Chat with us, powered by LiveChat Results/ Discussion. | Economics Write
+1(978)310-4246 credencewriters@gmail.com
  

Please  Read the study/research and analyze all its details. 
and resume  the results and discussion from the attached article

ORIGINAL ARTICLE

Chronic disease, risk factors and disability in adults aged
50 and above living with and without HIV: findings from
the Wellbeing of Older People Study in Uganda

Joseph O. Mugisha1,2*, Enid J. Schatz2, Madeleine Randell3,
Monica Kuteesa1, Paul Kowal4,5, Joel Negin3 and Janet Seeley1,6

1MRC/UVRI, Uganda Research Unit on AIDS, Uganda; 2Department of Health Sciences, University of
Missouri Columbia, Missouri, USA; 3School of Public Health, University of Sydney, Australia; 4World Health
Organization, Study on global AGEing and adult health, Geneva, Switzerland; 5Research Centre for Gender,
Health and Ageing, University of Newcastle, Australia; 6London School of Hygiene and Tropical Medicine,
London UK

Background: Data on the prevalence of chronic conditions, their risk factors, and their associations with
disability in older people living with and without HIV are scarce in sub-Saharan Africa.

Objectives: In older people living with and without HIV in sub-Saharan Africa: 1) to describe the prevalence
of chronic conditions and their risk factors and 2) to draw attention to associations between chronic

conditions and disability.

Methods: Cross-sectional individual-level survey data from people aged 50 years and over living with and
without HIV were analyzed from three study sites in Uganda. Diagnoses of chronic conditions were made

through self-report, and disability was determined using the WHO Disability Assessment Schedule

(WHODAS). We used ordered logistic regression and calculated predicted probabilities to show differences

in the prevalence of multiple chronic conditions across HIV status, age groups, and locality. We used linear

regression to determine associations between chronic conditions and the WHODAS.

Results: In total, 471 participants were surveyed; about half the respondents were living with HIV. The
prevalence of chronic obstructive pulmonary disease and eye problems (except for those aged 60�69 years) was
higher in the HIV-positive participants and increased with age. The prevalence of diabetes and angina was

higher in HIV-negative participants. The odds of having one or more compared with no chronic conditions were

higher in women (OR 1.6, 95% CI 1.1�2.3) and in those aged 70 years and above (OR 2.1, 95% CI 1.2�3.6).
Sleep problems (coefficient 14.2, 95% CI 7.3�21.0) and depression (coefficient 9.4, 95% CI 1.2�17.0) were
strongly associated with higher disability scores.

Conclusion: Chronic conditions are common in older adults and affect their functioning. Many of these
conditions are not currently addressed by health services in Uganda. There is a need to revise health care

policy and practice in Uganda to consider the health needs of older people, particularly as the numbers of

people living into older age with HIV and other chronic conditions are increasing.

Keywords: Africa; aging; aging disability; HIV/AIDS; older adults; non-communicable diseases; Uganda

Responsible Editor: Jennifer Stewart Williams, Umeå University, Sweden.

*Correspondence to: Joseph O. Mugisha, MRC/UVRI, Plot 51�59, Nakiwogo Road, Entebbe, Uganda,
Email: [email protected]

Received: 25 January 2016; Revised: 27 April 2016; Accepted: 27 April 2016; Published: 24 May 2016

Introduction
Chronic diseases are illnesses or conditions that require

ongoing medical attention and affect a person’s daily life

(1). Chronic diseases include cancers, cardiovascular

diseases, chronic respiratory diseases, diabetes, hyperten-

sion, mental disorders, and stroke. Other chronic impair-

ments that commonly affect people include arthritis;

rheumatism; and dental, vision, stomach, and intestinal

problems (2). In African countries, improved access to

antiretroviral treatment (ART) is increasing survival for

those with the human immunodeficiency virus (HIV).

Consequently, HIV is now considered a chronic condition

in many settings (3).

With shifts in the global burden of disease, chronic

diseases represent a substantial proportion of illnesses

even in low- and middle-income countries (LMICs) (4).

Global Health Action �

Global Health Action 2016. # 2016 Joseph O. Mugisha et al. This is an Open Access article distributed under the terms of the Creative Commons Attribution
4.0 International License (http://creativecommons.org/licenses/by/4.0/), allowing third parties to copy and redistribute the material in any medium or format and
to remix, transform, and build upon the material for any purpose, even commercially, provided the original work is properly cited and states its license.

