijönseppa S., Giampaoli S., Nissinen A., Kromhout
D.,, van den Bos G.A.M. Hierarchy levels, sum score and worsening of disability are related to depressive
symptomss in elderly men from three European countries.
JournalJournal of Aging and Health. In press.
ReprintedReprinted by Permission of Sage Publications, Inc.
ChapterChapter 6
Abstrac t t Objectiv ee To investigate the predictive value of hierarchy levels and sum score of disability, and
changee in disability on depressive symptoms.
Method ss Longitudinal data of 723 men aged 70 years and older of the Finland, Italy, and the
Netherlandss Elderly Study was collected in 1990 and 1995. Self-reported disability was based on
threee disability domains (instrumental activities, mobility, and basic activities) and depressive
symptomss on the Zung questionnaire.
Result ss Severity levels of disability were positively associated with depressive symptoms. Men
withh no disability scored 5 to 17 points lower (p<0.01) on depressive symptoms than those with
disabilityy in all domains. Among men with mild disability, those who had worsening of disability
statuss in the preceding 5 years scored 5 points higher (p=0.004) on depressive symptoms than
menn who improved.
Conclusio nn Hierarchic severity levels, sum score of disability, and preceding changes in disability
statuss are risk factors for depressive symptoms.
68 8
DisabilityDisability and depressive symptoms
Introductio n n Depressivee symptoms are considered to be the most common mental health problem in later life.
Depressivee symptoms are associated with lower quality of life and well-being,027128' a higher risk of
mortality'129130'' and a higher use of health care services.028' To prevent depressive symptoms,
moree knowledge is needed about risk factors for depressive symptoms.
Disabilityy among elderly people was associated with depressive symptoms in cross-sectional'131"136>
andd in longitudinal studies.*137"144' Disability in these studies was expressed as a sum score of
variouss disability items. It is however to be expected that besides the number of disabilities the
hierarchicc severity of the disability is also predictive of depressive symptoms. These severity levels
off disability can be expressed on the basis of different disability domains, that is instrumental
activities,, mobility, and basic activities of daily living, that are known to be hierarchically
associated.(8)) This means that people with disability in basic activities also have disability in the
otherr domains and that people with mobility disability are also disabled in instrumental activities.
Thee different hierarchy levels reflect the degree of dependence on other people. People with
disabilityy in basic activities need more help than people disabled in instrumental activities. We
thereforee hypothesised that the hierarchic severity of disability is a predictor of depressive
symptomss independent of the association of the sum score of disability with depressive symptoms.
Nott only current disability status, but also preceding change in disability status might influence
depressivee symptoms. Depressive symptoms are also influenced by changing circumstances, such
ass retirement, disability, or move to a nursing home.045146' Only a few of the mentioned longitudinal
studiess considered the effect of change in disability044' and reported that worsening of disability
statuss was related to depressive symptoms.038' The present study investigated whether current
disabilityy or change in disability is the predominant predictor of depressive symptoms.
Ann earlier study on the association between disability and depressive symptoms, carried out in
differentt countries worldwide, showed that the association was universal,047' but somewhat
strongerr in regions with a low prevalence of depressive symptoms. Studies comparing countries in
Europee showed that, in general, people in northern Europe reported fewer disabilities048' than
peoplee in southern Europe. Also depression score was known to be lower in northern Europe than
inn southern Europe.048149' In the current study we investigated whether the associations of
disabilityy with depressive symptoms differed among countries.
Thee aim of the present study was to investigate the association of the sum score and hierarchic
severityy levels of disability with depressive symptoms, and the association of the severity levels of
disabilityy with depressive symptoms, independent of the association of the sum score of disability.
Furthermoree the effect of preceding change in disability status on depressive symptoms was
69 9
ChapterChapter 6
investigated.. Finally, the various associations were compared between three European countries:
Finland,, the Netherlands, and Italy.
Method s s
Stud yy populatio n
Thee present study has a longitudinal design and used data of the Finland, Italy, and the
Netherlandss Elderly (FINE) Study, collected around 1990 and 1995. The FINE Study began in
19855 as a continuation of the Seven Countries Study,(36) and is focused on elderly men, born
betweenn 1900 and 1920. Detailed information about the FINE Study and its populations has been
reportedd elsewhere.(43)
Inn 1990, 1416 men were examined (response rates: Finland 90%, the Netherlands 78%, Italy
79%).. Between 1990 and 1995, 37% of the Finnish men died, 27% of the Dutch men died, and
20%% of the Italian men died. The response rates in 1995 were about the same as in 1990. The
presentt study included men who participated in both 1990 and 1995. A total of thirteen percent of
thee men were excluded from the analyses because of missing data on disability or depressive
symptoms.. The present study included 723 participants: Finland 221, the Netherlands 284, and
Italyy 218.
Disabilit yy (1990 and 1995)
Disabilityy was measured by a standardised questionnaire about daily routine activities.(8) Three
domainss were assessed:
•• instrumental activities of daily living (3 items): preparing meals, doing light and heavy
housework, ,
•• mobility (4 items): moving outdoors, using stairs, walking 400 meters, carrying a heavy object
forr 100 meters,
•• basic activities of daily living (6 items): walking indoors, getting in and out of bed, using toilet,
washingg and bathing, dressing and undressing, feeding oneself.
Thee participants were classified as being disabled on a certain item if they reported a need for help
orr were not able to perform that activity. A sum score was determined by counting the number of
disabilities.. The hierarchic severity level of disability was based on the ranking of the three
domains.. Disability in a domain was defined as disability in at least one item of that domain. The
domainss were hierarchically ordered as follows<8): Men who were disabled in basic activities were
alsoo disabled in mobility and instrumental activities of daily living. Men who were disabled in
mobilityy were also disabled in instrumental activities. The following hierarchic severity levels were
70 0
DisabilityDisability and depressive symptoms
distinguished:: 0) no disability, 1) mild disability: disability in instrumental activities only, 2) moderate
disability:: disability in instrumental activities and mobility, 3) severe disability: disability in
instrumentall activities, mobility and basic activities of daily living.
Changee in disability status was defined as change in hierarchic severity levels between 1990 and
19955 and was categorised as follows: 1) severe worsening: change of 2 or 3 levels to a worse
level;; for example, change from no disability to moderate or severe disability, 2) moderate
worsening:: change of 1 level to a worse level, 3) stable: stay in same level, 4) improving: change to
aa better level.
Depressiv ee symptom s (1990 and 1995)
Thee scale used to measure depressive symptoms in this study was the Self-rating Depression
Scalee (SDS) developed by Zung.<150) The questionnaire consisted of 20 items based on clinical
diagnosticc criteria most commonly used to characterise depressive disorders in terms of mood and
biologicall and psychological disturbances. In all, 10 items were worded symptomatically positive
andd 10 were worded symptomatically negative. Examples of items are: 'I feel down and sad', and 'I
cann think as clearly as before'. The items were coded as 1=never, 2=sometimes, 3=often, and
4=(almost)) always. For scoring the Self-rating Depression Scale, the positively worded items were
recoded,, so that a higher score indicated more depressive symptoms. An index for the SDS was
derivedd by dividing the sum of the item scores by 80 and multiplying it by 100, resulting in a range
fromm 25 to 100. Participants with more than 2 missing values on the 20 items were excluded. In
casee of 1 or 2 missing items, the mean score of the present items of the participant was given to
thee missing items.
Statistica ll analyse s
Forr each country the cross-sectional association between disability and depressive symptoms was
determinedd by calculating the mean level of depressive symptoms by sum score of disability and
byy hierarchic severity level of disability in 1990 and 1995. To determine the strength of these
associations,, the standardised beta and the explained variance were calculated by linear
regressionn analyses. The four categories of disability were put into the model as a continuous
variable.. To investigate whether the level of depressive symptoms differed between the countries,
regressionn analysis was performed with dummy variables for each country, with Finland as the
referencee group, adjusted for disability. To investigate whether the association between disability
andd depressive symptoms differed among countries, interaction terms of disability and country
weree added to the linear regression model.
Too investigate whether the hierarchic severity level of disability has additional value as predictor of
depressivee symptoms independent of the sum score of disabilities, cross-sectional analyses
71 1
ChapterChapter 6
(generall linear models) were carried out. The effect of hierarchic severity level of disability on
depressivee symptoms was investigated, stratified by the sum score of disability. For example,
amongg men with a sum score of two disabilities, the level of depressive symptoms of men with two
mildd disabilities (severity level 1) was compared with that of men with one mild and one moderate
disabilityy (severity level 2). Regression analysis was performed to determine the p for trend. For
menn with three disabilities and for men with four to six disabilities, the effect of hierarchic severity
levell was determined analogously. Men with seven or more disabilities were all in the level of
severee disability, so the analyses were not extended further.
Whetherr preceding change in disability was a predictor of depressive symptoms in addition to
currentt disability status was investigated by calculating the level of depressive symptoms for each
categoryy of change in the past 5 years, keeping current disability status (in 1995) constant. Firstly,
withinn the group of men with mild disability in 1995 (hierarchic severity level 1), men who had this
disabilityy already in 1990 were compared with men who developed this disability in the past 5 years
andd with men who improved in disability status during the past 5 years. The same analysis was
performedd among men with moderate disabilities in 1995 (hierarchic severity level 2) and among
menn with severe disabilities (hierarchic severity level 3). Regression analyses were performed to
determinee the p for trend. In these analyses, depressive symptoms in 1990 and country were
adjustedd for. To investigate whether the effects differed among countries, interaction terms
betweenn country and the independent variable were added.
Thee analyses were also adjusted for widowhood because widowhood appeared to be associated
withh both disability and depressive symptoms.
Statisticall analyses were performed using SAS version 8.2 (SAS Inc., Cary, NC). All tests were
two-tailedd and a p-value < 0.05 was considered to be statistically significant.
Result s s
Characteristicss of the study population are shown in table 6.1. Italian men were about 1.5 years
olderr and reported more disabilities in 1990 than Finnish and Dutch men (p = 0.0025). In both
years,, Italian men reported more depressive symptoms than Finnish and Dutch men (p < 0.0001).
Thesee differences were not explained by differences in age among the countries. Dutch men had
receivedd about six years more education and were less often widowed than Finnish and Italian
men. .
Aboutt 38% of men from Finland and the Netherlands reported worsening in disability between
19900 and 1995, compared to 28% of the Italian men. Although age and years of education varied
acrosss the populations, these variables did not confound the association between disability and
depressivee symptoms. The results are therefore shown unadjusted for these variables.
72 2
Tablee 6.1 Characteristics of the study populations.
Finland d
nn = 221
DisabilityDisability and depressive symptoms
Netherlands s
nn = 284
Italy y
nn = 218
1990 1990
Meann age (sd)
Meann number of years of education (sd)
Widowhoodd (% widowed)
Disabilityy sum score
meann sum score (sd)
Disabilityy hierarchic severity level
noo disability
mildd disability
moderatee disability
severee disability
Levell of depressive symptoms
Meann score (range 25-100)(sd)
Widowhoodd (% widowed)
Disabilityy sum score
meann sum score of disabilities
Disabilityy hierarchic severity level
noo disability
mildd disability
moderatee disability
severee disability
Levell of depressive symptoms
meann score (range 25-100)(sd)
Disabilityy change
improving g
stable e
moderatee worsening
severee worsening
** significantly different from Finland, tested by analysis tt significantly different from the Netherlands. tt significantly different from Italy.
75.0(4.1) )
4.4(3.1)T T
27% %
0.8(1.8)* *
65%* *
27% %
3%* *
5% %
45.44 (9.6)**
38% %
2.0(3.1) )
44% %
29% %
11%* *
16%* *
48.9(12.1)* *
8%* *
52% %
25%* *
14% %
off variance or chi
73 3
74.77 (4.3)
10.7(4.1)** *
19% %
0.8(1.6)* *
6 1 % %
30% %
8%* *
1% c c
41.8(8.7)** *
1995 1995
27% %
1.6(2.3) )
4 1 % * *
33%* *
19%** *
7%** *
45.77 (9.8)"*
1990-1995 1990-1995
7%* *
55% %
27%* *
11% %
-squaree test, p < 0.05.
