Abstract

Background: the World Health Organization estimates that more than one billion of the world's population are disabled. Disability is associated with increasing age and poverty, yet there are few reliable data regarding disability among the elderly in low-income countries. The aim of this study was to accurately document the prevalence of disability in those aged 70 years and over in a community-based setting in sub-Saharan Africa.

Methods: we performed a community-based study of people aged 70 years and over in Hai, Tanzania. Participants underwent disability assessment using the culturally non-specific Barthel index (BI), and also clinical assessment for neurological disorders and memory problems.

Results: in 2,232 participants, the age-adjusted prevalence of severe disability (BI < 15) was 3.7% (95% CI: 2.9–4.5) and the age-adjusted prevalence of moderate disability (BI: 15–18) was 6.2% [95% confidence interval (CI): 5.2–7.2]. Increasing age, female gender, memory problems and the presence of neurological disorders were all independent predictors of the presence of disability.

Conclusion: in this study, the average disability level was lower than seen in most high-income countries. This may reflect increased mortality from disabling disease in low-income countries. Disability is likely to increase as the population of low-income countries ages and disease survival improves.

Introduction

The Barthel index (BI) measures physical disability defined as; any impairment which restricts motor ability and in turn limits activities of daily living (ADL) [1]. The first World Report on Disability suggests more than one billion people experience disability worldwide [2]. However, there are few data on the prevalence of disability in low-income countries, particularly among older people [3].

Populations in low-income countries are ageing rapidly as they undergo demographic transition. In 2050 an estimated two billion people will be aged 60 years and over and 80% will reside in resource-poor settings [4]. This epidemiological transition will include increasing non-communicable disease (NCD), such as neurological disorders and dementia [4]. Collectively this provides a public health dilemma that low-income countries are ill-equipped to deal with [5]. Such diseases are often accompanied by disability and dependence, placing strains on the traditional supportive family network seen in many low-income countries. Prevalence of disability increases with age worldwide [6], but robust data from sub-Saharan Africa (SSA) are required.

The primary aim of this study was to document the prevalence of disability in those aged 70 years and over in a community-based setting in SSA, with a secondary aim to identify risk factors for disability.

Methods

Ethical approval for the study was obtained locally from Tumaini University ethics committee and nationally from the Tanzanian National Institute of Medical Research.

Setting

Data were collected between 1 November 2009 and 31 July 2010. The point prevalence date was the 1 January 2010. The Hai district of northern Tanzania is located on the southern side of Mount Kilimanjaro and includes a demographic surveillance site (DSS). There are regular population censuses within the DSS the most recent, completed on 1st June 2009, recorded the population of the 52 villages as 161,119, of whom 8,869 were 70 years and over. Villages within the district are classified as upland or lowland depending on their topography. Most villages within the DSS are served by either a small health centre or a dispensary supplying basic treatments. There are three small hospitals in the district, and a tertiary hospital in the adjacent district. The DSS is broadly representative of the rural population of Tanzania; people live within large family units, the majority are subsistence farmers and daily activities consist of work on their shamba (small holding), running the household and taking care of family members. They have little disposable income and for financial, diagnostic and supply reasons, few are on regular medication [7].

Study population

The study population was identified via a census. Collecting accurate information on patient age can be difficult in SSA, as few people have a birth certificate [7]. Age was calculated from birth year and confirmed using memory prompts. This method has been validated and shown to be accurate to within 3 years [8].

We planned to see one-quarter of the entire population aged 70 years and over in the DSS. Using a random number generator and stratification for upland and lowland villages, 12 villages, with a total census population of 2,425 aged 70 years and over were selected. Exclusions, refusals and additions are shown in Figure 1, giving a final cohort of 2,232 people. In the DSS, of those aged 70 years and over, 76.2% live in upland areas and 23.8% in lowland areas. We studied eight upland villages (n = 1,683, 75.4%) and four lowland villages (n = 549, 24.6%).