1

Citation: Glob Health Action 2016, 9: 31098 – http://dx.doi.org/10.3402/gha.v9.31098
(page number not for citation purpose)

http://creativecommons.org/licenses/by/4.0/

http://www.globalhealthaction.net/index.php/gha/article/view/31098

http://dx.doi.org/10.3402/gha.v9.31098

Few studies, however, have used individual-level data

to elucidate the prevalence of chronic conditions, risk

factors, and disability associated with chronic diseases in

older people in LMICs, and such research is particularly

scarce in sub-Saharan Africa. Comprehensive studies on

chronic diseases in LMICs primarily have concentrated

on younger and middle-aged people (5�10) with relatively
few focusing on older adults (2, 9, 11, 12).

In sub-Saharan Africa, the number and proportion

of older people is increasing and is projected to continue to

grow in coming decades (13, 14). This makes it parti-

cularly important to understand how chronic disease

impacts on older Africans’ lives. As African populations

age, the prevalence of individuals with chronic conditions

in these settings is likely to increase. In Uganda, for

example, the population of older people has continued

to grow rapidly (15). In addition, the number of older

people living with HIV in Uganda is also increasing (16) in

line with a global trend (17�19).
A number of studies have been conducted in sub-

Saharan Africa on chronic conditions in adults (7�9,
20�25). However, few provide information on concurrent
chronic conditions, including HIV (23), and fewer still

have simultaneously examined chronic diseases in older

people living with and without HIV (26). In Uganda, as

well, there are few data on health differences in chronic

conditions between older persons living with and without

HIV (27�29).
Chronic diseases can affect people of all age groups, but

they are more common and more likely to have negative

consequences in older adults. A 2005 study of mortality

and the burden of disease predicted an increase in deaths

for all ages worldwide due to chronic diseases (excluding

HIV) from 35 million deaths in 2005 to 41 million deaths

in 2015 (30). Nearly 60% of the deaths in each year are

estimated to occur among those aged 70-plus. Research

from southern Africa shows that chronic diseases (not

including HIV) are more prevalent among those aged

50-plus compared to those aged 18�49 (12). Another study
in South Africa showed that there were more chronic

conditions (excluding HIV) in later older age (65-plus)

than early older age (ages 50�65) (9).
With the exception of HIV, many chronic diseases share

common risk factors. These include excessive alcohol use,

tobacco use, unhealthy diets, and physical inactivity (31).

Current health behaviors, as well as the accumulated

impact of a lifetime of harmful health behaviors, con-

tribute to the higher likelihood of contracting a chronic

condition in older age (32, 33). Because the majority

of these risk factors are related to individual health

behaviors, most are potentially amenable to behavioral

interventions (34).

Using a unique dataset from Uganda, this paper

describes the prevalence of chronic diseases, including

angina; arthritis; chronic obstructive pulmonary disease

(COPD); depression; diabetes mellitus; and hypertension,

stroke, and vision problems, in older people living

with and without HIV. We also describe the prevalence

of related risk factors and association between chronic

disease and disability, using the World Health Organiza-

tion Disability Assessment Schedule (WHODAS 2.0) to

measure disability (35). This paper adds to the limited

body of literature on the prevalence and risk factors of

chronic conditions and how these impact on disability

in older Africans living with and without HIV.

Methods
Data for this analysis came from the second wave of the

longitudinal World Health Organization Study on global

AGEing and adult health (SAGE)-Wellbeing of Older

People Study (WOPS). The SAGE-WOPS HIV study in

Uganda was implemented in people aged 50 plus. To date,

two waves of data are available: the first wave (WOPS1)

conducted in 2009�2010 and the second wave (WOPS2)
conducted in 2012�2013. Details of the initial WOPS
recruitment are described elsewhere (26). Although data

from two waves of WOPS are available, only data from

WOPS2 are analyzed here because of inconsistencies in

available variables across the two waves. We therefore

present findings on a fuller set of more recent variables

rather than longitudinal data on a limited set of variables.

Interviews were conducted in three sites on the shores of

Lake Victoria � in the Kalungu and Masaka districts and
another in the Wakiso District, near Entebbe. The study

setting, study population, and data collection are also

described elsewhere (26, 36). Briefly, the WOPS1 sample

consisted of 510 older people (61.2% female, mean age

65 and age range 50�96 years). These included 1) older
persons who were living with HIV but not yet on ART;

2) older persons living with HIV and on ART for at least

1 year; 3) older persons who had a child living with HIV;

4) older persons who had a child who died of AIDS-related

illness; and 5) older persons who were not HIV-positive

themselves but had not lost a child due to HIV infection.

