Baseline characteristics
A total of 31 385 survey respondents reporting on HIV testing history were included: 14 911 and 16 474 records from Malawi and Zimbabwe and 14 027 and 17 358 among men and women respectively (Table 1). Of these, a total of 24 683 individuals were asked about HIVST with 6702 (21.4%, n=31 385) not asked about HIVST. An additional 15 individuals (0.06%, n=24 683) asked about HIVST had missing data related to questions on sexual activity used to determine HIV-related sexual risk. A total of 78.6% and 75.4% of people reported ever having tested for HIV in Malawi and Zimbabwe, respectively (p<0.001). More women, compared to men (31.2% vs. 16.5%; p<0.001), and more urban, compared to rural, residents (24.1% vs 21.0%; p<0.001) had tested previously. Compared to those aged ≥25 years, a larger proportion of those never-tested were young people (aged 15-24) (39.2% vs 14.1%; p<0.001).
Table 1. Baseline characteristics in Malawi and Zimbabwe, 2015-16
Variables*
|
Ever tested
(N=31 385)**
|
Ever self-test
(N= 24 683)**
|
Aware self-test
(N=24 683)**
|
N
|
%
|
p-value§
|
N
|
%
|
p-value§
|
n
|
%
|
p-value§
|
Total population
|
24 148
|
76.9
|
|
287
|
1.2
|
|
3118
|
12.6
|
|
Country
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
Malawi
|
11 726
|
75.4
|
|
141
|
1.0
|
|
1671
|
11.4
|
|
Zimbabwe
|
12 422
|
78.6
|
|
146
|
1.5
|
|
1447
|
14.5
|
|
Sex
|
|
|
<0.001
|
|
|
0.008
|
|
|
<0.001
|
Female
|
14 500
|
83.5
|
|
103
|
1.0
|
|
983
|
9.1
|
|
Male
|
9648
|
68.8
|
|
184
|
1.3
|
|
2135
|
15.3
|
|
Residence
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
Urban
|
7951
|
79.9
|
|
151
|
2.1
|
|
1516
|
21.2
|
|
Rural
|
16 197
|
75.9
|
|
136
|
0.8
|
|
1602
|
9.1
|
|
Age group (yrs)
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
15-19
|
3252
|
44.8
|
|
30
|
0.5
|
|
437
|
7.0
|
|
20-24
|
4703
|
80.7
|
|
42
|
0.9
|
|
562
|
12.1
|
|
25-29
|
4337
|
90.3
|
|
59
|
1.6
|
|
580
|
16.0
|
|
30-34
|
4070
|
91.1
|
|
55
|
1.7
|
|
497
|
15.2
|
|
35-39
|
3247
|
89.2
|
|
42
|
1.5
|
|
426
|
15.7
|
|
40-44
|
2446
|
87.4
|
|
35
|
1.7
|
|
303
|
15.0
|
|
45+
|
2093
|
80.8
|
|
24
|
1.1
|
|
313
|
14.7
|
|
Wealth
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
Poorest
|
3697
|
76.4
|
|
18
|
0.5
|
|
246
|
6.4
|
|
Poorer
|
4029
|
75.8
|
|
38
|
0.9
|
|
336
|
7.8
|
|
Middle
|
4252
|
75.3
|
|
26
|
0.6
|
|
279
|
8.3
|
|
Rich
|
5594
|
77.6
|
|
56
|
0.9
|
|
669
|
12.2
|
|
Richest
|
6576
|
78.5
|
|
149
|
2.3
|
|
1488
|
22.8
|
|
HIV status
|
|
|
<0.001
|
|
|
0.108
|
|
|
0.029
|
HIV negative
|
20 646
|
75.0
|
|
249
|
1.1
|
|
2760
|
12.5
|
|
HIV positive
|
2570
|
90.6
|
|
38
|
1.5
|
|
358
|
10.4
|
|
Marital status
|
|
|
<0.001
|
|
|
0.005
|
|
|
<0.001
|
Single
|
7595
|
58.9
|
|
99
|
0.9
|
|
1175
|
11.2
|
|
Married or cohabiting
|
16 553
|
89.5
|
|
188
|
1.3
|
|
1943
|
13.7
|
|
Employment
|
|
|
<0.001
|
|
|
0.033
|
|
|
<0.001
|
Not actively working
|
8719
|
70.1
|
|
85
|
1.0
|
|
823
|
9.4
|
|
Actively working
|
15 429
|
81.4
|
|
202
|
1.3
|
|
2295
|
14.4
|
|
Education
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
≤ Primary
|
10 617
|
74.8
|
|
70
|
0.6
|
|
927
|
7.5
|
|
≥ Secondary
|
13 531
|
78.7
|
|
217
|
1.8
|
|
2191
|
17.8
|
|
Literacy
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
Being illiterate
|
5211
|
73.5
|
|
44
|
0.7
|
|
488
|
8.0
|
|
Being literate
|
18 937
|
77.9
|
|
243
|
1.3
|
|
2630
|
14.2
|
|
Sexually active
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
Sexually inactive
|
9064
|
64.0
|
|
107
|
1.5
|
|
1225
|
17.7
|
|
Active in past 4-weeks
|
15 053
|
87.6
|
|
180
|
1.5
|
|
1890
|
14.2
|
|
HIV-related risk***