76.44 (3.5)*T
5.00 (2.4)*
28% %
1.3(1.7)'* *
53%' '
32% %
6% %
9%b b
49.2(11.1)** *
33% %
1.7(2.7) )
51%* *
24%* *
10%* *
15%* *
50.77 (10.8)*
20%** *
52% %
17%** *
11% %
ChapterChapter 6
Cross-sectiona ll associatio n betwee n disabilit y and depressiv e symptom s
Inn all countries, in both 1990 and 1995, the sum score and the hierarchic severity levels of disability
weree positively associated with depressive symptoms (table 6.2). For example in Finland, men with
threee or more disabilities scored about 11 points higher on depressive symptoms than men without
disabilities.. Concerning the hierarchic severity level of disability, for example Italian men with
severee disability scored 5 points higher on depressive symptoms than those with moderate
disability,, who scored 5 points higher than men with mild disability, who scored 5 points higher than
thosee without disability. All p-values showed statistically significant trends indicating more
depressivee symptoms with higher levels of disability (table 6.2). In general, the strength of the
associationn between the hierarchy score of disability and depressive symptoms seemed to be
slightlyy higher than that between the sum score and depressive symptoms, which is indicated by
thee standardised betas and the explained variance (table 6.2).
Regressionn analyses showed that after adjustment for disability severity, Italian men had the
highestt score on depressive symptoms and Dutch men had the lowest score in both survey years
(pp < 0.0001). Addition of interaction terms of disability and country to the models showed that the
associationn of disability and depressive symptoms differed significantly among the countries (p =
0.01).. In the Netherlands, the strength of the association was smaller than in the other countries.
Thiss difference in the strength of association reached only statistical significance in 1990.
Hierarchi cc severit y leve l of disabilit y as predicto r of depressiv e symptoms , independen t of
th ee sum scor e of disabilitie s
Wee also investigated the additional value of hierarchic severity level of disability for a given sum
score,, as predictor of depressive symptoms (table 6.3). Among men with a sum score of two
disabilities,, men who had only mild disabilities scored 4 points lower on the scale of depressive
symptomss than men who had one mild and one moderate disability. This difference was only
borderlinee significant (p = 0.09). Among men with three disabilities a similar trend was seen: men
withh mild disability scored 7 points lower on depressive symptoms than men with severe disability
(pp = 0.07). Among men with four to six disabilities the trend was similar (p = 0.11). Furthermore,
tablee 6.3 shows that in each hierarchic level of disability, depressive symptoms do not vary with the
numberr of disabilities (= sum score).
Additionn of interaction terms between severity level of disability and country did not show
significantt differences in the effect of disability on depressive symptoms between the countries (2
disabilities:: p for interaction = 0.73; 3 disabilities: p for interaction = 0.33).
74 4
DisabilityDisability and depressive symptoms
Tablee 6.2 Cross-sectional association of sum score and hierarchic severity level of disability with
depressivee symptoms in 19900 and 1995, adjusted
Finland d (nn = 221)
forr widowhood.
Netherlands s (nn = 284)
Italy y (nn = 218)
depressivee symptoms
Summ score
noo disability
11 disability
22 disabilities
33 or more disabilities
standardisedd beta
explainedd variance
Hierarchyy severity level
noo disability
mildd disability
moderatee disability
severee disability
standardisedd beta
explainedd variance
Summ score
noo disability
11 disability
22 disabilities
33 or more disabilities
standardisedd beta
explainedd variance
Hierarchicc severity level
noo disability
mildd disability
moderatee disability
severee disability
standardisedd beta
explainedd variance
43.1 1
48.9 9
53.1 1
54.4 4
0.39 9
16% %
42.7 7
48.9 9
56.0 0
54.8 8
0.40 0
17% %
43.3 3
48.0 0
56.2 2
57.9 9
0.51 1
27% %
43.1 1
49.7 7
56.1 1
58.7 7
0.49 9
25% %
1990:1990: depressive symptoms
41.6 6
43.7 7
45.3 3
44.8 8
0.15 5
3% %
40.7 7
43.0 0
44.6 6
45.8 8
0.16 6
4% %
1995:1995: depressive symptoms
42.4 4
43.8 8
50.0 0
51.6 6
0.39 9
15% %
42.6 6
45.1 1
51.0 0
56.6 6
0.42 2
18% %
46.0 0
50.8 8
51.2 2
55.7 7
0.34 4
11% %
45.9 9
50.5 5
55.6 6
60.4 4
0.40 0
16% %
47.8 8
49.3 3
50.1 1
60.0 0
0.44 4
22% %
47.5 5
48.8 8
54.8 8
64.4 4
0.53 3
29% %
75 5
ChapterChapter 6
Tablee 6.3 The effect of hierarchic severity level of disability on depressive symptoms in 1995, given a total
numberr of disabilities {=sum score), adjusted for widowhood and country.
Disabilityy severity level
Mildd disability
Moderatee disability
Severee disability
pp for trend
22 disabilities
(nn = 94)
49.6 6
53.7 7
0.09 9
Depressivee symptoms
33 disabilities
(nn = 57)
49.1 1
53.5 5
56.1 1
0.07 7
4-66 disabilities
(nn = 52)
52.2 2
56.7 7
0.11 1
** mild disability - in instrumental activities of daily living (3 items); moderate disability - in instrumental activities, and mobilityy (4 items); severe disability - in mobility, instrumental, and basic activities of daily living (6 items).
Tablee 6.4 Mean level of depressive symptoms by preceding change in disability, stratified by current level of
disabilityy status (1995), adjusted for depression score in 1990, becoming widowed, and country.
Disabilityy change
1990-1995 5
Severee worsening
Moderatee worsening
Stable e
Improving g
pp for trend
noo disability
(n=325) )
44.7 7
41.5 5
0.006 6
Currentt disability
mildd disability
(n=204) )
48.5 5
45.6 6
43.7 7
0.004 4
statuss (1995)
moderatee disability
(n=102) )
54.5 5
52.2 2
52.4 4
49.2 2
0.20 0
severee disability
(n=87) )
59.9 9
60.7 7
60.6 6
0.77 7
Chang ee in disabilit y and depressiv e symptom s
Thee analyses on the effect of change in disability were stratified by current disability status and
showedd that change in disability was predictive of depressive symptoms (table 6.4). The level of
depressivee symptoms in 1995 was 3 points higher among men who developed mild disability
duringg the past 5 years than among those who already had this disability in 1990. Furthermore,
menn who improved toward the status of mild disability during the past five years scored 2 points
lowerr level on depressive symptoms compared to those who remained stable. The p-value for the
trendd was statistically significant (p=0.004). Among men with moderate disability in 1995, the same
trendd was observed, although statistically significance was not reached (p=0.20). For men who
reportedd severe disability in 1995, depressive symptoms did not depend on preceding change in
disabilityy (p=0.77).
Furthermore,, current disability status had the highest effect, because men with worse current
disabilityy status had higher levels of depressive symptoms, independent of preceding change.
76 6
DisabilityDisability and depressive symptoms
Additionn of the interaction between country and change in disabilities to the model did not show
differencess in effects of change on depressive symptoms among countries (p-values for
interaction:: 0.70-0.81).
Discussio n n
Thiss study among men aged 70 to 89 years at baseline from Finland, the Netherlands, and Italy
showedd that both the hierarchic severity levels and the sum score of disability were associated with
depressivee symptoms. The strength of the association with hierarchic severity levels seemed to be
slightlyy higher than that with the sum score. The results testing the hypothesis that the hierarchic
severityy level of disability was a predictor of depressive symptoms independent of the sum score of
disabilityy were borderline statistically significant. Men with a worsening of disability status in the
previouss 5 years had more depressive symptoms than those who remained stable or improved,
whichh is in accordance with our hypothesis.
Somee methodological remarks need to be made. Although standardised questionnaires were used
inn the different countries, some differences among countries caused by translation or interpretation
off the items might still exist. Furthermore, the assessment of disability might be influenced by the
presencee of depressive symptoms, because self-reports are subject to individual's emotional
states.(132)) In the interpretation of the results, possible overestimation of the association between
disabilityy and depression must therefore be taken into account. However, the longitudinal analyses
inn which changes in disability status were investigated, sustained the cross-sectional observations.
Earlierr reports showed that in the Netherlands and Italy, non-respondents had more severe
disabilitiess than respondents.(55;56) In the present study, selection bias might also have occurred
becausee of exclusion of men with missing values. Men excluded from this study were 1 year older,
reportedd more disabilities and scored 5 points higher on the scale of depressive symptoms.
Furthermore,, exploration of the data showed that the association of disability and depressive
symptomss was somewhat stronger among men who were excluded than among men who were
included,, which might have led to underestimation of the observed association.
Depressionn score was highest in Italy and lowest in the Netherlands, which was in accordance with
ann earlier study in the same countries.*149' Another study reported higher levels of depressive
symptomss in southern Europe (Spain) compared to northern Europe (Finland, Sweden).048'
Althoughh depressive symptoms were shown to be associated with disabilities in the three
countries,, these disabilities did not fully explain the observed differences in depressive symptoms:
Afterr adjustment for disability status, Italian men still reported most depressive symptoms and
Dutchh men the fewest.
77 77
ChapterChapter 6
Inn an earlier report on the FINE Study, it was shown that independent of an objective measurement
off physical functioning, Dutch men reported more disabilities than Finnish and Italian men.(70) The
presentt results showed that given a level of disability, Dutch men reported the fewest depressive
symptoms.. Speculating on these observations we suggest that Dutch men perceive more
disabilitiess in daily living, but do not express unpleasant personal feelings and emotions about their
functioning.. These results suggest that cultural differences in perception and report of physical and
mentall functioning hamper cross-national comparisons of prevalence rates.
Inn addition to the results of our and other studies showing that the sum score of disabilities was a
predictorr of depressive symptoms, the present study showed that given the number of disabilities,
thee hierarchic severity level of disability also tended to be predictive. The hierarchic severity level
itselff was a strong predictor of depressive symptoms and seemed to be an even stronger predictor
thann the sum score. Disability in the more severe disability domains might be associated with more
feelingss of worthlessness or hopelessness, because men become more dependent on others.
Earlierr studies investigating different levels of disability (instrumental and basic activities)
separatelyy observed a strong association between disabilities in the instrumental activities and
depressivee symptoms, but not with disabilities in the basic activities,031133' which is in contrast with
thee results of the present study. The lack of a significant association with basic activities in these
studies,, however, might be because of the younger age (mean 70) and lower prevalence of
disabilitiess in basic activities. Prince et al.(134) determined depression scores per disability item and
observedd a general correspondence between depressive symptoms and the disadvantages
associatedd with the disability. For example, people with disability in climbing stairs had fewer
depressivee symptoms than those with disabilities in feeding, which is in accordance with our
results.. Berkman et al.(151> constructed a disability scale with different levels (e.g. physical
performance,, mobility, basic activities). Men with severe disability (in basic activities) reported
moree depressive symptoms than men with moderate disability (in mobility). In contrast with their
expectation,, men with impairment in physical performance (for example stooping or reaching) did
nott report fewer depressive symptoms than men with mobility disability. The investigators
suggestedd that there is no association between disability and depression at levels of such minor
disability.. The results from the present study showed that our categorisation of disability severity
wass reflected in the depressive symptoms, and suggest that the type of disability has slightly more
influencee on depressive symptoms than the number of disabilities.
Inn addition to current disability status, depressive symptoms were dependent on preceding change
inn disability. Our results suggest that the development of disability is a disruptive experience that
requiress readjustment and is therefore associated with more depressive symptoms. The results
78 8
DisabilityDisability and depressive symptoms
furthermoree suggest that in the course of time, men get used to their disability status, and therefore
reportt fewer depressive symptoms when disability has already been present for 5 years - the so-
calledd response shift.(11) This refers to a change in the meaning of one's self-evaluation of health
aspectss as a result of a change in the respondent's values, dependent on social expectations. It
seemss that this shift was not present in men with disability in basic activities, which suggests that
whenn disability is severe, depressive symptoms are determined by the severity of the current
disabilityy status, and not by preceding change. Furthermore, for all levels of disability severity it
wass shown that current disability status was a stronger predictor of depressive symptoms than
precedingg change in disability status.
Inn the present study, the effect of disability on depression could also be dependent on factors that
weree not taken into account. An earlier study showed that more subjective measures, e.g.
subjectivee health, are more related to depressive symptoms than disabilities/152' Furthermore, the
presencee of and satisfaction with social support were known to influence the effect of disabilities on
depression.*1321153» »
Ourr population consisted of male participants aged 70 years and older who were relatively healthy.
Att baseline, less than 10% reported severe disabilities and only 9% could be classified as at least
moderatee depressive (cut off point of 60 on the SDS).<150) This might because of non-response,
exclusionn of men with missing values, and men who died between 1990 and 1995. It is not
possiblee to generalise the results of the present study to the general population, i.e. women and
youngerr men. Women are known to have more depressive symptoms compared to men,(154) but
theyy are less susceptible to depressive symptoms when suffering physical health problems.(152)
Thee association between disability and depressive symptoms might therefore be somewhat weaker
inn women. Furthermore, the results probably do not hold for men younger than 70. An earlier study
showedd that the effect of disability on depressive symptoms was stronger among old-old men (75+)
thann among young-old men (55-64).<152>
Fromm the results of the present study we conclude that hierarchic disability severity and the
disabilityy sum score are major predictors for depressive symptoms. In identifying men who are at
higherr risk of depressive symptoms, preceding changes in disability should also be taken into
account. .