During WOPS2, we re-interviewed those respondents

who were still living in the area; 148 respondents were

lost to follow-up (these included 67 who had died, 25 who

emigrated from the study area, 17 who were found but

refused to participate, 9 who were too sick to participate,

4 who had travelled on the day of the interviews, 4 who

were too busy to participate in the interviews, and 22 who

could not be located). The follow-up rate was over 70%.

In WOPS2, we recruited an additional 100 older people

living with HIV attending the AIDS Support Organiza-

tion (TASO), a non-governmental organization (NGO) in

Masaka town, close to the Kalungu District site. All the

new recruits were randomly selected from older people

attending TASO. These additional recruits increased the

number of people living with HIV in the cohort. In order

to avoid misclassification of the study groups, all older

Joseph O. Mugisha et al.

2
(page number not for citation purpose)

Citation: Glob Health Action 2016, 9: 31098 – http://dx.doi.org/10.3402/gha.v9.31098

http://www.globalhealthaction.net/index.php/gha/article/view/31098

http://dx.doi.org/10.3402/gha.v9.31098

people who were HIV negative in WOPS1 were retested

for HIV using the Uganda Ministry of Health algorithms

for rapid HIV testing (37). The sample in this study is

stratified by HIV status between all those who were living

with HIV either in WOPS1 or WOPS2, and those who

were HIV negative in WOPS1 and remained so at the time

of testing in WOPS2.

Data collection

Study participants were either interviewed from home or

from a central hub (a central location in their village),

where a house was rented for survey activities. The

interviews were conducted by trained interviewers using

a validated questionnaire. After conducting the interviews,

the interviewers measured weight, height, blood pressure,

grip strengths, walking speed, and conducted a visual acuity

test. The WOPS questionnaire and other data collection

instruments were adapted from the WHO SAGE (38). All

instruments were pretested and piloted prior to use (26).

Variables

The components of the study questionnaire analyzed in

this paper include:

1. Sociodemographic characteristics: age, sex, marital

status, occupation (work status), education level,

and household assets.

2. Risk factors: smoking, alcohol use, stressful events,

sleep disorders, and body mass index (BMI).

3. Self-reported chronic conditions: self-reported diag-

noses of chronic conditions (including angina, ar-

thritis, cataract/eye sight problems, COPD, depression,

diabetes mellitus, hypertension, and stroke).

4. Objective measurements: weight, height, visual acuity

(using the Snellen charts), and blood pressure, mea-

sured three times in a sitting position.

Information from the interviews and assessments was

used to describe health states that included diagnoses,

risk factors, and impairments as described below. Dis-

ability was assessed using the 12-item version of WHO-

DAS 2.0 questionnaire (35).

Diagnoses

Hypertension

For all study participants, systolic and diastolic blood

pressures were measured three times with participants

in a sitting position using a Boso Medistar-S-wrist

blood pressure monitor. An average blood pressure

for the three readings was computed and used in the

analysis. Hypertension was defined according to the

World Health Organization (WHO) criteria (systolic

blood pressure ]140 mmHg and/or diastolic blood pres-

sure ]90 mmHg) (39).

For the conditions listed below, respondents were

asked a range of questions on diagnosis and symptoma-

tology for these chronic conditions, and their responses

determined the diagnosis used here.

Diabetes mellitus, COPD, and eyesight problems/

cataracts
For this analysis, prevalence estimates were based on

the self-report of a doctor’s diagnosis. Participants were

asked the following questions: Have you ever been told by

a doctor or a health worker that you have [condition]?

If yes, were you started on treatment and are you still on

treatment?

Stroke and angina

The prevalence for the conditions of stroke and angina

was determined through algorithms using symptom-

reporting (40, 41).

Depression

A diagnosis of depression was based on a diagnostic

algorithm, with participant responses scored using the

International Neuropsychiatric interview (MINI) criteria

(42�44). The criteria used for determining depression
were based on previous work using the MINI in Uganda

(45, 46). The following screening questions for a major

depressive episode were asked. For the past 2 weeks, were

you depressed or down, most of the day, nearly every day?

In the past 2 weeks, were you much less interested in most

things or much less able to enjoy the things you used to

enjoy, most of the time? If participants answered yes to

these questions, they were asked a number of additional

questions to ascertain a major depressive episode.