|
|
|
<0.001
|
|
|
<0.001
|
|
|
<0.001
|
Having low risk
|
8457
|
63.5
|
|
94
|
0.9
|
|
1102
|
10.3
|
|
Having moderate risk
|
13 092
|
88.7
|
|
134
|
1.2
|
|
1497
|
13.6
|
|
Having high risk
|
2570
|
78.8
|
|
59
|
2.0
|
|
516
|
17.2
|
|
*Ever tested refers to people surveyed on HIV testing history who reported they previously tested for HIV before the survey. Overall, 31 385 people were asked about their HIV testing history and 24 148 responded they had tested previously. Ever self-tested refers to people surveyed on HIV self-testing who reported they had previously self-tested. Overall, 24 683 people were asked whether they had self-tested and 287 reported they self-tested previously. Aware of self-testing refers to people surveyed who reported they were aware of HIV self-testing. Overall, 24 683 people were asked whether they were aware of self-testing and 3118 reported they were aware of self-testing.
**Out of 31 385 people surveyed, 31 348 were included as 37 people were missing information on sexual activity and HIV-related risk. Not all participants were systematically surveyed on self-testing questions. Out of 31 385 people surveyed 24 683 were asked about self-testing resulting in smaller sample size. Among these were 15 people reporting on self-testing who did not provide information on sexual activity and HIV risk. Population size asked about ever testing for HIV: 31 347 (HIV risk/sexual activity). Population size asked about awareness or ever self-testing for HIV: 24 668 (HIV risk/ activity).
***HIV risk is defined in this analysis includes reported sexual activity in the past four weeks, and the following high-risk exposures in the previous 12 months: multiple (i.e. ≥2) partners, any paid sex (asked to men), having received gifts, cash or other compensation in exchange for sex (asked to women), and having a sexually transmitted infection (STI). Individuals with any “high-risk” exposures were classified as “high-risk”, with the remaining respondents classified as “low risk” if reporting no sexually activity in the past four weeks, and as “moderate risk” otherwise.
§ P-value based on cluster-adjusted chi-squared test
The proportion of people who had ever self-tested was very low (1.2%) and similar in both countries. However, while awareness was low (12.6%) there was greater awareness of HIVST in Zimbabwe compared to Malawi (14.5% vs 11.4%; p<0.001) and for men compared to women (15.3% vs. 9.1%; p<0.001) (Table 1). Adults ≥30 years of age (≥30 years: 21.1% vs. <30 years: 9.1%; p<0.001), wealthier (richest: 22.8% vs poorest: 6.4%; p<0.001) and individuals with higher education (17.8% vs 7.5%; p<0.001) had greater awareness of self-testing than those aged <30 years, poorer, and individuals with lower education-levels.
Willingness to self-test could only be assessed among 7372 Zimbabwean men (48 men missing data on willingness) as only men were asked about willingness to self-test, and this question was not included in the Malawi DHS questionnaires.
Most Zimbabwean men (84.5%) were willing to self-test (Supplementary table S1 includes for baseline characteristics of Zimbabwean men on willingness to self-test, 2015-16). Men aged ≥25 reported greater willingness to self-test than men aged <25 (88.7% vs. 78.8%; p<0.001). High-risk men also reported greater willingness to self-test than low-risk men (78.8% vs 63.5%; p<0.001). Most men willing to self-test had tested in the past 12 months (88.5%). However, a high proportion of men who had not tested for HIV in the previous two or more years were also willing to self-test (86.4%).