79 9
7 7 Disability ,, self-rate d health , depressiv e
symptom ss and mortalit y
PublishedPublished as: van den Brink CL., Tijhuis MAR. , van den Bos G.A.M., Giampaoli S., Nissinen A., Kromhout
D.. The contribution of self-rated health and depressive symptoms to disability severity as predictor of 10-year
mortalityy in European elderly men.
AmericanAmerican Journal of Public Health 2005; 95(11):2029-2034
ReprintedReprinted with permission from the American Public Health Association
ChapterChapter 7
Abstrac t t Objectiv ee To investigate the effect of disability severity and the contribution of self-rated health
andd depressive symptoms to 10-year mortality.
Method ss Longitudinal data was collected of 1,141 men aged 70-89 years of the Finland, Italy, and
thee Netherlands Elderly Study from 1990 to 2000. Disability severity was classified into four
categories:: no disability, instrumental activities, mobility, and basic activities of daily living. Self-
ratedd health and depressive symptoms were classified into two and three categories respectively.
Multivariatee Cox proportional hazard models were used to calculate mortality risks.
Result ss Men with severe disability had a more than twofold (2.41; 95% confidence interval 1.84-
3.16)) higher risk of mortality than men without disability. Men who had severe disability and did not
feell healthy had the highest mortality risk (HR: 3.30; 95% CI: 2.52, 4.33). This risk was lower at
lowerr levels of disability and higher levels of self-rated health. The same trend was observed for
depressivee symptoms.
Conclusio nn For adequate prognoses on mortality or for developing intervention strategies, not
onlyy physical aspects of health, but also other health outcomes should be taken into account.
82 2
DisabilityDisability and mortality
Introductio n n Thee prediction of mortality in elderly people is a subject with a huge body of knowledge. There is
littlee debate about the importance of functional disability as a predictor of mortality. However, there
aree still unresolved issues in the pathway from disability to mortality, that are important for
enhancingg insight into long-term prognosis, planning health care facilities or for developing
interventionn strategies.
First,, disabilities in different domains, i.e. in instrumental activities/155"158' in mobility,059' and in
basicc activities*95160"163' are known to be associated with mortality risk. These earlier studies are
restrictedd to only one of the disability domains. These domains reflect differences in severity levels
off disability, but the relative impact of these domains on mortality is unknown. An earlier study that
incorporatedd disability in both mobility and basic activities reported that men with disability in basic
activitiess and mobility had a higher risk of mortality than those with disability in mobility only.(84) In
anotherr study about the association between disability and mortality, it was recommended to use
differentt severity levels of disability, e.g. instrumental activities, mobility, and basic activities, as
predictorss of mortality.*161' Although it seems plausible that mortality risk increases with the severity
levell of the disability, no earlier study incorporated the three severity levels in one classification,
andd it is not known whether there is a gradual or exponential increase in risk. In earlier studies in
whichh disability severity was classified in instrumental activities, mobility, and basic activities,
disabilityy severity was strongly associated with other health outcomes, such as performance-based
functionall limitations and chronic diseases.064165'
Inn addition to physical aspects of health, subjective aspects also may play a role in the association
withh mortality. From earlier research it is known that factors such as self-rated health and
depressivee symptoms are associated with disability<24;25;166;167> as well as with mortality/34155168"170'
However,, it is unclear how the combination of disability and more subjective health aspects
contributess to the mortality risk. A person's actual health and mood probably contribute to the
mortalityy risk besides disability.
Thee aim of the present study was to investigate severity levels of disability as predictor of mortality.
Furthermore,, we assessed how different combinations of levels of disability and self-rated health,
andd levels of disability and depressive symptoms contributed to mortality during a 10-year follow-
upp period. We had the opportunity to investigate the different associations in three European
countries,, i.e. Finland, the Netherlands, and Italy.
83 3
ChapterChapter 7
Method s s Stud yy populatio n
Thee present study has a longitudinal design and used data of the Finland, Italy, and the
Netherlandss Elderly (FINE) Study, collected in 1989, 1991, and 1990, respectively, with a mortality
follow-upp up to 2000. The FINE Study started in 1985 as a continuation of the Seven Countries
Study,<171)) focusing on elderly men, bom between 1900 and 1920. In 1985, there were 716
participantss from Finland, 887 from the Netherlands, and 682 from Italy.
Aroundd 1990, 1,416 men were examined {response rates: Finland, 90% of 523 survivors; the
Netherlands,, 78% of 718; Italy, 79% of 493). Six percent of the men were removed because of
missingg values on disability, 6% because of missing values on self-rated health, 6% because of
missingg values on depressive symptoms, and 2% because of missing values on both. In total,
1,1411 men were left for the analyses.
Inn 1985 in Finland, the research was approved by the Ethics Committee of the Kuopio University
Hospitall and in the Netherlands by the Medical Ethics Committee of the University of Leiden. In
Italyy an ethical committee at the local level approved the research. More information about the
FINEE Study and its populations has been reported elsewhere.(149)
Disabilit y y
Disabilityy was measured by a standardised questionnaire about daily routine activities.072' Three
domainss were assessed:
•• instrumental activities of daily living (3 items): preparing meals, doing light housework, and
doingg heavy housework;
•• mobility (4 items): moving outdoors, using stairs, walking 400 meters, carrying a heavy object
1000 meters;
•• basic activities of daily living (6 items): walking indoors, getting in and out of bed, using toilet,
washingg and bathing, dressing and undressing, and feeding oneself.
Thee participants were classified as being disabled on a certain item if they reported a need for help
orr were not able to perform that activity. Disability in a domain was defined as disability in at least
onee item of the domain. The domains were found to be hierarchically ordered.<172) Men who were
disabledd in basic activities were also disabled in mobility and instrumental activities of daily living.
Menn who were disabled in mobility were also disabled in instrumental activities. The following
severityy levels of disability status were distinguished: 0) no disability, 1) mild disability: disability in
instrumentall activities only, 2) moderate disability: disability in instrumental activities and mobility,
3)) severe disability: disability in instrumental activities, mobility, and basic activities of daily living.
Almostt 3% of the men did not fit the hierarchy. Fourteen men who reported disabilities in mobility,
84 4
DisabilityDisability and mortality
butt not in IADL, were classified in category 2. Nineteen men who reported to need help with IADL
andd BADL, but not with mobility were classified in category 3.
Self-rate dd healt h
GlobalGlobal self-rated health was assessed with a single-item question: 'We would like to know what
youu think about your health', with four answer categories: 1) healthy, 2) rather healthy, 3)
moderatelyy healthy, 4) not healthy. For the analyses self-rated health was dichotomised as healthy
andd not healthy, by combining category 1 with 2 and category 3 with 4.
Depressiv ee symptom s
Depressivee symptoms were measured by the Self-rating Depression Scale (SDS) developed by
Zung.(173)) The questionnaire consisted of 20 items developed from clinical diagnostic criteria most
commonlyy used to characterise depressive disorders in terms of mood and biological and
psychologicall disturbances. The items were scored from 1 to 4 on frequency of occurrence of the
symptoms.. An index for the Self-rating Depression Scale was derived by dividing the sum of the
itemss score by 80 and multiplying it by 100, resulting in a range from 25 to 100.
Chroni cc disease s
Informationn on prevalence of chronic diseases was collected for the following chronic conditions:
Inn addition, self-rated health was predictive of mortality (table 7.2). After addition of disability
severityy and depressive symptoms to the model, the mortality risk in the not healthy category was
23%% higher than in the healthy category. Addition of the prevalence of chronic diseases into the
modell slightly decreased the risk of self-rated health on mortality to 19%.
Forr depressive symptoms a similar trend was observed (table 7.2). Men in the highest fertile of
depressivee symptoms had a 4 2 % higher mortality risk than men in the lowest tertile, after
adjustmentt for disability, self-rated health, and prevalence of chronic diseases.
Tablee 7.2 Disability severity, self-rated health and depressive symptoms as predictors of 10-year mortality,
adjustedd for age and country.
Unadjustedd for thee other health
outcomes' '
Adjustedd for thee other health
outcomes* * HRR (95% CI) HRR (95% CI)
Adjustedd for other health outcomess and chronic
diseases* * HRR (95% CI)
Disabilityy severity
noo disability
mildd disability
moderatee disability
severee disability
Self-ratedd health
healthy y
nott healthy
Depressivee symptoms
lowestt tertile
middlee tertile
highestt tertile
1.00 0
1.34(1.11,, 1.61)
2.45(1.90,3.15) )
3.022 (2.34, 3.89)
1.00 0
1.63(1.35,1.98) )
1.00 0
1.29(1.05,, 1.59)
1.90(1.56,2.32) )
1.00 0
1.24(1.03,1.50) )
2.22(1.72,2.87) )
2.411 (1.84,3.16)
1.00 0
1.23(1.01,, 1.51)
1.00 0
1.19(0.97,, 1.46)
1.44(1.15,, 1.79)
1.00 0
1.18(0.98,1.43) )
2.06(1.59,2.67) )
2.28(1.74,3.00) )
1.00 0
1.19(0.97,, 1.46)
1.00 0
1.17(0.95,1.44) )
1.42(1.14,, 1.77)
Abbreviation:: CI, confidence interval. ** disability severity, self-rated health and depressive symptoms in three different models, tt independent association: disability severity, self-rated health and depressive symptoms tt independent association: disability severity, self-rated health and depressive symptoms adjustedd for chronic diseases.
togetherr in one model, togetherr in one model,
88 8
DisabilityDisability and mortality
severe e m i |dd disability
noo severity
self-ratedd health
Figur ee 7.1 Mortality risk3 for six different combinations of disability severity and self-rated health aa hazard ratio on z-axis, with men with no disability who felt healthy as reference group, adjusted for age, country, and chronicc diseases
re re
e e o o E E Lm Lm
re re
severe e mildd disabilit y
noo severit y
tertile ss depressiv e symptom s s
Figur ee 7.2 Mortality risk3 for nine different combinations of disability severity and depressive symptoms. 33 hazard ratio on z-axis, with men with no disability who felt healthy as reference group, adjusted for age, country, and chronicc diseases.
89 9
ChapterChapter 7
Combinatio nn of disabilit y and self-rate d healt h
Whenn participants were classified into six groups on the basis of categories of disability severity
andd self-rated health, mortality risks varied markedly (figure 7.1). The highest risk was observed
amongg those who had severe disability and did not feel healthy (HR: 3.30; 95% CI: 2.52, 4.33).
Thiss risk decreased with lower disability levels and with higher level of self-rated health. Among
thosee with mild or severe disability, (borderline) significant associations between self-rated health
andd mortality were noted (figure 7.1). Men withh mild disability who did not feel healthy had a hazard
ratioo of 1.36 (95% CI: 0.95, 1.93) compared with those who felt healthy. Among men with severe
disability,, those who did not feel healthy had a 45% higher risk (95% CI: 1.02, 2.04) than those
whoo felt healthy.
Thee p-value for interaction between disability and self-rated health was 0.11.
Combinationn of disability and depressive symptoms
Menn with severe disability in the two highest tertiles of depressive symptoms had a threefold higher
riskk of mortality compared with men without disability in the lowest tertile of depressive symptoms
(figuree 7.2). This risk decreased with lower levels of disability severity and was also lower among
menn with severe disability in the lowest tertile of depressive symptoms (HR: 2.16; 95% CI: 1.33,
3.51). .
Inn the lowest levels of disability severity, dose-response relationships between depressive
symptomss and mortality were found. Among men with no disability, those in the highest tertile of
depressivee symptoms had a hazard ratio of 1.62 (95% CI: 1.19, 2.20) with the lowest tertile as
referencee group. Men with mild disability in the highest tertile of depressive symptoms had a 59%
higherr mortality risk (95% CI: 1.11, 2.30) than those in the lowest tertile.
Thee p-value for interaction between disability and depressive symptoms was 0.34.
Discussio n n
Thee present study was designed to investigate disability severity and its combination with self-rated
healthh and depressive symptoms as risk factors for mortality among men aged 70 to 89 years at
baseline,, from Finland, the Netherlands, and Italy. The results showed that severity levels of
disability,, self-rated health, and depressive symptoms were independent predictors of 10-year
mortality.. The combinations of disability with self-rated health or with depressive symptoms were
stronglyy associated with mortality. For several levels of disability, dose-response relationships
betweenn self-rated health or depressive symptoms and mortality were observed.