Figure 1.

Identification of the study population.

Figure 1.

Identification of the study population.

Assessment and data collection

Procedures

Participants were seen at a place of their convenience (village health centres or patients' houses). Signed informed consent was obtained from each participant. In cases where patients were unable to consent due to cognitive impairment, written consent was obtained from a close relative. The BI was performed, through interview with the patient or next of kin (if the patient was unable to reliably answer the questions due to cognitive impairment). Age, sex, tribe and body mass index (BMI) were also recorded. This study was carried out in conjunction with other studies documenting the prevalence of neurological disorders and dementia, through direct assessment rather than self-reporting.

Measures

The BI [9] is an ordinal scale that asks questions regarding ten ADLs with questions given varying weights, the total score ranging from 0 to 20 [10]. The BI is widely used and recommended by the British Geriatrics Society and the UK Royal College of Physicians for routine use in the assessment of older people [11, 12]. Questions are more objective than those included in the Tanzanian Disability Survey [13]. While the World Health Organization (WHO)-DAS is the most recent proposed screening instrument for disability, we felt it was not appropriate as a rapid screening instrument given its length [14]. The BI is culturally and disease non-specific and simple modification allows easy use in low-income countries. Thus, direct comparison between disability levels worldwide can be made. It has good inter-rater reliability and can be consistently used in all disorders that result in physical disability. It has been proposed as the standard measure of disability and the gold standard against which new measures should be compared [10, 15, 16].

The BI was translated into Swahili and back-translated to ensure accuracy. Local census enumerators acted as non-medical investigators and were trained in its use. Efforts were made to make it culturally non-specific by minor modification of one question; ability in climbing stairs was changed to ability to walk up a steep hill.

Although other cut-offs have been used [17–19], Heslin et al. [20] define severe, moderate and mild/no disability as a BI of less than 15, 15–18 and 19 or 20, respectively. We utilised these cut-offs in our analyses.

A validated neurological screening instrument (consisting of 21 questions with a high sensitivity and specificity) followed by a neurological history and examination (to confirm or refute a neurological diagnosis) and further validation by a neurologist, movement disorder expert or geriatrician identified cases of neurological disease [21]. These were diagnosed based on WHO-ICD 10 and other specialist classifications. To provide data on cognitive function, participants (or their next of kin) were asked two questions, Do you have persistent problems understanding not because of a hearing problem? Has there been deterioration in your memory that stops you doing your normal daily activities without assistance?’ For the purposes of this study, a positive answer to either or both questions was taken as evidence of memory problems. These two questions were designed as a practical and simple screen for memory problems within our cohort in the context of the current disability study. They were not designed to provide a formal diagnosis of cognitive problems.

Statistical analysis

Age standardisation of crude prevalence figures to the WHO world population was performed using the direct method [22]. Confidence intervals (CIs) for prevalence and for odds ratios were calculated based on the assumptions of the binomial distribution and for continuous variables based on the normal distribution. A logistic regression model was constructed to help identify independent predictors of moderate or severe disability. Stepwise methods were used to construct the model and the robustness of the model checked by examination of residuals, Cook's distances, loading of variables on individual eigenvalues and collinearity diagnostic tests. Two-tailed tests and a significance level of 5% were used throughout.

Results

The BI was performed on 2,232 study participants. Of these, 976 (43.7%) were male. The mean age was 77.9 (95% CI: 77.4–78.3) for males and 77.7 (95% CI: 77.3–78.1) for females. The mean BMI was 20.9 (95% CI: 20.7–21.2) for males and 22.2 (95% CI: 21.9–22.5) for females. The majority tribe was Chagga, 2,032 (91.0%); the remainder included Pare and Maasai tribes.