Ever testing for HIV
HIV-related age, HIV status and sexual risk appeared to modify effects in the multivariable analysis across a number of variables (Table 2). Collinearity affected results of multivariable analysis, notably between age and HIV-related sexual risk, marital status and HIV-related sexual risk, age and education level and education level and literacy.
Table 2. Univariable and multivariable associations between sociodemographic factors and ever testing for HIV in Malawi and Zimbabwe, 2015-16
Variables
|
Univariable (weighted)
n=31 375*
|
Multivariable (weighted)
n=31 347*
|
OR
|
95% CI and
p-value
|
aOR
|
95% CI and
p-value
|
Country
|
|
|
|
|
Zimbabwe
|
1
|
|
1
|
|
Malawi
|
1.18
|
1.10-1.28
|
1.26
|
1.15-1.38
|
Sex
|
|
|
|
|
Female
|
1
|
|
1
|
|
Male
|
0.42
|
0.39-0.45
|
0.39
|
0.36-0.42
|
Age (years)
|
|
|
|
|
15-19
|
1
|
p<0.001§
|
1
|
p<0.001§
|
20-24
|
5.40
|
4.87-5.98
|
4.37
|
3.91-4.87
|
25-29
|
11.89
|
10.48-13.48
|
8.24
|
7.18-9.46
|
30-34
|
14.19
|
12.38-16.27
|
8.86
|
7.63-10.29
|
35-39
|
10.42
|
9.01-12.05
|
6.41
|
5.47-7.50
|
40-44
|
8.44
|
7.27-9.80
|
5.30
|
4.50-6.25
|
45+
|
5.16
|
4.50-5.91
|
3.72
|
3.20-4.34
|
Residence
|
|
|
|
|
Urban
|
1
|
|
1
|
|
Rural
|
0.83
|
0.76-0.90
|
1.00
|
0.88-1.14
|
HIV status
|
|
|
|
|
HIV negative
|
1
|
|
1
|
|
HIV positive
|
3.44
|
2.98-3.97
|
2.11
|
1.80-2.49
|
Marital status
|
|
|
|
|
Single
|
1
|
|
1
|
|
Married or cohabiting
|
6.07
|
5.62-6.54
|
**
|
**
|
Wealth
|
|
|
|
|
Poorest
|
1
|
p<0.003§
|
1
|
p<0.581§
|
Poor
|
0.94
|
0.84-1.05
|
1.02
|
0.89-1.16
|
Middle
|
0.91
|
0.81-1.01
|
1.09
|
0.96-1.25
|
Rich
|
1.04
|
0.93-1.16
|
1.05
|
0.91-1.20
|
Richest
|
1.09
|
0.97-1.23
|
1.11
|
0.94-1.31
|
Employment
|
|
|
|
|
Not actively working
|
1
|
|
1
|
|
Actively working
|
1.81
|
1.69-1.95
|
1.16
|
1.06-1.26
|
Education
|
|
|
|
|
≤ Primary
|
1
|
|
1
|
|
≥ Secondary
|
1.26
|
1.17-1.35
|
**
|
**
|
Literacy
|
|
|
|
|
Being illiterate
|
1
|
|
1
|
|
Being literate
|
1.30
|
1.20-1.40
|
1.63
|
1.50-1.78
|
HIV-related risk***
|
|
|
|
|
Having low risk
|
1
|
p<0.001§
|
1
|
p<0.001§
|
Having moderate risk
|
4.58
|
4.23-4.96
|
2.15
|
1.96-2.36
|
Having high risk
|
2.14
|
1.92-2.40
|
1.54
|
1.80-2.49
|
* Both samples were weighted based on standard DHS weights; Strata = 56; PSU=1256. Univariable draws from a total of 31 385 observations, population size 31 375. Multivariable draws from a total of 31 348 observations and population size of 31 338. This excludes 37 people who did not report on sexual activity and risk behaviours and are missing from the “HIV risk category”.
**Represents variables which were not included in the multivariable analysis due to identified collinearity.
*** HIV risk is defined in this analysis includes reported sexual activity in the past four weeks, and the following high-risk exposures in the previous 12 months: multiple (i.e. ≥2) partners, any paid sex (asked to men), having received gifts, cash or other compensation in exchange for sex (asked to women), and having a sexually transmitted infection (STI). Individuals with any “high-risk” exposures were classified as “high-risk”, with the remaining respondents classified as “low risk” if reporting no sexually activity in the past four weeks, and as “moderate risk” otherwise.