Arthritis
First, participants were asked if a health worker had ever

diagnosed or told them that they have arthritis. If the

answer was yes, they were asked about medication use or

any other treatment for arthritis in the last 2 weeks and

the last 12 months, and about symptoms, such as aching,

stiffness, or swelling around the joints that were not

related to injury and lasted for 1 month. Prevalence was

determined using a diagnostic algorithm (40).

HIV

During WOPS1, participants were selected in the five

categories described above. In order to avoid misclassifi-

cation during WOPS2, all participants seen in WOPS1

who were previously HIV negative were subjected to

repeat HIV testing. HIV testing was done using an algo-

rithm for HIV-1 testing using three HIV-1 rapid tests as

recommended by the Uganda Ministry of Health. The

algorithm for HIV rapid testing consisted of an initial

screening with the rapid test Determine HIV1/2. If the test

result was negative the participant was given a diagnosis

of HIV negative with no further rapid testing. If the test

result was positive, the sample was retested with the rapid

test HIV-1/2 Stat-Pak. If both tests gave a positive result

the participant was given a diagnosis of HIV positive with

Chronic conditions and disability in older people with and without HIV in Uganda

Citation: Glob Health Action 2016, 9: 31098 – http://dx.doi.org/10.3402/gha.v9.31098 3
(page number not for citation purpose)

http://www.globalhealthaction.net/index.php/gha/article/view/31098

http://dx.doi.org/10.3402/gha.v9.31098

no further rapid testing. If the tests gave discordant results

(i.e. one positive and the other negative), the sample was

further evaluated with the rapid test Uni-Gold Recombi-

nant HIV-1/2. For those samples assessed by all three

tests, two positive test results were interpreted as a positive

diagnosis. If two of the three tests gave negative results,

then the participant was diagnosed as being negative for

HIV. The two resulting categories for our analysis below

are those who tested HIV positive and those who tested

HIV negative.

Risk factors

Risk factors included tobacco use (if participants were

using tobacco, they were asked about the duration of

use), the method of tobacco consumption (whether they

were smoking or using chew or snuff), and the quantity

of tobacco consumed on each of the previous 7 days.

Alcohol use was determined by asking whether partici-

pants had ever or were currently consuming alcohol, the

duration of use, and the types of alcohol consumed. BMI

was determined from weight and height measurements

taken at the time of the survey. BMI was calculated by

dividing weight in kilograms by height in meters squared.

Disability

Questions necessary to generate the 12-item version of

WHODAS 2.0 were asked in the interview (47�49). These
questions gather information across six domains: cogni-

tion, mobility, self-care, getting along, life activities, and

participation, asking about difficulty in these domains

during the 30 days preceding the interview. The possible

responses for each question were on a five-point scale:

‘none’, ‘mild’, ‘moderate’, ‘severe’, and ‘extreme or cannot

do’. The WHODAS 2.0 algorithm was used to compute

an overall score [range 0�100] for each respondent, with a
higher score indicative of greater level of disability (47).

Ethical issues

Ethical approval to conduct this study was obtained from

the Uganda Virus Research Institute Science and Ethics

Committee, the Uganda National Council for Science

and Technology, and WHO’s Ethical Review Committee.

All participants gave a written and thumb-printed con-

sent to participate in the study. For non-literate partici-

pants, an impartial third party witnessed the entire

consent process and counter-signed the consent document

on which the participant had placed their thumb-print.

Statistical methods and data analysis

All analyses were conducted in Stata 13 (Stata Corp,

College Park, Tx, USA). We did not use any imputation

methods for missing data. However, the majority of

variables had two or fewer missing cases, only three

variables had more than 10 missing cases: BMI (11), stroke

(12), and current employment status (17). All descriptive

statistics and sample sizes are presented as un-weighted

values, with a p value of B0.05 considered statistically

significant (all p values are two-sided). We did not apply

sampling weights. The study sample was selected ran-

domly from lists of older people in the study population.

Analyses for descriptive statistics and risk factors were

stratified by HIV status for each of the following

characteristics: sociodemographic variables (mean age,

gender, locality, employment status, marital status, and

highest level of education), all past and current use of

tobacco, all past and current alcohol use, mean BMI,

sleep problems, and antiretroviral (ART) use-conditional

on HIV status. Analyses for chronic conditions (angina,

arthritis, diabetes, COPD, depression, eye problems,

hypertension, and stroke) were stratified by HIV status

and age group; chi-square statistics highlight whether

there were significant differences (1) across chronic con-

ditions by HIV status and age group, and (2) significant

differences between risk factors and HIV status. Median

differences in age and BMI were calculated for the two

respondent groups due to the data not being normally

distributed. Wilcoxon rank-sum analyses were used

to compare median differences in age and BMI for

the two respondent groups. We conducted an ordered

logistic regression and calculated predicted probabilities

to show the differences in the number of chronic con-

ditions across HIV status, gender, age group, and locality.