90 0
DisabilityDisability and modality
Theree was a selection on health caused by non-response and by removing men with missing
values.. Men removed because of missing values on self-rated health or depressive symptoms had
moree disabilities than men included in the study. Furthermore, men removed because of missing
valuess for disability, self-rated health, depressive symptoms, or who were nonrespondents, had a
higherr mortality rate than men who were included. The exclusions might have led to
underestimationn of the strength of the associations.
Ass far as we know, there are few investigations of the association between different disability
domainss and mortality. Bernard et al. investigated the three domains separate from each other and
foundd an association between disability and mortality only for the instrumental and basic
activities.074)) Khokhar et al. showed that men with disabilities in basic activities and mobility
{severee disability) had a higher risk of mortality than men with disabilities in mobility only (moderate
disability)/84** These results were in accordance with our findings. Our study, using one additional
disabilityy domain (instrumental activities), showed that the classification of disability severity on the
basiss of three domains, i.e. instrumental activities, mobility, and basic activities of daily living, was
aa strong predictor of mortality risk. According to an earlier study, concerning mobility disability, self-
reportedd disability even predicted mortality as well as more objective measurements of disability,
suchh as gait speed.(92)
Afterr adjustment for self-rated health and depressive symptoms, disability was still a significant
predictorr of mortality. In accordance with earlier studies, also self-rated health(34174~176) and
depressivee symptoms(155) were, independently of disability, associated with mortality. Disability,
however,, had a stronger association with mortality than self-rated health and depressive symptoms
inn the present study. In contrast, earlier studies showed the strongest association for self-rated
healthh and mortality.<174;175> These divergent findings might have been caused by differences in the
measurementt or distribution of self-rated health. Furthermore, these studies had a shorter follow-
upp period (3-5 years), and self-rated health(96) and depressive symptoms(155) are known to be better
predictorss of mortality in short-term studies (3-5 years). However, it is also possible that our
classificationn of disability, encompassing several domains, was a better predictor of mortality than
thatt in the other studies, in which the disability domains were investigated separately. In addition,
thee present study was restricted to male subjects from three European countries, while the other
studiess included both sexes and also other countries.
First,, although there were differences in disability, self-rated health, and depressive symptoms
betweenn the three countries, the associations between these health aspects and mortality were not
differentt and can therefore be generalised to European men. Second, women have a higher risk of
disabilityy than men<161) and a lower risk of mortality.055' Furthermore, depression seemed to be
associatedd with a higher mortality risk in men than in women.(100) Some studies found that self-
91 1
ChapterChapter 7
ratedd health is a better predictor of mortality in men(175;177178) and other studies have found the
reverse.(162)) Considering these results, it is not justified to generalise our results to women.
Wee assumed that disability reflects the consequences of several underlying diseases and might
thereforee be associated with mortality. The prevalence of chronic diseases was associated with
bothh disability and mortality, and may therefore confound the association between disability and
mortality.. It is also possible that adjustment for these disease removes the association between
disabilityy and mortality. We therefore reported the associations of disability, self-rated health and
depressivee symptoms with mortality, both adjusted and unadjusted for the prevalence of chronic
diseases.. The strength of the association between disability and mortality only slightly decreased
afterr adjustment, which shows that disability is associated with mortality also independent of these
chronicc diseases. Disability reflects more aspects of overall health and functioning than the
disabilityy impact of chronic diseases alone.
Thee interaction terms between disability and self-rated health or depressive symptoms were not
statisticallyy significant, which means that the associations of self-rated health and depressive
symptomss with mortality did not differ among the levels of disability severity. At the lower levels of
disabilityy however, clear dose-response relationships were observed between depressive
symptomss and mortality, that were not present at the most severe disability level. Nevertheless,
menn with severe disability had a high mortality risk, and the small numbers of men in these
differentt categories made it difficult to interpret the findings. In contrast, self-rated health was
associatedd with mortality only in the higher levels of disability. These results suggest that positive
healthh perceptions (self-rated health) and less depressive symptoms may postpone mortality.
Knowledgee about risk factors for mortality in old age is important for enhancing insight into
prognosis,, planning long-term facilities and developing intervention strategies. Firstly, disability is
ann important risk factor. Men with only mild disability have an increased mortality risk, and further
deteriorationn of disability should be encountered effectively to prevent a much higher risk.
Secondly,, although depressive symptoms are often unrecognised/179' depression as well as self-
ratedd health are both important. Intervention strategies should therefore not only focus on
preventionn of deterioration of physical disability, but also on reinforcing mental functioning.
Interventionss on disability, self-rated health, and depressive symptoms will not only postpone
mortality,, but will also improve quality of life.
Fromm the results of the present study we conclude that in elderly men the risk of mortality increases
withh severity level of disability. Furthermore, self-rated health and depressive symptoms increase
thee mortality risk at different levels of disability. This knowledge may be helpful for enhancing
insightt into long-term prognosis, planning health care facilities, and developing intervention
strategies. .
92 2
8 8 Genera ll discussio n
ChapterChapter 8
Thiss thesis addresses functional disability among European elderly men. We aimed to identify risk
groupss and risk factors for functional disability and to quantify health (care) impacts of functional
disability.. Firstly, the assessment of disability was validated by relating self-reported disability to
performance-basedd functioning, and by comparisons between three different countries. Thereafter,
wee focused on the main research questions of this thesis. Widowhood was studied as a risk group
forr disability, and physical activity as risk factor. We investigated whether use of formal home care
wass according to disability-related needs. The health impact of disability was investigated by
determiningg its effect on depressive symptoms and on mortality.
Thee data for our study came from the Finland, Italy and the Netherlands Elderly (FINE) Study, that
consistss of 2285 men born between 1900 and 1920. Four surveys were carried out in 1984-1985,
1989-1991,, 1994-1995, and 1999-2000.
Wee focused on functional disability: dependency in instrumental activities of daily living, mobility, or
basicc activities of daily living. Functional disability is a growing public health problem, related to the
increasee in the number of elderly people and the ensuing rise in prevalence of chronic diseases.
Disabilityy seems to be an irreversible process for the individual. Although previous research has
shownn that disability is a dynamic process in which people can recover,(180) this recovery is often
short-lasting,, because people who recover are at high risk for recurrent disability/181* Particularly at
oldd age, people hardly recover from disability/182183' In order to anticipate the growing burden for
thee individual as well as society as a whole, more insight is needed into different aspects of
disabilityy among elderly people.
Inn this chapter the main findings of the thesis are summarised. Methodological considerations are
discussedd and the implications for public health and health care, such as preventive strategies and
consequencess for care delivery, are addressed.
Mainn finding s
Inn table 8.1 the main findings for each chapter described in this thesis are summarised.
Assessmen tt of disabilit y (chapter 2)
Self-reportedd disability is often used to compare health status between countries. We examined
whetherr there is cross-cultural variation in the self-report of disability, independent of differences in
physicall functioning. The association of self-reported disability with performance tests was
2.. Omran AR. The epidemiologic transition. A theory of the epidemiology of population change. Milbank Memm Fund Q 1971;49(4):509-38.
3.. Imhoff EvN. Wat zijn de belangrijkste verwachtingen voor de toekomst? In: Volksgezondheid Toekomstt Verkenning, Nationaal Kompas Volksgezondheid. Bilthoven: RIVM, <http://www.nationaalkompas.nl>> Bevolking, 14 februari 2003.
4.. Caldwell JC. Population health in transition. Bull World Health Organ 2001 ;79{2):159-60.
5.. Perenboom RJM, Mulder YM, Herten LMv, Oudshoorn K, Hoeymans N. Trends in gezonde levensverwachting:: Nederland 1983-2000. Leiden: TNO Preventie En Gezondheid; 2002.
6.. Centraal Bureau voor de Statistiek. Prognose-intervallen van de bevolking naar leeftijdsgroep. http://statline.cbs.nll (bezocht 6 mei 2005).
7.. Verbrugge LM, Jette AM. The disablement process. Soc Sci Med 1994;38(1):1-14.
8.. Hoeymans N, Feskens EJ, van den Bos GAM, Kromhout D. Measuring functional status: cross-sectionall and longitudinal associations between performance and self-report (Zutphen Elderly Study 1990-1993).. J Clin Epidemiol 1996;49(10):1103-10.
9.. Angel R, Thoits P. The impact of culture on the cognitive structure of illness. Cult Med Psychiatry 1987;11(4):465-94. .
10.. Jee M, Or Z. Health outcomes in OECD countries: a framework of health indicators for outcome-orientedd policymaking. Paris: Organisation for Economic Co-Operation and Development; 1999.
11.. Daltroy LH, Larson MG, Eaton HM, Phillips CB, Liang MH. Discrepancies between self-reported and observedd physical function in the elderly: the influence of response shift and other factors. Soc Sci Medd 1999;48( 11): 1549-61.
12.. Kivinen P, Sulkava R, Halonen P, Nissinen A. Self-reported and performance-based functional status andd associated factors among elderly men: the Finnish cohorts of the Seven Countries Study. J Clin Epidemioll 1998;51 (12): 1243-52.
13.. Guralnik JM, Simonsick EM, Ferrucci L, Glynn RJ, Berkman LF, Blazer DG, Scherr PA, Wallace RB. A shortt physical performance battery assessing lower extremity function: association with self-reported disabilityy and prediction of mortality and nursing home admission. Journal of Gerontol A Med Sci 1994;49(2):M85-M94. .
14.. Centraal Bureau voor de Statistiek. Bevolking per regio naar leeftijd, geslacht en burgerlijke staat. http://statline.cbs.nll (bezocht 6 mei 2005).
15.. Maas IAM, Gijsen R, Lobbezoo IE, Poos MJJC. Volksgezondheid Toekomst Verkenning 1997. 1. De gezondheidstoestand:: een actualisering. Maarssen: RIVM / Elsevier/ De Tijdstroom; 1997.
16.. Wisocki PA, Skowron J. The effects of gender and culture on adjustment to widowhood. Handbook of
17.. Goldman N, Korenman S, Weinstein R. Marital status and health among the elderly. Soc Sci Med 1995;40(12):: 1717-30.
18.. Mendes de Leon CF, Kasl SV, Jacobs S. Widowhood and mortality risk in a community sample of the elderly:: a prospective study. J Clin Epidemiol 1993;46(6):519-27.
19.. Deeg DJH. Sex differences in IADL in the Netherlands: functional and situational disability. In Robine JM,, Mathers CD, Bone MR Eds. Calculation of Health Expectancies: Harmonization, Consensus Achievedd and Future Perspectives. Montrouge, France: Colleque INSERM/John Libbey Eurotext Ltd; 1993:p.. 203-13.
20.. U.S. Department of Health and Human Services. Physical Activity and Health: A Report of the Surgeon General.. Atlanta, GA: U.S. Department of Health and Human Services, Centers for Disease Control andd Prevention, National Center for Chronic Disease Prevention and Health Promotion; 1996.
21.. Spirduso WW, Cronin DL. Exercise dose-response effects on quality of life and independent living in olderr adults. Med Sci Sports Exerc 2001 ;33(6 Suppl):S598-S608.
22.. Tanasescu M, Leitzmann MF, Rimm EB, Willett WC, Stampfer MJ , Hu FB. Exercise type and intensity inn relation to coronary heart disease in men. JAMA 2002;288{16): 1994-2000.
23.. Andersen R, Aday LA. Access to medical care in the U.S.: realized and potential. Med Care 1978;16(7):533-46. .
24.. Bruce ML. Depression and disability in late life: directions for future research. Am J Geriatr Psychiatry 20011 ;9{2):102-12.
25.. Cole MG, Dendukuri N. Risk factors for depression among elderly community subjects: a systematic revieww and meta-analysis. Am J Psychiatry 2003; 160(6): 1147-56.
26.. Stek ML, Gussekloo J, Beekman AT, van Tilburg W, Westendorp RG. Prevalence, correlates and recognitionn of depression in the oldest old: the Leiden 85-plus study. J Affect Disord 2004;78(3):193-200. .
27.. Hartman-Stein PE, Potkanowicz ES. Behavioral determinants of healthy aging: good news for the baby boomerr generation. Online J Issues Nurs 2003;8(2):6.
28.. Hirvensalo M, Rantanen T, Heikkinen E. Mobility difficulties and physical activity as predictors of mortalityy and loss of independence in the community-living older population. J Am Geriatr Soc 2000;48(5):493-8. .
29.. Ginsberg GM, Hammerman-Rozenberg R, Cohen A, Stessman J. Independence in instrumental activitiess of daily living and its effect on mortality. Aging (Milano) 1999; 11(3): 161-8.
30.. Wolinsky FD, Callahan CM, Fitzgerald JF, Johnson RJ. Changes in functional status and the risks of subsequentt nursing home placement and death. J Gerontol 1993;48(3):S94-101.