§ P-value based on Wald Test. P-values for variables with more than two categories are shown.
On multivariable analysis, after assessing for collinearity, being Malawian was associated with ever having tested for HIV included (adjusted odds ratio (aOR)=1.26; 95% confidence interval [CI]: 1.15-1.38, p<0.001) (Table 2). However, men had substantially lower odds of having ever tested for HIV compared to women (aOR=0.36; 95%CI: 0.32-0.41, p<0.001).And individuals between 30 and 34 years of age had greater odds of ever having tested for HIV compared to 15-19 year olds (aOR=8.86; 95%CI: 7.63-10.29, p<0.001). Additional factors associated with ever having tested fore HIV included: an HIV-positive test result in the survey (HIV-positive vs. HIV-negative: aOR =2.11, 95%CI:1.80-2.49, p<0.001), employment (actively working vs not actively working: aOR=1.16; 95%CI: 1.06-1.26, p<0.001), literacy (being literate vs being illiterate: aOR=1.63, 95%CI:1.50-1.78, p<0.001), and reporting more HIV-related sexual risk behaviours (moderate vs. low: aOR=2.15; 95%CI:1.96-2.36, p<0.001, and high-risk vs. low: aOR=1.54; CI:1.80-2.49, p<0.001).
Use and awareness of self-testing
A complete analysis of ever self-testing is shown in supplementary table S2. Table 3 provides outcomes from univariable and multivariable analysis for awareness of HIV self-testing.
In the multivariable analysis, men aged 30-34 years old had greater odds of past self-testing use compared to younger men (age 15-19) (aOR= 2.89; 95%CI: 1.47-5.68, p<0.002) (Table S2). Across wealth quintiles, being wealthier was also associated with previous self-testing (p<0.001) with wealthiest individuals having the greatest odds of past self-testing (aOR for richest vs. poorest=3.59; 95%CI:1.79-7.18, p<0.001).
On multivariable analysis, respondents in Malawi and those from a rural setting were less likely to be aware of HIVST compared with Zimbabweans and urban participants (Table 3). However, the following variables were significantly associated with being aware of HIVST: being male (male vs female: aOR=1.55;95%CI:1.37-1.75, p<0.001), older than 15-19 years (when compared with those aged 25-29 years: aOR=1.76; 95%CI:1.43-2.17, p<0.001 and aged 35-39 years: aOR=1.69; 95%CI: 1.34-2.12, p<0.001), wealthier (richest vs poorest: aOR=3.03; 95%CI: 2.46-3.73, p<0.001), employment (actively working vs not actively working: aOR=1.25; 95%CI: 1.12-1.42, p<0.001), literate (being literate vs being illiterate: aOR= 1.17; 95%CI: 1.01-1.36, p<0.035) and having previously tested for HIV (ever tested vs never tested: aOR=1.89; 95%CI:1.65-2.17, p<0.001).
Table 3. Univariable and multivariable associations between sociodemographic factors and awareness of HIV self-testing in Malawi and Zimbabwe, 2015-16
Variables
|
Univariable (weighted)
n=24 683*
|
Multivariable (weighted)
n=24 668*
|
OR
|
95% CI and
p-value
|
aOR
|
95% CI and
p-value
|
Country
|
|
|
|
|
Zimbabwe
|
1
|
|
1
|
|
Malawi
|
0.76
|
0.67-0.87
|
0.82
|
0.70-0.94
|
Sex
|
|
|
|
|
Female
|
1
|
|
1
|
|
Male
|
1.73
|
1.54-1.92
|
1.55
|
1.37-1.75
|
Age
|
|
|
|
|
15-19
|
1
|
p<0.001§
|
1
|
p<0.001§
|
20-24
|
1.79
|
1.50-2.12
|
1.35
|
1.12-1.62
|
25-29
|
2.52
|
2.11-3.00
|
1.76
|
1.43-2.17
|
30-34
|
2.44
|
2.02-2.94
|
1.66
|
1.32-2.08
|
35-39
|
2.46
|
2.04-2.97
|
1.69
|
1.34-2.12
|
40-44
|
2.09
|
1.70-2.55
|
1.45
|
1.14-1.86
|
45+
|
2.00
|
1.64-2.46
|
1.31
|
1.04-1.66
|
Residence