We defined the number of chronic conditions using an

algorithm that grouped respondents into three categories

being zero chronic conditions; one condition; or two or

more conditions. However, HIV was not considered a

chronic condition for the purposes of these counts. We

tested the proportional odds assumption for ordered

logistic regression. This assumes that the coefficients that

describe the relationship between the lowest versus all

higher categories of the response variable are the same as

those that describe the relationship between the next

lowest category and all higher categories. For this, we

used the omodel command in Stata and achieved a non-

significant result, meaning that there was no difference in

coefficients between models. For each respondent group,

mean WHODAS 2.0 scores were determined for each

chronic condition. T-tests were run within each respon-

dent group to compare WHODAS scores for those with

or without a chronic condition diagnosis.

Linear regression analyses were used to determine

existing associations between sociodemographic factors,

chronic conditions, and risk factors to WHODAS scores.

Univariate analyses first determined significant main

effects as well as interaction terms between HIV status

and other factors before a multiple linear regression with

these variables was undertaken. Although HIV was not

significant in the univariate analysis, we left it in the final

model as an a priori confounder together with age and

gender. In the linear regression modeling, HIV negative

was used as the reference category. Thus, compared to

Joseph O. Mugisha et al.

4
(page number not for citation purpose)

Citation: Glob Health Action 2016, 9: 31098 – http://dx.doi.org/10.3402/gha.v9.31098

http://www.globalhealthaction.net/index.php/gha/article/view/31098

http://dx.doi.org/10.3402/gha.v9.31098

those who were HIV negative, HIV-positive individuals

were expected to have higher WHODAS scores (meaning

more disability). For all the univariate and multivariate

analyses, a significance level of 0.05 was used. Model fit

was assessed by examining residuals from the model. For

this analysis, a robust regression analysis was used.

Results

Sociodemographic characteristics of study

participants

Sociodemographic characteristics of the study popula-

tion by HIV status are provided in Table 1. In total, the

median age for the 471 participants was 63 (50�101). The
majority of the sample was female (62.6%), widowed, still

working, and had less than primary school education.

About half of the study participants (51.8%) were HIV

positive. The HIV-positive respondents tended to be

younger. Only about 10% of older persons living with

HIV were aged 70 or older, whereas over half of the HIV-

negative sample was in the older age groups. Locality

differences by HIV status are in part due to the sampling

strategies.

Chronic conditions by HIV status
Several differences in the percentage of individuals report-

ing chronic conditions, other than HIV, were evident

between the two respondent groups (Table 2). When

comparing by HIV status, the prevalence of COPD and

eye problems (except for those aged 60�69 years) were
higher in the HIV-positive participants and prevalence of

diabetes and angina were higher in HIV-negative partici-

pants. When comparing across age groups within HIV

status, significant differences were present for eye pro-

blems and hypertension, which generally increased with

age, and multi-morbidity for which the prevalence was

higher in those with advanced age. The percentage of

people with COPD decreased with age for both groups,

with a higher starting point and a steeper decline in the

percentage for the HIV-positive group.

The odds of having at least one or one or more,

compared with no chronic conditions (other than HIV)

Table 1. Sociodemographic factors by HIV status

HIV�(N �244) HIV�(N �227)

Demographics N % N %

Gender

Male 97 39.8 79 34.8

Female 147 60.3 148 65.2

Age

50�59 135 55.3 33 14.5

60�69 82 33.6 69 30.4

70�79 23 9.4 82 36.1

80 � 4 1.6 43 18.9

Locality

Wakiso 64 26.2 105 46.3

Kalungu 73 29.9 120 52.9

Masaka 107 43.9 2 0.9

Marital status

Never married 3 1.2 9 4.0

Cohabitating/married 77 31.6 70 30.8

Divorced/separated 57 23.4 49 21.6

Widowed 107 43.9 99 43.6

Current employment status (n �241) (n �226)

Still working 213 88.4 166 73.5

No longer working 28 11.6 60 26.6

Education level (n �242) (n �212)