Crude and age-adjusted prevalence rates of severe disability (BI < 15) were 4.3% (95% CI: 3.4–5.1) and 3.7% (95% CI: 2.9–4.5). Crude and age-adjusted prevalence rates of moderate disability (BI: 15–18) were 6.9% (95% CI: 5.8–8.0) and 6.2% (95% CI: 5.2–7.2). The age-specific prevalence of moderate or severe disability combined for males and females are shown in Table 1.

Table 1.

Prevalence of moderate or severe disability (BI < 19) within our cohort

Age bands Males
 
Females
 
Cases Denominator population Prevalence % (95% CI) Cases Denominator population Prevalence % (95% CI) 
70–74 21 384 5.5 (3.2–7.7) 25 493 5.1 (3.1–7.0) 
75–79 18 285 6.3 (3.5–9.1) 24 340 7.1 (4.3–9.8) 
80–84 10 135 7.4 (3.0–11.8) 33 209 15.8 (10.8–20.7) 
85+ 41 172 23.8 (17.5–30.2) 77 214 36.0 (29.6–42.4) 
Total 90 976 9.2 (7.4–11.0) 159 1,256 12.7 (10.8–14.5) 
Crude prevalence (male and female) 249 2,232 11.2 (9.8–12.5)    
Age-adjusted prevalence (male and female)a — — 9.9 (8.7–11.2)    
Age bands Males
 
Females
 
Cases Denominator population Prevalence % (95% CI) Cases Denominator population Prevalence % (95% CI) 
70–74 21 384 5.5 (3.2–7.7) 25 493 5.1 (3.1–7.0) 
75–79 18 285 6.3 (3.5–9.1) 24 340 7.1 (4.3–9.8) 
80–84 10 135 7.4 (3.0–11.8) 33 209 15.8 (10.8–20.7) 
85+ 41 172 23.8 (17.5–30.2) 77 214 36.0 (29.6–42.4) 
Total 90 976 9.2 (7.4–11.0) 159 1,256 12.7 (10.8–14.5) 
Crude prevalence (male and female) 249 2,232 11.2 (9.8–12.5)    
Age-adjusted prevalence (male and female)a — — 9.9 (8.7–11.2)    

aTo WHO world population.

In both genders, disability levels increased with greater age. Of those aged 85 years and over 30.6% had moderate or severe disability, compared with only 7.1% of those aged less than 85 years, OR = 5.76 (95% CI: 4.36–7.63). In total, 4.6% of females and 3.8% of males were severely disabled, whereas 12.7% of females and 9.2% of males were moderately or severely disabled. The odds of a female having moderate or severe disability were 1.43 (95% CI: 1.09–1.88) times that of a male.

Three hundred and forty-nine cases were identified as having a neurological disorder. Data relating to the presence of a neurological disorder are presented in Table 2. The odds of someone being moderately or severely disabled if a neurological disorder was present were 4.58 (95% CI: 3.45–6.09) that of someone who did not have a neurological disorder. Of the 56 cases who were identified as having memory problems, 24 (42.9%) had severe disability, 14 (25.0%) had moderate disability and 18 (32.1%) had no disability. The odds of someone being moderately or severely disabled if they had memory problems were 19.66 (95% CI: 11.02–35.06) that of someone who did not have memory problems.

Table 2.