31.. Reuben DB, Rubenstein LV, Hirsch SH, Hays RD. Value of functional status as a predictor of mortality:
109 9
resultss of a prospective study. Am J Med 1992;93(6):663-9.
32.. Ramasubbu R, Patten SB. Effect of depression on stroke morbidity and mortality. Can J Psychiatry 2003;48(4):250-7. .
33.. Schulz R, Drayer RA, Rollman BL. Depression as a risk factor for non-suicide mortality in the elderly. Bioll Psychiatry 2002;52(3):205-25.
34.. Benyamini Y, Idler EL. Community studies reporting association between self-rated health and mortalityy - Additional studies, 1995-1998. Res Aging 1999;21 (3):392-401.
35.. Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. J Healthh Soc Behav 1997;38:21-37.
36.. Keys A, Aravanis C, Blackburn H, van Buchem FSP, Buzina R, Djordjevic BS, Dontas AS, Fidanza F, Karvonenn MJ, Kimura N, et al. Epidemiological studies related to coronary heart disease; characteristicss of men aged 40-59 in seven countries. Acta Med Scan 1967;460(suppl): 1-392.
37.. World Health Organization. World Health Report 2000. Geneva: World Health Organization; 2000.
38.. Nagi S. Disability concepts revisited: implications for prevention. In Pope AM, Tarlov AR Eds. Disability inn America. Toward a National Agenda for Prevention. Washington, DC: National Academy Press; 1991. .
39.. Sager MA, Dunham NC, Schwantes A, Mecum L, Halverson K, Harlowe D. Measurement of activities off daily living in hospitalized elderly: a comparison of self-report and performance-based methods. J Amm Geriatr Soc 1992;40(5):457-62.
40.. Simonsick EM, Kasper JD, Guralnik JM, Bandeen-Roche K, Ferrucci L, Hirsch R, Leveille S, Rantanen T,, Fried LP. Severity of upper and lower extremity functional limitation: scale development and validationn with self-report and performance-based measures of physical function. J Gerontol B Psycholl Sci Soc Sci 2001;56B(1):S10-S19.
41.. Langlois JA, Maggi S, Harris T, Simonsick EM, Ferrucci L, Pavan M, Sartori L, Enzi G. Self-report of difficultyy in performing functional activities identifies a broad range of disability in old age. J Am Geriatr Soc1996;44(12):1421-8. .
42.. Kempen Gl, Steverink N, Ormel J, Deeg DJ. The assessment of ADL among frail elderly in an intervieww survey: self- report versus performance-based tests and determinants of discrepancies. J Gerontoll B Psychol Sci Soc Sci 1996;51B(5):P254-P260.
43.. Bijnen FC, Feskens EJ, Caspersen CJ, Giampaoli S, Nissinen AM, Menotti A, Mosterd WL, Kromhout D.. Physical activity and cardiovascular risk factors among elderly men in Finland, Italy, and the Netherlands.. Am J Epidemiol 1996;143(6):553-61.
44.. World Health Organization. The elderly in eleven countries. Copenhagen: World Health Organization; 1983. .
45.. König-Zahn C, Furer JW, Tax B. Het meten van de gezondheidstoestand. 2 Lichamelijke gezondheid, socialee gezondheid. Assen: Van Gorcum; 1994.
110 0
References References
46.. Cornoni-Huntley J, Brock DB, Ostfeld AM, Taylor JO, Wallace RB . Established populations for epidemiologicc studied of the elderly. Resource data book. New Haven: National Institute on Aging.
47.. Rose GA, Blackburn H. Cardiovascular survey methods. Monogr Ser World Health Organ 1968;56:1-188. .
48.. Sadana R, Mathers CD, Lopez AD, Murray CJL, Iburg K. Comparative analyses of more than 50 householdd surveys on health status. Geneva: World Health Organization; 2000.
49.. Cavelaars AE, Kunst AE, Geurts JJ, Crialesi R, Grotvedt L, Helmert U, Lahelma E, Lundberg O, Mathesonn J, Mielck A, et al. Differences in self reported morbidity by educational level: a comparison off 11 western European countries. J Epidemiol Community Health 1998;52(4):219-27.
50.. Russell K, Jewell N. Cultural impact of health-care access: challenges for improving the health of Africann Americans. J Community Health Nurs 1992;9{3):161-9.
51.. Jylha M, Guralnik JM, Ferrucci L, Jokela J, Heikkinen E. Is self-rated health comparable across culturess and genders? J Gerontol B Psychol Sci Soc Sci 1998;53B(3):S144-S152.
52.. Huijbregts PPCW, Feskens EJM, RSsSnen L Diet, energy intake, and self-rated health in elderly men. InIn Dietary Patterns and Health in the Elderly: a North-South Comparison in Europe [Dissertation]. Den Haag:: CIP-Data, Koninklijke Bibiotheek; 1997: p. 93-107.
53.. Meredith LS, Siu AL. Variation and quality of self-report health data. Asians and Pacific Islanders comparedd with other ethnic groups. Med Care 1995 Nov;33{11):1120-31.
54.. Alonso J, Black C, Norregaard JC, Dunn E, Andersen TF, Espallargues M, Bemth-Petersen P, Andersonn GF. Cross-cultural differences in the reporting of global functional capacity: an example in cataractt patients. Med Care 1998;36(6):868-78.
55.. Hoeymans N, Feskens EJ, Van Den Bos GA, Kromhout D. Non-response bias in a study of cardiovascularr diseases, functional status and self-rated health among elderly men. Age Ageing 1998;27(1):: 35-40.
56.. Giampaoli S, Menotti A. Performance ed autosufficienza nella popolazione italiana. In La Salute DegN Italian!.. Roma: La Nuova Italia Scientifica. 1993: p. 287-96.
57.. Merrill SS, Seeman TE, Kast SV, Berkman LF. Gender differences in the comparison of self-reported disabilityy and performance measures. J Gerontol A Biol Sci Med Sci 1997;52A(1):M19-M26.
58.. Spector WD, Katz S, Murphy JB, Fulton JP. The hierarchical relationship between activities of daily livingg and instrumental activities of daily living. J Chronic Dis 1987;40(6):481-9.
59.. World Health Organization. The World Health Organization Collection on Long-Term Care. Current andd Future Long-Term Care Needs. 2002.
60.. Guralnik JM, Alecxih L, Branch LG, Wiener JM. Medical and long-term care costs when older persons becomee more dependent. Am J Public Health 2002;92(8): 1244-5.
61.. Central Bureau of Statistics. [Increasing incidence of widowhood among men]. Voorburg/Heerlen, the
111 1
Netherlands:: CBS; 2002.
62.. Joung IMA, Stronks K, van de Mheen H, van Poppel FWA, van der Meer JBW, Mackenbach JP. The contributionn of intermediary factors to marital status differences in self-reported health. Journal of Marriagee and the Family 1997;59:476-90.
63.. van Poppel F, Joung I. Long-term trends in marital status mortality differences in The Netherlands 1850-1970.. J Biosoc Sci 2001;33(2):279-303.
64.. Ben-Shlomo Y, Smith GD, Shipley M, Marmot MG. Magnitude and causes of mortality differences betweenn married and unmarried men. J Epidemiol Community Health 1993;47{3):200-5.
65.. De Jong Gierveld J, Dykstra PA. Life transitions and the network of personal relationships. Theoretical andd methodological issues. In Duck, S. Ed. Advances in Personal Relationships. Jessica Kingsley Publisherss Ltd; 1993. p. 195-227.
66.. Antonucci TC, Lansford JE, Schaberg L , Baltes M, Takahashi K, Dartigues JF, Smith J, Akiyama H, Fuhrerr R. Widowhood and illness: a comparison of social network characteristics in France, Germany, Japan,, and the United States. Psychol Aging 2001;16(4):655-65.
67.. O'Bryant SL, Hansson RO. Widowhood. In R. Blieszner & V.H. Bedford Eds. Handbook of Aging and thee Family. Westport: CT: Greenwood Press; 1995: p. 440-59.
68.. Jacobzone, S. Ageing and care for frail elderly persons: an overview of international perspectives. Paris:: Organisation for economic co-operation and development; 1999.
69.. Sociaal en Cultureel Planbureau. Sociaal en Cultureel Rapport 2000. Nederland in Europa. The Hague,, the Netherlands: Sociaal en Cultureel Planbureau; 2000.
70.. van den Brink CL, Tijhuis M, Kalmijn S, Klazinga NS, Nissinen A, Giampaoli S , Kivinen P, Kromhout D,, van den Bos GA. Self-reported disability and its association with performance-based limitation in elderlyy men: a comparison of three European countries. J Am Geriatr Soc 2003;51(6):782-8.
71.. Glass TA, Kasl SV, Berkman LF. Stressful life events and depressive symptoms among the elderly. J Agingg Health 1997;9(1):70-89.
72.. Umberson D, Wortman CB, Kessler RC. Widowhood and depression: explaining long-term gender differencess in vulnerability. J Health Soc Behav 1992;33(1):10-24.
73.. Billings AG, Moos RH. Stressful life events and symptoms: a longitudinal model. Health Psychol 1982;1:99-117. .
74.. Bieliauskas LA, Counte MA, Glandon GL. Inventorying stressing life events as related to health change inn the elderly. Stress Medicine 1995;11(2):93-103.
75.. Grand A, Grosclaude P, Bocquet H, Pous J, Albarede JL. Predictive value of life events, psychosocial factorss and self-rated health on disability in an elderly rural French population. Soc Sci Med 1988;27(12):1337-42. .
76.. Umberson D. Gender, marital status and the social control of health behavior. Soc Sci Med 1992;34-
112 2
References References
8:907-17. .
77.. Lichtenberg PA, MacNeill SE, Mast BT. Environmental press and adaptation to disability in hospitalizedd iive-alone older adults. Gerontologist 2000 Oct;40{5):549-56.
78.. Speare AJ, Avery R, Lawton L. Disability, residential mobility, and changes in living arrangements. J Gerontoll 1991 ;46(3):S133-42.
79.. Zunzunegui MV, Beland F, Otero A. Support from children, living arrangements, self-rated health and depressivee symptoms of older people in Spain. Int J Epidemiol 2001 ;30(5): 1090-9.
80.. Joung IMA, Mheen Hvd, Stronks K, Poppel FWAv, Mackenbach JP. Differences in self-reported morbidityy by marital status and by living arrangement. Intern J Epidemiol 1994;23(1):91-7.
81.. Fry PS. Predictors of health-related quality of life perspectives, self-esteem, and life satisfactions of olderr adults following spousal loss: an 18- month follow-up study of widows and widowers. Gerontologistt 2001 ;41(6):787-98.
82.. Erlichman J, Kerbey A, James W. Physical activity and its impact on health outcomes. Paper 1: The impactt of physical activity on cardiovascular disease and all-cause mortality: an historical perspective. Obesityy Reviews 2002;3:257-71.
83.. Lee I, Skerrett P. Physical activity and all-cause mortality: what is the dose-response relation? Med Sci Sportss Exerc 2001 ;33(6 Suppl):S459-71.
84.. Bijnen FC, Caspersen CJ, Feskens EJ, Saris WH, Mosterd WL, Kromhout D. Physical activity and 10-yearr mortality from cardiovascular diseases and all causes: The Zutphen Elderly Study. Arch Intern Medd 1998;158(14):1499-505.
85.. Miller ME, Rejeski WJ, Reboussin BA, Ten Have TR, Ettinger WH. Physical activity, functional limitations,, and disability in older adults. J Am Geriatr Soc 2000;48{10):1264-72.
86.. Ferrucci L, Izmirlian G, Leveille S, Phillips CL, Corti MC, Brock DB, Guralnik JM. Smoking, physical activity,, and active life expectancy. Am J Epidemiol 1999;149(7):645-53.
87.. Wu SC, Leu SY, Li CY. Incidence of and predictors for chronic disability in activities of daily living amongg older people in Taiwan. J Am Geriatr Soc 1999;47(9): 1082-6.
88.. Huang Y, Macera CA, Blair SN, Brill PA, Kohl HW 3rd, Kronenfeld JJ. Physical fitness, physical activity,, and functional limitation in adults aged 40 and older. Med Sci Sports Exerc 1998;30(9): 1430-5.
89.. Unger JB, Johnson CA, Marks G. Functional decline in the elderly: evidence for direct and stress-bufferingg protective effects of social interactions and physical activity. Ann Behav Med 1997; 19(2): 152-60. .
91.. Young DR, Masaki KH, Curb JD. Associations of physical activity with performance-based and self-reportedd physical functioning in older men: the Honolulu Heart Program. J Am Geriatr Soc
113 3
1995;43(8):845-54. .
92.. LaCroix AZ, Guralnik JM, Berkman LF, Wallace RB, Satterfield S. Maintaining mobility in late life. II. Smoking,, alcohol consumption, physical activity, and body mass index. Am J Epidemiol 1993;137(8):858-69. .