|
|
|
|
|
Urban
|
1
|
|
1
|
|
Rural
|
0.33
|
0.29-0.39
|
0.64
|
0.55-0.77
|
Ever tested
|
|
|
|
|
No
|
1
|
|
1
|
|
Yes
|
2.18
|
1.94-2.45
|
1.89
|
1.65-2.17
|
HIV status
|
|
|
|
|
HIV negative
|
1
|
|
1
|
|
HIV positive
|
1.12
|
0.95-1.31
|
0.89
|
0.75-1.06
|
Marital status
|
|
|
|
|
Single
|
1
|
|
1
|
|
Married or cohabiting
|
1.26
|
1.13-1.39
|
**
|
**
|
Wealth
|
|
|
|
|
Poorest
|
1
|
p<0.001§
|
1
|
p<0.001§
|
Poor
|
1.26
|
1.04-1.53
|
1.24
|
1.02-1.51
|
Middle
|
1.26
|
1.03-1.53
|
1.25
|
1.02-1.53
|
Rich
|
1.87
|
1.53-2.28
|
1.49
|
1.20-1.84
|
Richest
|
4.30
|
3.54-5.22
|
3.03
|
2.46-3.73
|
Employment
|
|
|
|
|
Not actively working
|
1
|
|
1
|
|
Actively working
|
1.63
|
1.47-1.82
|
1.25
|
1.12-1.42
|
Education
|
|
|
|
|
≤Primary
|
1
|
|
1
|
|
≥ Secondary education
|
2.69
|
2.38-3.04
|
**
|
**
|
Literacy
|
|
|
|
|
Unable to read
|
1
|
|
1
|
|
Can read
|
1.84
|
1.59-2.12
|
1.17
|
1.01-1.36
|
HIV risk***
|
|
|
|
|
Low risk
|
1
|
p<0.001§
|
1
|
p<0.518
|
Moderate risk
|
1.37
|
1.24-1.53
|
1.03
|
0.90-1.17
|
High risk
|
1.75
|
1.51-2.03
|
1.10
|
0.93-1.31
|
* Both samples were weighted based on standard Demographic Health Survey weights; Strata = 56; PSU=1256. Not all participants were systematically surveyed on self-testing questions. Out of 31 385 people surveyed 24 683 were asked about self-testing resulting in smaller sample size. Among those reporting on HIV self-testing, 15 did not provide information on sexual activity and HIV risk. Population size asked about awareness or ever self-testing for HIV: 24 668 (HIV risk), 24 668 (sexual activity).
**Represents variables which were not included in the multivariable analysis due to identified collinearity.
*** HIV risk is defined in this analysis includes reported sexual activity in the past four weeks, and the following high-risk exposures in the previous 12 months: multiple (i.e. ≥2) partners, any paid sex (asked to men), having received gifts, cash or other compensation in exchange for sex (asked to women), and having a sexually transmitted infection (STI). Individuals with any “high-risk” exposures were classified as “high-risk”, with the remaining respondents classified as “low risk” if reporting no sexually activity in the past four weeks, and as “moderate risk” otherwise.
§ P-value based on Wald Test. P-values for variables with more than two categories are shown.
Willingness to self-test among Zimbabwean men
The relationship between willingness to test and socio-economic variables (wealth and actively working) and HIV status, substantially differed according to both high and low HIV-related sexual risk (Table 4): see for example univariable OR for HIV status and employment. Thus, we adapted our planned multivariable analysis to account for effect-modification between HIV-related sexual risk categorization and socioeconomic variables. On multivariable analysis, comparing men with high HIV-related sexual risk behaviours to those with low HIV-related sexual risk behaviour, high-risk men were more likely than low-risk men to express willingness to self-test if they were also from higher socioeconomic quintiles, not working, in rural settings and tested previously (interaction terms: socioeconomic status, p = 0.066, rural residence, p=0.071, employment p=0.003, literacy, p=0.225, married, p=0.401, aware of self-test, p=0.605, previous testing, p=0.001, and HIV status p = 0.162) (Table 4).