The relationship of disability and neurological disorders

 Cases with severe disability, % (95% CI) Cases with moderate disability, % (95% CI) Cases with no disability, % (95% CI) 
Whole study population (= 2,232) n = 95, 4.3% (3.4–5.1) n = 154, 6.9% (5.9–8.0) n = 1,983, 88.8% (87.9–90.2) 
Study population without neurological disorders (n = 1,883) n = 45, 2.4% (1.7–3.1) n = 104, 5.5% (4.5–6.6) n = 1,734, 92.1% (90.9–93.3) 
Study population with neurological disorders (n = 349) n = 50, 14.3% (10.7–18.0) n = 50, 14.3% (10.7–18.0) n = 249, 71.4% (66.6–76.1) 
Tremor n = 12, 10.9% n = 12, 10.9% n = 86, 78.2% 
Parkinsonism n = 3, 21.4% n = 5, 35.7% n = 6, 42.9% 
Cerebellar disorders n = 3, 27.3% n = 2, 18.2% n = 6, 54.5% 
Dyskinesias n = 1, 20.0% n = 2, 40.0% n = 2, 40.0% 
Headaches that affect ADLs n = 3, 3.3% n = 11, 12.0% n = 78, 84.8% 
Stroke n = 22, 40.7% n = 12, 22.2% n = 20, 37.0% 
Epilepsy n = 1, 10.0% n = 2, 20.0% n = 7, 70.0% 
Cord lesions n = 4, 80.0% n = 0 n = 1, 20.0% 
Polyneuropathies n = 2, 4.8% n = 6, 14.3% n = 34, 81.0% 
Cranial nerve pathology n = 0 n = 0 n = 14, 100% 
Upper limb mononeuropathies n = 1, 6.7% n = 0 n = 14, 93.3% 
Lower limb mononeuropathies n = 1, 20.0% n = 2, 40.0% n = 2, 40.0% 
Polio n = 0 n = 1, 20.0% n = 4, 80.0% 
Muscle wasting and atrophy n = 9, 52.9% n = 2, 11.8% n = 6, 35.3% 
 Cases with severe disability, % (95% CI) Cases with moderate disability, % (95% CI) Cases with no disability, % (95% CI) 
Whole study population (= 2,232) n = 95, 4.3% (3.4–5.1) n = 154, 6.9% (5.9–8.0) n = 1,983, 88.8% (87.9–90.2) 
Study population without neurological disorders (n = 1,883) n = 45, 2.4% (1.7–3.1) n = 104, 5.5% (4.5–6.6) n = 1,734, 92.1% (90.9–93.3) 
Study population with neurological disorders (n = 349) n = 50, 14.3% (10.7–18.0) n = 50, 14.3% (10.7–18.0) n = 249, 71.4% (66.6–76.1) 
Tremor n = 12, 10.9% n = 12, 10.9% n = 86, 78.2% 
Parkinsonism n = 3, 21.4% n = 5, 35.7% n = 6, 42.9% 
Cerebellar disorders n = 3, 27.3% n = 2, 18.2% n = 6, 54.5% 
Dyskinesias n = 1, 20.0% n = 2, 40.0% n = 2, 40.0% 
Headaches that affect ADLs n = 3, 3.3% n = 11, 12.0% n = 78, 84.8% 
Stroke n = 22, 40.7% n = 12, 22.2% n = 20, 37.0% 
Epilepsy n = 1, 10.0% n = 2, 20.0% n = 7, 70.0% 
Cord lesions n = 4, 80.0% n = 0 n = 1, 20.0% 
Polyneuropathies n = 2, 4.8% n = 6, 14.3% n = 34, 81.0% 
Cranial nerve pathology n = 0 n = 0 n = 14, 100% 
Upper limb mononeuropathies n = 1, 6.7% n = 0 n = 14, 93.3% 
Lower limb mononeuropathies n = 1, 20.0% n = 2, 40.0% n = 2, 40.0% 
Polio n = 0 n = 1, 20.0% n = 4, 80.0% 
Muscle wasting and atrophy n = 9, 52.9% n = 2, 11.8% n = 6, 35.3% 

A logistic regression model for predictors of moderate or severe disability was constructed, as shown in the Supplementary data available in Age and Ageing online, Appendix 1. Age, sex, the presence of neurological disability and the presence of memory problems were identified as significant independent predictors of disability. Overall the model predicted 30.1% (Nagelkerke's R2) of the variability in whether a person had moderate or severe disability or not. Sex was the weakest predictor of disability, whereas age and the presence of memory problems were the strongest predictors. Someone with memory problems was almost 20 times more likely to have moderate or severe disability than someone with no memory problems.