93.. Avlund K, Lund R, Holstein BE, Due P, Sakari-Rantala R , Heikkinen RL. The impact of structural and functionall characteristics of social relations as determinants of functional decline. J Gerontol B Psychol Scii Soc Sci 2004;59(1 ):S44-S51.
94.. Caspersen CJ, Bloemberg BP, Saris WHM , Merritt RK, Kromhout D. The prevalence of selected physicall activities and their relation with coronary heart disease risk factors in elderly men: the Zutphen Study,, 1985. Am J Epidemiol 1991; 133(11): 1078-92.
95.. Westerterp KR, Saris WHM, Bloemberg BPM, Kempen K, Caspersen CJ, Kromhout D. Validation of thee Zutphen Physical Activity Questionnaire for the Elderly with doubly labeled water. Med. Sci. Sports Exerc.. 1992;24:S68.
96.. Kemper HCG, Ooijendijk WTM, Stiggelbout M. [Consensus about the Dutch physical activity guideline].. Tijdschr Soc Geneeskd 2000;78:180-3.
97.. Pate RR, Pratt M, Blair SN, Haskell WL, Macera CA, Bouchard C, Buchner D, Ettinger W, Heath GW , Kingg AC, et al. Physical activity and public health. A recommendation from the Centers for Disease Controll and Prevention and the American College of Sports Medicine. JAMA 1995;273(5):402-7.
98.. Schoenborn CA, Barnes P.M. Leisure-time physical activity among adults: United States, 1997-98. Advancee data from vital and health statistics; no. 325. Hyattsville, Maryland: National Center for Health Statisticss 2002.
99.. Hildebrandt VH, Ooijendijk WTM, Stiggelbout M. [Trend report on physical activity and health 1998/1999].. Lelystad, NL: Koninklijke Vermande; 1999.
100.. Shephard R. Absolute versus relative intensity of physical activity in a dose-response context. Med Sci Sportss Exerc 2001 ;33(6 Suppl):S400-18.
101.. Schuit AJ, van Loon AJ, Tijhuis M, Ocké M. Clustering of lifestyle risk factors in a general adult population.. Prev Med 2002;35(3):219-24.
102.. Ferraro KF, Su YP, Gretebeck RJ, Black DR, Badylak SF. Body mass index and disability in adulthood:: a 20-year panel study. Am J Public Health 2002;92(5):834-40.
103.. Oslin DW. Alcohol use in late life: disability and comorbidity. J Geriatr Psychiatry Neurol 2000;; 13(3): 134-40.
104.. Lamarca R, Ferrer M, Andersen PK, Liestol K, Keiding N, Alonso J. A changing relationship between disabilityy and survival in the elderly population: differences by age. J Clin Epidemiol 2003;56(12):1192-201. .
105.. Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Socc Behav 1995;36(1):1-10.
114 4
References References
106.. Algera M, Francke AL, Kerkstra A, van der Zee J. Home care needs of patients with long-term conditions:: literature review. J Adv Nurs 2004;46(4):417-29.
107.. van den Bos GA, Smits JP, Westert GP, van Straten A. Socioeconomic variations in the course of stroke:: unequal health outcomes, equal care? J Epidemiol Community Health 2002;56(12):943-8.
108.. Jacobi CE, Triemstra M, Rupp I, Dinant HJ, Van Den Bos GA. Health care utilization among rheumatoidd arthritis patients referred to a rheumatology center: unequal needs, unequal care? Arthritis Rheumm 2001 ;45(4):324-30.
109.. Scholte op Reimer WJM, de Haan RJ, Rijnders PT, Limburg M, van den Bos GAM. Unmet care demandss as perceived by stroke patients: deficits in health care? Qual Health Care 1999;8:30-5.
110.. de Boer AG, Wijker W, de Haes HC. Predictors of health care utilization in the chronically ill: a review off the literature. Health Policy 1997;42{2):101-15.
111.. de Haan R, Limburg M, van der Meulen J, van den Bos GA. Use of health care services after stroke. Quall Health Care 1993;2(4): 222-7.
112.. Kempen Gl, Suurmeijer TP. Professional home care for the elderly: an application of the Andersen-Newmann model in The Netherlands. Soc Sci Med 1991;33(9):1081-9.
113.. Wolinsky FD, Johnson RJ. The use of health services by older adults. J Gerontol 1991 ;46{6):S345-S357. .
114.. Evashwick C, Rowe G, Diehr P, Branch L. Factors explaining the use of health care services by the elderly.. Health Serv Res 1984;19(3):357-82.
115.. Westert GP, Satariano WA, Schellevis FG, van den Bos GA. Patterns of comorbidity and the use of healthh services in the Dutch population. Eur J Public Health 2001;11(4):365-72.
116.. Liu K, Manton KG, Aragon C. Changes in home care use by disabled elderly persons: 1982-1994. J Gerontoll B Psychol Sci Soc Sci 2000;55(4):S245-S253.
117.. Scholte op Reimer WJM, de Haan RJ, Limburg M, van den Bos GAM. Use of long-term health care afterr stroke. In Long-term use after stroke [dissertation]. Amsterdam; 1999.
118.. Stoddart H, Whitley E, Harvey I, Sharp D. What determines the use of home care services by elderly people?? Health Soc Care Community 2002;10{5):348-60.
119.. Portrait F, Lindeboom M, Deeg D. The use of long-term care services by the Dutch elderly. Health Econ2000;9(6):513-31. .
120.. Geerlings SW, Pot AM, Twisk JWR, Deeg DJH. Predicting transitions in the use of informal and professionall care by older adults. Aging and Society 2005;25:111-30.
121.. Cantor MH. Neighbours and friends: an overlooked resource in the informal support system. Res Agingg 1979;1:434-63.
122.. Sussman M. The family life of old people, in Binstock, R. & Shanas, EH Eds. Handbook of Aging and
115 5
thee Social Sciences. New York: Van Nostrand Reinhold, 1976.
123.. Kempen Gl, Suurmeijer TP. Factors influencing professional home care utilization among the elderly. Socc Sci Med 1991 ;32(1):77-81.
124.. Feskens E, Bloemberg B, Pijls L, Kromhout D. A longitudinal study on elderly men: the Zutphen Study. InIn Schroots JJF Ed. Aging, Health and Competence. Amsterdam: Elsevier Science Publishers; 1993.
125.. Berg Jeths A vd, Timmermans JM, Hoeymans N, Woittiez IB. Ouderen nu en in de toekomst. Gezondheid,, verpleging en verzorging 2000-2020. Bilthoven: RIVM/ Bohn Stafleu Van Loghum; 2004.
126.. Algera M, Francke AL, Schreeuwenberg P, Zee Jvd. The match between Dutch chronic patients' felt needd and home care delivered and its determinants. In All You Need Is... Home Care [Dissertation]. Utrecht;; 2005.
127.. Ruo B, Rumsfeld JS, Hlatky MA, Liu H, Browner WS, Whooley MA. Depressive symptoms and health-relatedd quality of life. JAMA 2003;290(2):215-21.
128.. Beekman AT, Penninx BW, Deeg DJ, de Beurs E, Geerling SW , van Tilburg W. The impact of depressionn on the well-being, disability and use of services in older adults: a longitudinal perspective. Actaa Psychiatr Scand 2002; 105(1 ):20-7.
129.. Ganguli M, Dodge HH, Mulsant BH. Rates and predictors of mortality in an aging, rural, community-basedd cohort: the role of depression. Arch Gen Psychiatry 2002;59(11): 1046-52.
130.. Blazer DG, Hybels CF, Pieper CF. The association of depression and mortality in elderly persons: a casee for multiple, independent pathways. J Gerontol A Biol Sci Med Sci 2001;56<8):M505-M509.
131.. Cummings SM, Neff JA, Husaini BA. Functional impairment as a predictor of depressive symptomatology:: the role of race, religiosity, and social support. Health Soc Work 2003;28(1):23-32.
132.. Jang Y, Haley WE, Small BJ, Mortimer J A. The role of mastery and social resources in the associationss between disability and depression in later life. Gerontologist 2002;42(6):807-13.
133.. Steffens DC, Hays JC, Krishman K.R.R. Disability in geriatric depression. Am J Geriatr Psychiatry 1999;7:34-40. .
134.. Prince MJ, Harwood RH, Blizard RA, Thomas A, Mann AH. Impairment, disability and handicap as risk factorss for depression in old age. The Gospel Oak Project V. Psychol Med 1997;27(2):311-21.
135.. Alexopoulos GS, Vrontou C, Kakuma T, Meyers BS, Young RC, Klausner E, Clarkin J . Disability in geriatricc depression. Am J Psychiatry 1996;153(7):877-85.
136.. Forsell Y, Jorm AF, Winblad B. Association of age, sex, cognitive dysfunction, and disability with major depressivee symptoms in an elderly sample. Am J Psychiatry 1994; 151 (11): 1600-4.
137.. Kessler RC, Berglund P, Dernier O, Jin R, Koretz D, Merikangas KR, Rush AJ, Walters EE, Wang PS . Thee epidemiology of major depressive disorder. Results from the National Comorbidity Survey Replicationn (NCS-R). JAMA 2003;289(23):3095-105.
116 6
References References
138.. Ormel J, Rijsdijk FV, Sullivan M, van Sonderen E, Kempen Gl. Temporal and reciprocal relationship betweenn IADL/ADL disability and depressive symptoms in late life. J Gerontol B Psychol Sci Soc Sci 2002;57(4):P338-P347. .
139.. Geerlings SW, Beekman ATF, Deeg DJH, Van Tilburg W. Physical health and the onset and persistencee of depression in older adults: an eight-wave prospective community-based study. Psychol Medd 2000;30:369-80.
140.. Prince MJ, Harwood RH, Thomas A, Mann AH. A prospective population-based cohort study of the effectss of disablement and social milieu on the onset and maintenance of late-life depression. The Gospell Oak Project VII. Psychol Med 1998;28:337-50.
141.. Henderson AS, Korten AE, Jacomb PA, Mackinnon AJ, Jorm AF , Christensen H, Rodgers B. The coursee of depression in the elderly: a longitudinal community-based study in Australia. Psychol Med 1997;27(1):119-29. .
142.142. Zeiss AM, Lewinsohn PM, Rohde P, Seeley JR. Relationship of physical disease and functional impairmentt to depression in older people. Psychol Aging 1996;11(4):572-81.
143.. Beekman ATJ, Deeg DJH, Smit JH, Van Tilburg W. Predicting the course of depression in the older population:: results from a community-based study in The Netherlands. J Affect Disord 1995;34:41-9.
144.. Kennedy GJ, Kelman HR, Thomas C. The emergence of depressive symptoms in late life: the importancee of declining health and increasing disability. J Community Health 1990;15(2):93-104.
145.. Kraaij V, Arensman E, Spinhoven P. Negative life events and depression in elderly persons: a meta-analysis.. J Gerontol B Psychol Sci Soc Sci 2002;57(1):P87-P94.
146.. Lindeboom M, Portrait F, van den Berg GJ. An econometric analysis of the mental-health effects of majorr events in the life of older individuals. Health Econ 2002; 11 (6):505-20.
147.. Simon GE, Goldberg DP, Von Korff M, Ustun TB. Understanding cross-national differences in depressionn prevalence. Psychol Med 2002;32(4):585-94.
148.. Heslin JM, Soveri PJ, Winoy JB, Lyons RA, Buttanshaw AC, Kovacic L, Daley JA, Gonzalo E. Health statuss and service utilisation of older people in different European countries. Scand J Prim Health Care 20011 ;19(4):218-22.
149.. De Leo D, Diekstra RF, Lonnqvist J, Trabucchi M, Cleiren MH, Frisoni GB, Dello Buono M, Haltunen A, Zucchettoo M, Rozzini R, et al. LEIPAD, an internationally applicable instrument to assess quality of life inn the elderly. Behav Med 1998;24(1): 17-27.
150.. Zung WWK. A self-rating depression scale. Arch Gen Psychiatry 1965;12:63-70.
151.. Berkman LF, Berkman CS, Kasl S, Freeman DH Jr, Leo L, Ostfeld AM, Cornoni-Huntley J, Brody JA. Depressivee symptoms in relation to physical health and functioning in the elderly. Am J Epidemiol 1986;124(3):372-88. .
152.152. Beekman AT, Kriegsman DM, Deeg DJ, van Tilburg W. The association of physical health and depressivee symptoms in the older population: age and sex differences. Soc Psychiatry Psychiatr Epidemioll 1995;30{1):32-8.
117 7
References References
153.. Allen SM, Mor V. The prevalence and consequences of unmet need. Contrasts between older and youngerr adults with disability. Med Care 1997;35( 11):1132-48.