No formal education 35 14.5 53 23.5

Less than primary 96 39.7 113 50.0

Completed primary 43 17.8 16 7.1

Incomplete secondary 40 16.5 16 7.1

Completed secondary 15 6.2 14 6.2

Higher education than secondary 3 1.2 6 2.7

College/university or more 10 4.1 8 3.5

Chronic conditions and disability in older people with and without HIV in Uganda

Citation: Glob Health Action 2016, 9: 31098 – http://dx.doi.org/10.3402/gha.v9.31098 5
(page number not for citation purpose)

http://www.globalhealthaction.net/index.php/gha/article/view/31098

http://dx.doi.org/10.3402/gha.v9.31098

are shown in Table 3. The odds of having one or more than

one chronic condition were significantly higher in women

and the oldest age group. The predicted probabilities of

having one or more chronic conditions (other than HIV)

in Table 4 give similar findings. Predicted probabilities are

higher in women and in those aged 70 years and above.

Risk factors by HIV status

Several significant differences in the percentage of respon-

dents reporting or having risk factors for chronic conditions

(other than HIV) by HIV status were also evident (Table 5).

BMI was higher for HIV-negative respondents compared

to those who were HIV positive. This, however, may be a

result of HIV status rather than a risk factor for chronic

conditions. A higher proportion of HIV-negative respon-

dents said they currently use both tobacco and alcohol

compared to HIV-positive respondents. A higher proportion

of HIV-negative respondents also experienced mild sleep

problems as compared to HIV-positive respondents.

Linear regression of WHODAS scores

We found no interaction effects between HIV and other

factors before undertaking the multiple regression analy-

sis. Tables 5 and 6 show that there are several significant

differences in the proportion of chronic conditions (other

than HIV) and risk factors between respondents living

with and without HIV. These reached significance in the

Table 2. Percentage of chronic conditions by age and HIV status

50�59 (N �168) 60�69 (N �151) 70�(N �152)

HIV�

(N �135) (%)

HIV �

(N �33) (%)

HIV�

(N �82) (%)

HIV �

(N �69) (%)

HIV�

(N �27) (%)

HIV �

(N �125) (%) p Value by age

p Value by

HIV status

Hypertension

Yes 23.7 48.5 30.5 27.5 33.3 56.8 0.00 0.00

Diabetes

Yes 2.2 9.1 0.0 8.7 3.7 8.1 0.287* 0.001*

Arthritis

Yes 6.7 9.1 6.1 2.9 7.4 4.9 0.743* 0.316

Angina

Yes 0.9 0.0 1.4 5.2 0.0 4.8 0.225* 0.05*

COPD

Yes 10.4 3.0 7.3 1.5 3.7 1.6 0.026* 0.002*

Eye problems

Yes 4.4 3.0 4.9 7.3 18.5 16.1 0.001* 0.017

Depression

Yes 12.6 3.0 8.5 7.3 7.4 7.2 0.464 0.114

Stroke

Yes 1.5 3.0 1.2 0.0 3.7 4.0 0.140* 0.533*

Number of conditions

None 52.6 42.4 51.2 55.1 55.6 28.0 0.00* 0.004*

One 44.4 57.6 47.6 44.9 33.3 67.2

More than one 3.0 0.0 1.2 0.0 11.1 4.8

*Fisher’s exact test used due to small cell size.

Note: HIV not treated as a chronic condition throughout all tables.

Bold values are statistically significant at pB0.05.

Table 3. Ordered multivariate logistic regression of one or

more chronic conditions
a

Independent variable OR (95% CI) p

HIV status

Positive*

Negative 1.4 (0.9�2.2) 0.149

Gender

Male*

Female 1.6 (1.1�2.3) 0.024

Age group

50�59*

60�69 0.9 (0.5�1.4) 0.538

70 � 2.1 (1.2�3.6) 0.006

Locality

Wakiso*

Kalungu 0.5 (0.4�0.8) 0.005

Masaka 1.0 (0.6�1.8) 0.982

aZero chronic conditions is the reference group. HIV status,

gender, age group, and locality are included in the final model.

*values in italic are statistically significant at pB0.05.

Joseph O. Mugisha et al.

6
(page number not for citation purpose)

Citation: Glob Health Action 2016, 9: 31098 – http://dx.doi.org/10.3402/gha.v9.31098

http://www.globalhealthaction.net/index.php/gha/article/view/31098

http://dx.doi.org/10.3402/gha.v9.31098

univariate analyses (Table 5), however, when controlling

for all other variables, many of the associations between

these variables and the WHODAS score were no longer

significant. These included current tobacco use, HIV

infection, and arthritis diagnosis.

Table 6 shows the factors that were significantly

associated with WHODAS. A diagnosis of depression

was associated with …

error: Content is protected !!