Discussion

Reported disability prevalence rates around the world vary significantly because of different definitions, study designs and the lack of a universal assessment instrument. This makes inter-study comparisons difficult.

The use of the (self-reported) BI, with identical cut-offs, in similarly aged patients (those aged 70 years and over) allows direct comparison to a study of health status of people aged 70–94 years in eight European districts. Of 4,004 subjects, 9 and 26% were severely and moderately disabled, respectively [20]. These differences may reflect increased mortality from disabling disease in low-income countries and increased risk factors for disability such as greater age, obesity, physical inactivity and higher prevalence of NCD in high-income countries.

Difficulty interpreting quoted disability levels is evident on reviewing the Tanzanian Disability Survey from 2008 [13] where subjective questioning was used and the report of ‘some’ difficulty with seeing, hearing, walking, self-care or communicating defined disability. They report disability levels of 41.3% for the 70–74 age group, rising to 55.6% for those aged 80 years and over. In our experience, the majority of these people will still be fully independent and therefore it is difficult to contextualise these results and fully assess their significance.

Other studies from both high- and low-income countries support the hypothesis that disability rates are higher in the developed world; however, the different measures used again complicate interpretation and highlight the unique nature of our study [23–26].

Age, gender and the presence of neurological disorders were identified as independent risk factors of physical disability. The Cognitive Function and Ageing Study [26] supports the former association, reporting that 38% of all disabled people were aged 85 years or over. They noted the economic and social burden that an ageing population represents. The 10/66 dementia research group supports the latter association concluding that dementia, stroke, limb impairment and arthritis were associated with disability [3].

Interventions to reduce the burden of disability in SSA should be aimed at both the individual and the societal level [1]. Many disabling conditions can be prevented or treated readily and inexpensively. Identification of those with disability, and diseases that result in disability, is clearly an important component in reducing the morbidity burden [27]. Clearly models of rehabilitation from high-income countries may have limitations in this environment where, for example, a community-based rehabilitation strategy may be more appropriate.

Study limitations

There is little consensus regarding suitable cut-offs of the BI for the level of disability. Uyttenboogaart et al. compare the BI to the modified Rankin score (MRS) demonstrating that an MRS score of 1 (no significant disability) corresponds with a BI score of 19–20, an MRS score of 2 (slight disability) is equivalent to a BI score of 18 and an MRS score of 3 (moderate disability) to a BI score of 15 [17]. Celani et al. suggest that a score of 19–20 indicates those not requiring help from another person [18]. Finally, Kay et al. conclude that a BI score of 16 or less was the optimal cut-off for self-reported dependency [19]. There is consensus that MRS of 2 or less reflects independence (equivalent to BI 18 or more). Since the cut-off of Heslin et al. [20] allows direct comparison with our data we used these in our analyses.

As a result of its ease of use, the BI does not assess the social and psychological aspect of disability. The WHO defines disability as an umbrella term covering impairments, activity limitations and participation restrictions, physical disability (as measured by the BI) is only a part of this; however, it is said to be most representative of elderly people's ability to function independently [1, 28, 29].

The BI was administered by interview rather than performance-based testing. This may reduce its reliability, particularly in the elderly where self-reported scores may be higher [30]. The question relating to memory was designed to pick up those whose ADLs were affected by this and so may over-estimate the influence of memory problems on disability. It was beyond the scope of this study to evaluate the contribution of all risk factors for disability and consequently a fully inclusive assessment of risk factors cannot be presented. Finally, although our population was broadly representative of rural Tanzania, we are unable to comment on the generalisability of our results to urban populations, where risk factor profiles are likely to be substantial different.

Conclusion

This is the first large-scale community-based study from SSA examining disability rates in older people. We have revealed a lower prevalence of disability in the older people in Tanzania in comparison with Europe. This may reflect increased mortality from disease in SSA. It may also highlight different risk factors for disability within high-income countries. Disability levels are likely to increase in low-income countries as their populations' age and disease survival rates increase, placing a greater burden on wider society. Our results support those of the WHO World Health Survey and Global study on Ageing in relation to age, gender and the role of NCD. Disability in low-income countries will become increasingly problematic as the population ages and NCDs become more prevalent.