154.. Hybels CF, Pieper CF, Blazer DG. Sex differences in the relationship between subthreshold depressionn and mortality in a community sample of older adults. Am J Geriatr Psychiatry 2002; 10 (3):283-91. .
155.. Idler EL, Russell LB, Davis D. Survival, functional limitations, and self-rated health in the NHANES I Epidemiologicc Follow-up Study, 1992. First National Health and Nutrition Examination Survey. Am J Epidemioll 2000;152(9):874-83.
156.. Dorn J, Cemy F, Epstein L, Naughton J, Vena J, Winkelstein V, Schisterman E, Trevisan M. Work and leisuree time physical activity and mortality in men and women from a general population sample. Ann Epidemioll 1999;9(6):366-73.
157.. Ostbye T, Taylor DH, Jung SH. A longitudinal study of the effects of tobacco smoking and other modifiablee risk factors on ill health in middle-aged and old Americans: results from the Health and Retirementt Study and Asset and Health Dynamics among the Oldest Old survey. Prev Med 2002; 34(3):334-45. .
158.. Williamson GM, Schulz R. Symptoms of depression in elderly persons: beyond the effects of physical illnesss and disability. In Facts and Research in Gerontology; 1995.
159.. Hybels CF, Blazer DG, Pieper CF. Toward a threshold for subthreshold depression: an analysis of correlatess of depression by severity of symptoms using data from an elderly community sample. Gerontologistt 2001 ;41 (3):357-65.
160.. AijSnseppa S, Kivinen P, Helkala EL, Kivela SL, Tuomilehto J, Nissinen A. Serum cholesterol and depressivee symptoms in elderly Finnish men. Int J Geriatr Psychiatry 2002;17(7):629-34.
161.. Andersen LB, Schnohr P, Schroll M, Hein HO. All-cause mortality associated with physical activity duringg leisure time, work, sports, and cycling to work. Arch Intern Med 2000;160:1621-8.
162.. Keysor J J. Does late-life physical activity or exercise prevent or minimize disablement? A critical revieww of the scientific evidence. Am J Prev Med 2003;25:129-36.
163.. Blair SN, Wei M. Sedentary habits, health, and function in older women and men. Am J Health Promot 2000;; 15(1): 1-8.
164.. Ormel J, Lindenberg S, Steverink N, Vonkorff M. Quality of life and social production functions: a frameworkk for understanding health effects. Soc Sci Med 1997;45(7):1051-63.
165.. Hoeymans N, Feskens EJ, Kromhout D, van den Bos GA. The contribution of chronic conditions and disabilitiess to poor self-rated health in elderly men. J Gerontol A Biol Sci Med Sci 1999;54(10):M501-M506. .
166.. Gama EV, Damian JE, Perez de Molino J, Lopez MR, Lopez Perez M, Gavira Iglesias FJ. Association off individual activities of daily living with self-rated health in older people. Age Ageing 2000;29(3):267-70. .
167.. Alexander NB, Guire KE, Thelen DG, Ashton-Miller JA, Schultz AB, Grunawalt JC, Giordani B. Self-
118 8
References References
reportedd walking ability predicts functional mobility performance in frail older adults. J Am Geriatr Soc 2000;48(11):: 1408-13.
168.. Kushiro W, Yokoyama T, Date C, Yoshiike N, Tanaka H. Association of activities of daily living and indicess of mental status with subsequent 20-year all-cause mortality in an elderly Japanese population. Nursingg and Health Sciences 2002;4 (suppl):A5.
169.. Kelly-Hayes M, Jette AM, Wolf PA, D'Agostino RB, Odell PM. Functional limitations and disability amongg elders in the Framingham Study. Am J Public Health 1992;82(6):841-5.
170.. Schoevers RA, Geerlings Ml, Beekman AT, Penninx BW, Deeg DJ, Jonker C, Van Tilburg W. Associationn of depression and gender with mortality in old age. Results from the Amsterdam Study of thee Elderly (AMSTEL). Br J Psychiatry 2000; 177:336-42.
171.. Ferrer M, Lamarca R, Orfila F, Alonso J. Comparison of performance-based and self-rated functional capacityy in Spanish elderly. Am J Epidemiol 1999; 149 (3):228-35.
172.172. Addington-Hall J, Kalra L. Who should measure quality of life? BMJ 2001 ;322:1417-20.
173.. Rowe JW. Health care of the elderly. N Engl J Med 1985;312(13):827-35.
174.. Cross-national comparisons of the prevalences and correlates of mental disorders. WHO International Consortiumm in Psychiatric Epidemiology. Bull World Health Organ 2000; 78(4):413-26.
175.. Clark J. Preventive home visits to elderly people. Their effectiveness cannot be judged by randomised controlledd trials. BMJ 2001;323{7315):708.
176.. Van der Bij AK, Laurant MG, Wensing M. Effectiveness of physical activity interventions for older adults:: a review. Am J Prev Med 2002;22(2): 120-33.
177.. Deeg DJ, Kriegsman DM. Concepts of self-rated health: specifying the gender difference in mortality risk.. Gerontologist 2003;43{3):376-86; discussion 372-5.
178.. Bickenbach JE, Chatterji S, Badley EM, Ustun TB. Models of disablement, universalism and the internationall classification of impairments, disabilities and handicaps. Soc Sci Med 1999;48(9):1173-87. .
179.. Colsher PL, Wallace RB. Data quality and age: health and psychobehavioral correlates of item nonresponsee and inconsistent responses. J Gerontol 1989;44(2):P45-P52.
180.. Hardy SE, Gill TM. Factors associated with recovery of independence among newly disabled older persons.. Arch Intern Med 2005;165:106-12.
181.. Hardy SE, Gill TM. Recovery from disability among community-dwelling older persons. JAMA 2004;2911 (13): 1596-602.
182.182. Al Snih S, Markides KS, Ostir GV, Ray L, Goodwin JS. Predictors of recovery in activities of daily living amongg disabled older Mexican Americans. Aging Clin Exp Res 2003; 15(4):315-20.
183.. Gill TM, Robison JT, Tinetti ME. Predictors of recovery in activities of daily living among disabled older
119 9
References References
personss living in the community. J Gen Intern Med 1997;12(12):757-62.
184.. Crawford SL, Jette AM, Tennstedt SL. Test-retest reliability of self-reported disability measures in older adults.. J Am Geriatr Soc 1997;45(3):338-41.
185.. Hoeymans N, Wouters ER, Feskens EJM, van den Bos GAM, Kromhout D. Reproducibility of performance-basedd and self-reported measures of functional status. J Gerontol A Biol Sci Med Sci 1997;52(6):M363-8. .
186.. Kempen Gl, van Sonderen E. Psychological attributes and changes in disability among low-functioning olderr persons: does attrition affect the outcomes? J Clin Epidemiol 2002;55(3):224-9.
187.. McCurry SM, Gibbons LE, Bond GE, Rice MM, Graves AB, Kukull WA, Teri L, Higdon R, Bowen JD, McCormickk WC, et al. Older adults and functional decline: a cross-cultural comparison. Int Psychogeriatrr 2002; 14(2): 161 -79.
188.. Schoevers RA, Beekman AT, Deeg DJ, Hooijer C, Jonker C, van Tilburg W. The natural history of late-lifee depression: results from the Amsterdam Study of the Elderly (AMSTEL). J Affect Disord 2003;76(1-3):5-14. .
189.. Aijanseppa S, Notkola IL, Tijhuis M, van Staveren W, Kromhout D, Nissinen A. Physical functioning in elderlyy Europeans: 10 year changes in the north and south: the HALE project. J Epidemiol Community Healthh 2005;59(5):413-9.
190.. Conn VS, Minor MA, Burks KJ, Rantz MJ, Pomeroy SH. Integrative review of physical activity interventionn research with aging adults. J Am Geriatr Soc 2003;51:1159-68.
191.. Cyarto EV, Moorhead GE, Brown WJ. Updating the evidence relating to physical activity intervention studiess in older people. J Sci Med Sport 2004;7(1 Suppl ):30-8.
192.. Latham NK, Bennett DA, Stretton CM, Anderson CS. Systematic review of progressive resistance strengthh training in older adults. J Gerontol A Biol Sci Med Sci 2004;59(1):48-61.
193.. Singh MA. Exercise to prevent and treat functional disability. Clin Geriatr Med 2002;18(3):431-62, vi-vii. .
194.. Andersen RE, Wadden TA, Bartlett SJ, Zemel B, Verde TJ, Franckowiak SC. Effects of lifestyle activity vss structured aerobic exercise in obese women: a randomized trial. JAMA 1999;281(4):335-40.
195.. Dunn AL, Marcus BH, Kampert JB, Garcia ME, Kohl HW 3rd, Blair SN. Comparison of lifestyle and structuredd interventions to increase physical activity and cardiorespiratory fitness: a randomized trial. JAMAA 1999;281(4):327-34.
196.. Shumway-Cook A, Patla A, Stewart A, Ferrucci L, Ciol MA, Guralnik JM. Environmental components of mobilityy disability in community-living older persons. J Am Geriatr Soc 2003;51(3):393-8.
197.. Brach JS, VanSwearingen JM, FitzGerald SJ, Storti KL, Kriska AM. The relationship among physical activity,, obesity, and physical function in community-dwelling older women. Prev Med 2004;39(1):74-80. .
120 0
References References
198.. Bratzler DW, Oehlert WH, Austelle A. Smoking in the elderly-it's never too late to quit. J Okla State Medd Assoc 2002;95(3): 185-91; quiz 192-3.
199.. Fries JF. Reducing disability in older age. JAMA 2002;288(24):3164-6.
200.. Andrews GR. Promoting health and function in an ageing population. BMJ 2001 ;322(7288):728-9.
201.. Stuck AE, Walthert JM, Nikolaus T, Bula CJ, Hohmann C, Beck JC. Risk factors for functional status declinee in community-living elderly people: a systematic literature review. Soc Sci Med 1999; 48(4):445-69. .
202.. Stuck AE, Egger M, Hammer A, Minder CE, Beck JC. Home visits to prevent nursing home admission andd functional decline in elderly people: systematic review and meta-regression analysis. JAMA 2002;287(8):: 1022-8.
203.. Elkan R, Kendrick D, Dewey M, Hewitt M, Robinson J, Blair M, Williams D, Brummell K. Effectiveness off home based support for older people: systematic review and meta-analysis. BMJ 2001;323(7315):719-25. .
204.. van Haastregt JC, Diederiks JP, van Rossum E, de Witte LP, Crebolder HF. Effects of preventive homee visits to elderly people living in the community: systematic review. BMJ 2000;320(7237):754-8.
205.. Williams K. The transition to widowhood and the social regulation of health: consequences for health andd health risk behavior. J Gerontol B Psychol Sci Soc Sci 2004;59(6):S343-S349.
206.. Caserta MS, Lund DA, Rice SJ. Pathfinders: a self-care and health education program for older widowss and widowers. Gerontologist 1999;39(5):615-20.
207.. Van Exel J, Koopmanschap MA, Van Wijngaarden JD, Scholte Op Reimer WJ. Costs of stroke and strokee services: Determinants of patient costs and a comparison of costs of regular care and care organisedd in stroke services. Cost Eff Resour Alloc 2003;1(1):2.
208.. Langhorne P, Dennis MS. Stroke units: the next 10 years. Lancet 2004;363(9412):834-5.
209.. Langhorne P, Duncan P. Does the organization of postacute stroke care really matter? Stroke 2001;32(1):268-74. .
210.. Anderson C, Ni Mhurchu C, Brown PM, Carter K. Stroke rehabilitation services to accelerate hospital dischargee and provide home-based care: an overview and cost analysis. Pharmacoeconomics 2002;20(8):537-52. .
211.. Bohlmeijer E, Smit F, Smits C, Cuijpers P. Integrale aanpak depressiepreventie bij ouderen. Utrecht: Trimbos-Instituut;; 2005.
212.. Alexopoulos GS, Buckwalter K, Olin J, Martinez R, Wainscott C, Krishnan KR. Comorbidity of late life depression:: an opportunity for research on mechanisms and treatment. Biol Psychiatry 2002;52(6):543-58. .
213.. Cameron ID, Kurrle SE. 1: Rehabilitation and older people. Med J Aust 2002;177(7):387-91.
121 1
References s
214.. Klausner EJ, Clarkin JF, Spielman L, Pupo C, Abrams R, Alexopoulos GS. Late-life depression and functionall disability: the role of goal- focused group psychotherapy. Int J Geriatr Psychiatry 1998;13<10):707-16. .
215.. van Exel NJ, Koopmanschap MA, van den Berg B, Brouwer WB, van den Bos GA. Burden of informal caregivingg for stroke patients. Identification of caregivers at risk of adverse health effects. Cerebrovasc Diss 2005; 19(1 ):11-7.