Table 4. Univariable and multivariable associations between sociodemographic factors and willingness to self-test among men in Zimbabwe, by having low, moderate and high HIV-related risk, 2015-16
Variables
|
Univariable (weighted)
|
Multivariable (weighted)
|
Having low risk
(n=3142)*
|
Having moderate risk
(N=2988)*
|
Having high risk
(n=1241)*
|
Having low risk
(n=3142)*
|
Having moderate risk
(N=2988)*
|
Having high risk
(n=1241)*
|
OR
|
95% CI and
p-value
|
OR
|
95% CI and
p-value
|
OR
|
95% CI and
p-value
|
aOR
|
95% CI and
p-value
|
aOR
|
95% CI and
p-value
|
aOR
|
95% CI and
p-value
|
Age (years)
|
|
|
|
|
|
|
|
|
|
|
|
|
15-19
|
1
|
p<0.001§
|
1
|
p=0.063§
|
1
|
p=0.028§
|
1
|
p=0.106§
|
1
|
p=0.343§
|
1
|
p=0.030§
|
20-24
|
1.78
|
1.37-2.30
|
1.49
|
0.70-3.23
|
2.44
|
1.21-4.92
|
1.47
|
1.11-1.92
|
1.31
|
0.60-2.85
|
2.71
|
1.32-5.57
|
25-29
|
2.00
|
1.36-2.96
|
2.43
|
1.18-4.99
|
2.09
|
0.99-4.41
|
1.50
|
1.00-2.27
|
1.89
|
0.90-3.95
|
2.66
|
1.23-5.75
|
30-34
|
2.01
|
1.11-3.63
|
2.52
|
1.24-5.09
|
2.86
|
1.35-6.05
|
1.44
|
0.79-2.64
|
1.98
|
0.96-4.07
|
3.82
|
1.82-8.00
|
35-39
|
1.69
|
0.96-2.99
|
2.31
|
1.10-4.85
|
3.77
|
1.75-9.14
|
1.17
|
0.65-2.10
|
1.92
|
0.91-4.07
|
4.87
|
2.14-11.07
|
40-44
|
1.82
|
0.88-3.78
|
2.52
|
1.16-5.44
|
2.21
|
0.95-5.16
|
1.27
|
0.59-2.72
|
2.09
|
0.96-4.59
|
3.02
|
1.18-7.71
|
45+
|
1.61
|
0.93-2.81
|
1.70
|
0.87-3.33
|
1.94
|
0.93-4.07
|
1.05
|
0.56-1.95
|
1.44
|
0.72-2.88
|
2.46
|
1.09-5.54
|
Residence
|
|
|
|
|
|
|
|
|
|
|
|
|
Urban
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
Rural
|
0.81
|
0.64-1.02
|
1.18
|
0.89-1.55
|
1.33
|
0.86-2.06
|
0.71
|
0.49-1.03
|
1.14
|
0.74-1.76
|
3.56
|
1.61-7.90
|
Wealth
|
|
|
|
|
|
|
|
|
|
|
|
|
Poorest
|
1
|
p=0.128§
|
1
|
p=0.113§
|
1
|
p=0.981§
|
1
|
p=0.102§
|
1
|
p=0.260§
|
1
|
p=0.080§
|
Poor
|
1.04
|
0.75-1.45
|
1.87
|
1.13-3.09
|
1.16
|
0.58-2.30
|
1.02
|
0.74-1.41
|
1.72
|
1.02-2.91
|
1.27
|
0.64-2.50
|
Middle
|
0.97
|
1.00-1.90
|
1.27
|
0.85-1.88
|
0.96
|
0.49-1.89
|
0.98
|
0.71-1.35
|
1.20
|
0.78-1.84
|
1.04
|
0.51-2.10
|
Rich
|
1.38
|
0.70-1.34
|
1.12
|
0.71-1.77
|
1.03
|
0.52-2.05
|
1.04
|
0.73-1.47
|
1.03
|
0.60-1.77
|
2.64
|
1.07-6.53
|
Richest
|
1.02
|
0.74-1.42
|
1.12
|
0.75-1.67
|
1.10
|
0.57-2.12
|
0.65
|
0.42-1.02
|
1.02
|
0.54-1.94
|
3.74
|
1.39-10.03
|
Employment
|
|
|
|
|
|
|
|
|
|
|
|
|
Not actively working
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
Actively working
|
1.64
|
1.35-1.99
|
1.19
|
0.86-1.63
|
0.72
|
0.44-1.18
|
1.41
|
1.13-1.77
|
1.12
|
0.78-1.61
|
0.57
|
0.34-0.95
|
HIV status
|
|
|
|
|
|
|
|
|
|
|
|
|
HIV negative
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
HIV positive
|
1.82
|
1.19-2.79
|
0.94
|
0.56-1.59
|
0.76
|
0.43-1.35
|
1.41
|
0.87-2.30
|
0.84
|
0.49-1.42
|
0.67
|
0.37-1.21
|
Marital status
|
|
|
|
|
|
|
|
|
|
|
|
|
Single
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
Married or cohabiting
|
0.59
|
0.40-89
|
0.72
|
0.47-1.10
|
0.72
|
0.49-1.06
|
**
|
**
|
**
|
**
|
**
|
**
|
Education
|
|
|
|
|
|
|
|
|
|
|
|
|
≤Primary
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
≥ Secondary
|
1.52
|
1.22-1.89
|
1.20
|
0.91-1.58
|
1.19
|
0.77-1.86
|
**
|
**
|
**
|
**
|
**
|
**
|
Literacy
|
|
|
|
|
|
|
|
|
|
|
|
|
Being illiterate
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
Being illiterate
|
1.23
|
0.98-1.55