Key points

  • This is the first large-scale community-based study from SSA examining disability rates in older people.

  • Lower prevalence of disability in the older people in Tanzania in comparison with Europe.

  • Disability levels are likely to increase in low-income countries as their population ages and disease survival rates increase.

Authors' contributions

Design/conception: R.W.W. and F.D. Literature search: R.W.W. and F.D. Data collection: F.D., M.J.D., G.O. A.R.L., C.D. and S.-M.P. Data analysis: W.K.G. and F.D. Interpretation of results: R.W.W., W.K.G. and F.D. Writing of paper and review: R.W.W., W.K.G., F.D., G.O., W.H., P.C., A.R.L, C.D. and S.-M.P.

Conflicts of interest

None declared.

Funding

This work was supported by a research fellowship from the Dunhill Foundation and the Royal College of Physicians, London. The sponsors of this study had no role in designing the study; in the collection, analysis and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication.

Supplementary data

Supplementary data mentioned in the text is available to subscribers in Age and Ageing online.

Acknowledgements

We wish to acknowledge the help of all health care workers, officials, carers and family members who assisted in examination, assessment, data collection and input.

References

1
World Health Organization
The International Classification of Functioning Disability and Health
2001
Geveva
World Health Organization
2
World Health Organization
The World Report on Disability
2011
Geneva, Switzerland
World Health Organization
3
Sousa
RM
Ferri
CP
Acousta
D
, et al.  . 
Contribution of chronic diseases to disability in elderly people in countries with low and middle incomes: a 10/66 Dementia Research Group population-based survey
Lancet
 , 
2009
, vol. 
374
 (pg. 
1823
-
30
)
4
World Health Organization
The World Health Report 2008: primary health care—now more than ever
2008
Geneva, Switzerland
World Health Organization
5
Heligman
L
Chen
N
Babakol
O
Shifts in the structure of population and deaths in less developed regions
Epidemiologic Transitions: Policy and Planning Implications for Developing Countries
 , 
1993
Washington, DC
National Academy Press
(pg. 
9
-
41
)
6
Martin
J
Metzer
H
Elliot
D
The prevalence of disability in adults, OPCS surveys of disability in Great Britain
1988
HMSO
 
Report 1
7
Adult Morbidity and Mortality Project (AMMP)
Policy implications of adult morbidity and mortality; final report
2004
Dar-es-Salaam
Tanzanian Ministry of Health
 
8
Paraïso
MN
Houinato
D
Guerchet
M
, et al.  . 
Validation of the use of historical events to estimate the age of subjects aged 65 years and over in Cotonou (Benin)
Neuroepidemiology
 , 
2010
, vol. 
35
 (pg. 
12
-
6
)
9
Mahoney
F
Barthel
D
Functional evaluation: the Barthel index
Md Med J
 , 
1965
, vol. 
14
 (pg. 
56
-
61
)
10
Colin
C
Wade
DT
Davies
S
Home
V
The Barthel ADL index: a reliability study
Int Dis Study
 , 
1988
, vol. 
10
 (pg. 
61
-
3
)
11
British Geriatric Society
Rehabilitation of Older People. BGS Best Practice Guide
 , 
2009
London
British Geriatric Society
 