216.. Brouwer WB, van Exel NJ, van de Berg B, Dinant HJ, Koopmanschap MA, van den Bos GA. Burden of caregiving:: evidence of objective burden, subjective burden, and quality of life impacts on informal caregiverss of patients with rheumatoid arthritis. Arthritis Rheum 2004;51(4):570-7.
217.. Jacobi CE, van den Berg B, Boshuizen HC, Rupp I, Dinant HJ, van den Bos GA. Dimension-specific burdenn of caregiving among partners of rheumatoid arthritis patients. Rheumatology (Oxford) 2003; 42(10):1226-33. .
218.. Nijboer C, Triemstra M, Tempelaar R, Mulder M, Sanderman R, van den Bos GA. Patterns of caregiver experiencess among partners of cancer patients. Gerontologist 2000;40(6):738-46.
219.. Schulz R, Beach SR. Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMAA 1999;282(23):2215-9.
220.. Scholte op Reimer WJ, de Haan RJ, Rijnders PT, Limburg M, van den Bos GA. The burden of caregivingg in partners of long-term stroke survivors. Stroke 1998;29(8):1605-11.
221.. Koopmanschap MA, van Exel NJ, van den Bos GA, van den Berg B, Brouwer WB. The desire for supportt and respite care: preferences of Dutch informal caregivers. Health Policy 2004;68(3):309-20.
122 2
Summar y y
Summary Summary
Duringg the 20th century, functional disability in elderly people has become a major public health
problemm in the Western world due to demographic and epidemiologic transitions. The demographic
transition,, i.e. a shift from patterns of high fertility and mortality rates to low fertility and delayed
mortality,, has led to a growing number of elderly people. The post-war baby boom (people born
betweenn 1946 and 1955) will further increase the number of elderly people in the next decades.
Thee epidemiologic transition, i.e. a change in patterns of health, disease, and mortality, has led to
ann increase in the prevalence of chronic diseases and disability. Disability has a major impact on
qualityy of life and the demand on health care. It is therefore time now to focus on disability in
researchh and health care. The ageing of the population is a success story, but presents society
withh new challenges related to the independency of elderly people.
Thiss thesis aimed to identify risk groups and risk factors for functional disability and to quantify
healthh (care) impacts of functional disability. Firstly, the assessment of disability was validated by
relatingg self-reported disability to performance tests. Thereafter, widowers were studied as a risk
groupp for disability, and physical activity as a risk factor for disability. Furthermore, we investigated
whetherr use of formal home care was according to disability-related needs. The health impact of
disabilityy was investigated by estimating its relationship with depressive symptoms and mortality. In
thee discussion section, the main findings were presented, and some methodological issues and
implicationss for preventive strategies and health care programs were addressed.
Wee defined functional disability as needing help in daily activities. These activities could be
categorisedd into three domains: 1). instrumental activities of daily living (e.g. preparing meals,
doingg housework); 2). mobility (e.g. moving outdoors, using stairs); 3). basic activities of daily living
(e.g.. dressing, using toilet). Disability severity was based on the hierarchical order of these three
disabilityy domains.
Thee data for our study came from the Finland, Italy and the Netherlands Elderly (FINE) Study, that
consistss of 2285 men born between 1900 and 1920. The FINE Study is a prospective study on risk
factorss and health in elderly men. Four surveys were carried out in 1984-1985, 1989-1991, 1994-
1995,, and 1999-2000.
Self-reportedd disability is often used to compare health status between countries. We therefore
investigatedd cross-sectionally whether self-reported disability and its association with performance-
basedd tests is comparable between countries, using data of the second survey (chapter 2). The
scoree on the performance tests was based on a standing balance test, walking speed test, chair
standd test, and shoulder rotation test. In all three countries, statistically significant associations
betweenn self-reported disability and performance-based tests were found, in the sense that more
124 4
Summary Summary
disabilitiess were related to less performance according to the tests. Dutch men reported more
disabilitiess than men from Finland and Italy, after adjustment for performance-based scores. From
thiss study we concluded that the association between self-reported disability and performance tests
iss comparable between countries, but that cross-cultural variation is present in self-reported
disability,, adjusted for performance-based scores. This suggests that the comparison of health
statuss between countries can not be based on prevalence figures of self-reported disability.
Thee number of widowed men is increasing because the life expectancy of men is rising faster than
thatt of women. We studied longitudinally the relationship between becoming widowed and the
onsett of disability (chapter 3). We started with men who were married at baseline and compared
thee disability status five years later between those who were still married and those who had
becomee widowed. Men who had become widowed had a higher risk of disability in instrumental
activities,, and in mobility, but not in basic activities of daily living. Moreover, men who became
widowedd during the past five years had a higher risk of disability in instrumental activities than
thosee who had been widowed for a longer time. We also studied whether household composition
amongg widowers was associated with disability by comparing widowers living alone with those who
livedd with others. Widowers living alone tended to have more disability in instrumental activities and
lesss disability in mobility compared to widowers living with others. We concluded that widowhood in
elderlyy men is a risk factor for dependency in instrumental activities and mobility and that therefore
thee increase in the number of widowers will lead to higher demands on health care.
Althoughh it was known that physical activity is inversely associated with disability, it was not known
whichh aspects of physical activity explain that association. We therefore studied longitudinally the
relationshipp between two aspects of physical activity, i.e. duration and intensity, and disability onset
(chapterr 4). Only men without disability at baseline were incorporated. Data on physical activity,
collectedd by questionnaire, were based on activities like walking, bicycling, playing billiards, and
gardening.. Men in the highest fertile of duration of physical activity (median of 205 minutes per
day)) had a lower risk of disability than those in the lowest fertile (median of 39 minutes per day),
afterr adjustment for other lifestyle factors. Intensity of physical activity was not associated with
disabilityy onset. We concluded from this study that a physically active lifestyle is important even in
oldd age. This can be achieved by adopting enjoyable activities into daily life. Although the Dutch
physicall activity guideline recommends 30 minutes of physical activity a day, our results suggest
thatt spending more is even better.
Elderlyy people desire care at home rather than institutionalisation. The demand of home care will
increasee because of the rising numbers of elderly people. We evaluated among the Dutch
125 5
Summary Summary
participantss in 2000 whether use of formal home care was according to need, using the Andersen
modell {chapter 5). Need factors (chronic diseases, self-rated health, disability), predisposing (age,
maritall status), and enabling factors (occupation, education, informal care) were incorporated as
predictorss of use of home nursing and home help. Disability was strongly associated with use of
homee nursing. Also educational level was associated with use of home nursing, which suggests
inequity.. Home help use was associated with marital status. Married men had lower use of home
helpp than men not married. Because support by the spouse decreases the demand for formal
homee help, there is no firm evidence for inequity in the use of home help.
Becausee the onset of disability can be considered as a life event, it is a risk factor for depressive
symptoms.. We studied cross-sectionally the association between disability and depressive
symptomss and longitudinally the effect of change in disability (chapter 6). Severity of disability was
determinedd in two ways: by a hierarchy score of the disability domains and by a sum score of
disability.. Depressive symptoms were determined by the Zung questionnaire. Both disability scores
weree significantly associated with depressive symptoms in all three countries. The association
betweenn the hierarchy score of disability and depressive symptoms seemed to be somewhat
strongerr than that between the sum score and depressive symptoms. We also investigated
whetherr depressive symptoms depended on the domains of the disabilities, or whether the
disabilitiess could simply be summed. Although not statistically significant, the results showed that
thee domains of disability were indeed important. Longitudinally, changes in disability were
associatedd with depressive symptoms. Men who had worsening of disability status in the preceding
fivee years reported more depressive symptoms than men who improved in disability status. Health
professionalss should be aware of the risk of depressive symptoms among elderly men with
disability,, especially those with a severe worsening in the past years.
Disabilityy is an important predictor of mortality. We explored this association, focusing on the role
off subjective health aspects. We studied the effect of disability and the additional contribution of
self-ratedd health and depressive symptoms to 10-year mortality (chapter 7). All health aspects
weree significantly associated with mortality. In addition, men with severe disability who felt not
healthy,, had a higher mortality risk than men with severe disability who felt healthy. Among men
withoutt disability, those with more depressive symptoms had a higher mortality risk than those with
lessless depressive symptoms. The association between self-rated health and mortality was
particularlyy present among men with severe disability, while the association between depressive
symptomss and mortality was clearest among men with less severe disability. We concluded that for
adequatee prognoses of mortality, or for developing intervention strategies among elderly people,
nott only physical health aspects, but also subjective health outcomes should be taken into account.
126 6
Summary Summary
Afterr summarising the most important findings of our research, some methodological issues were
addressed,, i.e. the validity of our assessment of disability and internal and external validity (chapter
8).. Furthermore, implications for public health (care) were discussed. Public health programs must
respondd to the challenges created by the growing burden of disability. We addressed strategies to
preventt or postpone disability, related to the topics of our thesis. An attractive target for
interventionss designed to prevent or postpone disability, is the increase of duration of physical
activityy among elderly people. Public health advice should inform people that regular physical
activityy is desirable and can be achieved by adopting enjoyable activities. Furthermore, widowers
shouldd be offered programs that provide health and wellness information, teach new self-care skills
andd focus on psychological aspects, such as coping and prevention of depression. Because elderly
peoplee with disability are a risk group for depression, care programs should also take into account
psychosociall aspects. Finally, the impact of the changing health care system (introduction of the
Sociall Support Act) was discussed. In a few years it should be evaluated whether people who
needd care actually receive care in the new system. In addition, informal caregivers should be
supported,, in order to alleviate the growing burden and to provide a sustainable system of long-
termm care and social support.
Althoughh many people wish to grow old, ageing is often accompanied by health problems, for
examplee functional disability. In this thesis, several starting points for preventing or postponing
disabilityy and other health problems in old age have been identified. Also adequate care and
psychosociall support programs deserve attention in anticipating the growing burden of elderly
peoplee with disability. These issues should get high priority in research and policy in the years to
come,, in order to sustain the autonomy and independent living of the elderly.
127 7
Samenvattin g g
Samenvatting Samenvatting
Ditt proefschrift gaat over functionele beperkingen bij oudere mannen. Functionele beperkingen zijn
gedefinieerdd als het niet zelfstandig kunnen uitvoeren van algemene dagelijkse activiteiten.
Functionelee beperkingen bij ouderen zijn in de loop van de twintigste eeuw uitgegroeid tot een
omvangrijkk volksgezondheidprobleem, als gevolg van demografische en epidemiologische
transities.. Zo heeft de demografische transitie, de verschuiving van hoge naar lage geboorte- en
sterftecijfers,, geleid tot een toenemende vergrijzing van onze samenleving. Onder invloed van de
naoorlogsee geboortegolf zal het aantal ouderen in de komende decennia verder stijgen. Daarnaast
heeftt de epidemiologische transitie, de verschuiving in mortaliteit- en morbiditeitpatronen van
infectieziektenn naar chronische ziekten, geleid tot een hogere levensverwachting en een toename
vann het aantal personen met functionele beperkingen. Functionele beperkingen hebben een grote
weerslagg op de kwaliteit van leven en het zorggebruik van ouderen.
Hett doel van dit proefschrift is het identificeren van risicogroepen en risicofactoren voor functionele
beperkingenn bij oudere mannen. Daarnaast beoogt dit proefschrift de invloed van functionele
beperkingenn op andere aspecten van de gezondheid, en de gezondheidszorg te bepalen.
Allereerstt is de validiteit van zelf-gerapporteerde beperkingen bestudeerd. Vervolgens is
onderzochtt of weduwnaars een risicogroep vormen en of lichamelijke inactiviteit een risicofactor is
voorr functionele beperkingen. Voorts is geanalyseerd of het gebruik van thuiszorg gerelateerd is
aann de met functionele beperkingen samenhangende behoefte aan zorg. Ten slotte is de invloed
vann functionele beperkingen op depressieve symptomen en sterfte bestudeerd.
Dee gegevens voor dit proefschrift zijn afkomstig uit de FINE Studie, een internationaal prospectief
onderzoek,, uitgevoerd onder 2285 mannen die geboren zijn tussen 1900 en 1920 uit Finland, Italië
enn Nederland. De FINE Studie is gericht op determinanten van gezondheid bij oudere mannen. De
resultatenn die in dit proefschrift zijn beschreven, zijn gebaseerd op gegevens die in vier
onderzoeksrondess zijn verzameld: 1984-1985,1989-1991, 1994-1995, en 1999-2000.
Functionelee beperkingen zijn bepaald op basis van verschillende activiteiten in drie domeinen:
1.. instrumentele activiteiten van het dagelijks leven (IADL; bijvoorbeeld koken en huishoudelijke