|
1.66
|
1.18-2.32
|
1.36
|
0.83-2.22
|
1.16
|
0.91-1.48
|
1.55
|
1.07-2.25
|
1.32
|
0.78-2.23
|
Ever tested
|
|
|
|
|
|
|
|
|
|
|
|
|
No
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
Yes
|
1.74
|
1.40-2.15
|
2.00
|
1.47-2.72
|
1.40
|
0.88-2.20
|
1.48
|
1.18-1.85
|
1.87
|
1.37-2.55
|
1.20
|
0.76-1.90
|
Aware of self-test
|
|
|
|
|
|
|
|
|
|
|
|
|
No
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
1
|
|
Yes
|
1.35
|
0.95-1.92
|
0.96
|
0.69-1.34
|
1.00
|
0.56-1.78
|
1.09
|
0.76-1.55
|
0.94
|
0.66-1.33
|
0.89
|
0.50-1.60
|
* Weighted analysis using standard Demographic Health Survey (DHS) sample weights: Sample size = 7041.0867; Strata = 19; PSU=400. Out of 7420 men surveyed, 7372 reported on willingness to self-test. 48 men did not respond and 1 not provide information on sexual activity (HIV risk). Sexual activity was not reported by 1 respondent and could not be used in HIV risk variable. These variables have a total sample size of 7 371.
**Represents variables which were not included in the multivariable analysis due to identified collinearity.
*** HIV risk is defined in this analysis includes reported sexual activity in the past four weeks, and the following high-risk exposures in the previous 12 months: multiple (i.e. ≥2) partners, any paid sex (asked to men), having received gifts, cash or other compensation in exchange for sex (asked to women), and having a sexually transmitted infection (STI). Individuals with any “high-risk” exposures were classified as “high-risk”, with the remaining respondents classified as “low risk” if reporting no sexually activity in the past four weeks, and as “moderate risk” otherwise.
§ P-value based on Wald Test. P-values for variables with more than two categories are shown.
On multivariable analysis of men at high HIV-related sexual risk, willingness to self-test increased with age (p=0.030) with the strongest association for those aged 35-39 compared to those aged 15-19 (aOR=4.87; 95%CI: 2.14-11.07, p<0.001). Similarly, willingness to self-test among men with high HIV-related sexual -risk increased in rural settings (rural vs urban: aOR=3.56, 95%CI: 1.61-7.90, p=0.002) and with greater wealth quintile (wealthiest vs least wealthy: aOR=3.74, 95%CI:1.39-10.53, p=0.009).
While actively working men with high HIV-related risk were less willingness to self-test (actively working vs not actively working: aOR: 0.57, 95%CI: 0.34-0.95, p=0.030), employed low-risk men were more willing to self-test than unemployed low-risk men (aOR 1.41; 95%CI: 1.13-1.77, p=0.003). The association with previous testing and willingness to test was also more pronounced for low-risk men (ever tested vs never tested: aOR 1.48; 95%CI: 1.18-1.85, p<0.001) than high-risk men (ever tested vs never tested: aOR 1.20; 95%CI: 0.76-1.90, p=0.435), while associations with age (p=0.106) and wealth (p=0.102) were less pronounced than for high-risk men (Table 4, described above).
We additionally conducted a stratified analysis to investigate whether willingness to self-test varied by past HIV-testing behaviour (i.e. previously tested or not) (supplementary table S3). Patterns of willingness to self-test were similar for the 2,437/7,372 (33.1%) men who had never previously tested as for those with at least one past HIV test, with greater willingness in older men.