12
Royal College of Physicians
Report of joint workshops of the Royal College of Physicians and the British Geriatrics Society. Standardised assessment scales for elderly people
1992
Royal College of Physicians
13
National Bureau of Statistics
Tanzanian Disability Survey Report
2010
Dar es Salaam, Tanzania
National Bureau of Statistics
14
Sousa
RM
Dewey
ME
Acosta
D
, et al.  . 
Measuring disability across cultures–the psychometric properties of the WHODAS II in older people from seven low- and middle-income countries. The 10/66 Dementia Research Group population-based survey
Int J Methods Psychiatr Res
 , 
2010
, vol. 
19
 (pg. 
1
-
17
)
15
Sainsbury
A
Seebass
G
Bansal
A
Young
J
Reliability of the Barthel index when used with older people
Age Ageing
 , 
2005
, vol. 
34
 (pg. 
228
-
32
)
16
Gosman-Hedstrom
G
Svensson
E
Parallel reliability of the functional independence measure and the Barthel ADL index
Disability and Rehabilitation
 , 
2000
, vol. 
22
 (pg. 
702
-
15
)
17
Uyttenboogaart
M
Stewart
RE
Vroomen
PCAJ
De Keyser
J
Luijckx
GJ
Optimizing cutoff scores for the Barthel index and the modified Rankin Scale for defining outcome in acute stroke trials
Stroke
 , 
2005
, vol. 
36
 (pg. 
1984
-
7
)
18
Celani
M
Cantisani
T
Righetti
E
Spizzichino
L
Ricci
S
Different measures for assessing stroke outcome: an analysis from the international stroke trial in Italy
Stroke
 , 
2002
, vol. 
33
 (pg. 
218
-
23
)
19
Kay
R
Wong
KS
Perez
G
Woo
J
Dichotomizing stroke outcomes based on self-reported dependency
Neurology
 , 
1997
, vol. 
49
 (pg. 
1694
-
6
)
20
Heslin
JM
Soveri
PJ
Winoy
JB
, et al.  . 
Health status and service utilisation of older people in different European countries
Scand J Prim Health Care
 , 
2001
, vol. 
19
 (pg. 
218
-
22
)
21
Dewhurst
F
Dewhurst
MJ
Orega
G
, et al.  . 
A screening questionnaire to measure the prevalence of neurological disorders in the elderly in low income countries: a community-based pilot study
Eur J Neurol
 , 
2011
, vol. 
18
 
Suppl s2
pg. 
542
 
22
Ahmad
OB
Boshchi-Pinto
C
Lopez
AD
Murray
CJL
Lozano
R
Inoue
M
Age standardisation of rates. A new WHO standard
2001
Geneva
World Health Organisation
 
GPE discussion paper series: No 31
23
Allain
TJ
Wilson
AO
Gomo
AR
, et al.  . 
Morbidity and disability in elderly Zimbabweans
Age Ageing
 , 
1997
, vol. 
26
 (pg. 
115
-
21
)
24
Fitaw
Y
Boersma
JMF
Prevalence and impact of disability in north-western Ethiopia
Disabil Rehabil
 , 
2006
, vol. 
28
 (pg. 
949
-
53
)
25
Statistics New Zealand
Disability counts
1998
Wellington, New Zealand
Statistics New Zealand
26
Melzer
D
McWilliams
B
Brayne
C
, et al.  . 
Medical Research Council Cognitive Function and Aging Study (MRC CFAS) and Resource Implications Study (RIS MRC CFAS), Writing committee
Profile of disability in elderly people: estimates from a longitudinal population study
BMJ
 , 
1999
, vol. 
318
 (pg. 
1108
-
11
)
27
World Health Organization
Community-based rehabilitation (CBR) guidelines
2010
Geneva
World Health Organization
28
Mont
D
Measuring health and disability
Lancet
 , 
2007
, vol. 
369
 (pg. 
1658
-
63
)
29
Crepaldi
G
Maggi
S
Rozzini
R
Trabucchi
M
Reducing disability among the elderly in Europe
Lancet
 , 
1998
, vol. 
351
 pg. 
375
 
30
Sinoff
G
Ore
L
The Barthel activities of daily living index: self-reporting versus actual performance in the old-old (> or = 75 years)
J Am Geriatr Soc
 , 
1997
, vol. 
45
 (pg. 
832
-
6
)

Supplementary data

Comments

0 Comments