Reach and Relevance
The sociodemographic characteristics of the sample are presented in Table 2 – total sample of 2 198. The mean age of respondents was 13 years (Standard Deviation: SD = 0.82). There were slightly more males (51%) than females (49%). There was a significant difference in age with more 13-14-year olds (70%) than 10–12 year olds (30%) (p = 0.004) as well as sex (p = 0.01) between those exposed and not exposed to SBC (60% females and 40% males). A quarter had completed grade 5. There were significantly more respondents with grade 7 (last year of primary school) amongst the exposed (4%) compared to those not exposed (22% in grade 5) (p = 0.007).
The survey indicated 12% of the 10-14-year olds respondents were exposed to Soul Buddyz clubs and its booklets for over a year (11% = 354 256). More respondents aged 13–14 years (8.9% N = 248 259) than 10–12 years (3.8% N = 105 998) were reached with the programme. Reach was slightly higher among females (7.6% n = 211 997) than males (5.1% n = 142 261). Four percent of respondents, equating to 108 256 people, reported that they had ever belonged to a SBC. Most respondents who had ever belonged to a SBC indicated that they were part of an SBC for one year (54% N = 55 490).
Table 2
Socio-demographic characteristics of the SBC members vs non members
|
Exposed to SBC
|
Unexposed to SBC
|
Total
|
P-value
|
|
No.
|
%
|
No.
|
%
|
No.
|
%
|
|
Age
|
|
10–12 years
|
78
|
30%
|
746
|
41%
|
832
|
40%
|
0.004
|
13–14 years
|
182
|
70%
|
1073
|
59%
|
1247
|
60%
|
Total
|
260
|
100.0
|
1819
|
100.0
|
2079
|
100.0
|
Sex
|
|
Male
|
104
|
40%
|
928
|
51%
|
1034
|
49.7
|
0.01
|
Female
|
156
|
60%
|
891
|
49%
|
1045
|
50.3
|
Total
|
260
|
100.0
|
1819
|
100.0
|
2079
|
100.0
|
Population group
|
|
Black African
|
229
|
1
|
1510
|
83%
|
1746
|
84%
|
0.27
|
White
|
8
|
0
|
109
|
6%
|
104
|
5%
|
Coloured
|
18
|
0
|
164
|
9%
|
187
|
9%
|
Indian/Asian
|
5
|
0
|
36
|
2%
|
42
|
2%
|
Total
|
260
|
100.0
|
1819
|
100.0
|
2079
|
100.0
|
Province
|
|
Eastern Cape
|
23
|
9%
|
255
|
14%
|
291
|
14%
|
0.07
|
Free State
|
18
|
7%
|
109
|
6%
|
125
|
6%
|
Gauteng
|
81
|
31%
|
364
|
20%
|
437
|
21%
|
KwaZulu-Natal
|
52
|
20%
|
400
|
22%
|
457
|
22%
|
Limpopo
|
29
|
11%
|
236
|
13%
|
249
|
12%
|
Mpumalanga
|
10
|
4%
|
109
|
6%
|
125
|
6%
|
North West
|
23
|
9%
|
127
|
7%
|
146
|
7%
|
Northern Cape
|
5
|
2%
|
36
|
2%
|
42
|
2%
|
Western Cape
|
18
|
7%
|
182
|
10%
|
208
|
10%
|
Total
|
260
|
100.0
|
1819
|
100.0
|
2079
|
100.0
|
Grade
|
|
Grade 1
|
5
|
2%
|
18
|
1%
|
20
|
1%
|
0.007
|
Grade 2
|
5
|
2%
|
54
|
3%
|
41
|
2%
|
Grade 3
|
5
|
2%
|
143
|
8%
|
143
|
7%
|
Grade 4
|
31
|
12%
|
286
|
16%
|
327
|
16%
|
Grade 5
|
59
|
23%
|
447
|
25%
|
511
|
25%
|
Grade 6
|
52
|
20%
|
411
|
23%
|
470
|
23%
|
Grade 7
|
80
|
31%
|
321
|
18%
|
409
|
20%
|
Grade 8
|
21
|
8%
|
107
|
6%
|
123
|
6%
|
Total
|
258
|
100.0
|
1786
|
100.0
|
2044
|
100.0
|
There was strong qualitative evidence supporting the relevance of the programme. Children live in a context where there is drug and alcohol abuse, sexual abuse, lack of communication between children and their parents, HIV and AIDS and exposure to unsafe places. By speaking directly to these issues, SBC was deemed to be relevant as illustrated in the quote below:
Figure 1: SBC FGD, girls 10–12 years, Gauteng, February 2019
The picture above illustrates the perceptions of how SBC activities responded to the health environment of the children who participated in the FGDs. The picture talks about SBC school sessions to discuss the effects of alcohol, and visits to local taverns to discourage sale of alcohol to children. The quotation below further illustrates the relevance of SBC:
‘We talk about having to be responsible towards others. We talk about everything other people have done to us. We go to the garden every day and talk about soul buddyz. Every Tuesday they give us homework to do. We have learnt a lot of things from Soul buddyz.’ SBC FGD mixed 10–12 year olds, Gauteng, January 2019
Key aspects of success and sustained participation include that the clubs are child-led, provide peer support, are contextualised to the issues at hand and engage a passionate facilitator that is approachable and recognises the contribution of young people to their own development, and as change agents in their own lives. The relationships between facilitators and the Club members was also identified as a critical success factor. Where club members felt that they could trust facilitators, they felt that they could disclose things to the facilitator, and they would get help. Club members had the following to say about facilitators:
“I like them – they understand us”. SBC FGD mixed 10–12 year olds, Western Cape, February2019
Table 3
Associations between socio - demographic characteristics, HIV knowledge and stigma, MMC and HIV testing among SBC membership
|
Correct HIV prevention knowledge
|
Medical Male Circumcision
|
Recent HIV test
|
HIV stigmatizing attitudes
|
yes
|
no
|
p value
|
yes
|
no
|
p value
|
yes
|
no
|
p value
|
Has stigma
|
No stigma
|
p value
|
Age
|
|
|
10–12 years
|
68%
|
32%
|
n = 2198
p = .0002
|
32%
|
68%
|
n = 1026 p = .43
|
17%
|
83%
|
n = 1985 p = .06
|
25%
|
75%
|
n = 2069
p = .0002
|
13–14 years
|
77%
|
23%
|
36%
|
64%
|
22%
|
78%
|
16%
|
84%
|
Sex
|
|
|
|
|
Male
|
69%
|
31%
|
n = 2198 p = .003
|
N/A - only males
|
23%
|
77%
|
n = 1985 p = .003
|
20%
|
80%
|
n = 2069 p = .84
|
Female
|
77%
|
23%
|
16%
|
84%
|
81%
|
19%
|
Correct HIV prevention knowledge
|
Exposed to SBC
|
Unexposed to SBC
|
Total
|
P-value
|
|
No.
|
%
|
No.
|
%
|
No.
|
%
|
|
No
|
35
|
13%
|
452
|
25%
|
499
|
24%
|
0.0001
|
Yes
|
235
|
87%
|
1357
|
75%
|
1580
|
76%
|
Total
|
270
|
100
|
1809
|
100
|
2079
|
100
|
HIV stigmatizing attitudes
|
|
|
|
|
No stigma
|
238
|
89%
|
1413
|
79%
|
1665
|
81%
|
0.005
|
|
has stigma
|
29
|
11%
|
376
|
21%
|
391
|
19%
|
|
Total
|
267
|
100
|
1789
|
100
|
2056
|
100
|
|
Ever had sex
|
|
|
|
|
|
No
|
260
|
98%
|
1754
|
99%
|
1996
|
98%
|
0.33
|
|
Yes
|
5
|
2%
|
18
|
1%
|
41
|
2%
|
|
Total
|
265
|
100
|
1772
|
100
|
2037
|
100
|
|
Condom use at first sex
|
|
|
|
|
|
Did not
|
2
|
49%
|
8
|
46%
|
11
|
47%
|
no p-value
|
|
Used condom
|
3
|
51%
|
10
|
54%
|
12
|
53%
|
|
Total
|
5
|
100
|
18
|
100
|
23
|
100
|
|
Medical Male Circumcision
|
|
|
|
|
|
No
|
46
|
45%
|
625
|
68%
|
674
|
66%
|
0.0005
|
|
Yes
|
56
|
55%
|
294
|
32%
|
347
|
34%
|
|
Total
|
102
|
100
|
919
|
100
|
1021
|
100
|
|
Recent HIV test
|
|
|
|
|
|
|
No
|
187
|
73%
|
1391
|
81%
|
1579
|
80%
|
0.02
|
|
Yes
|
69
|
27%
|
326
|
19%
|
395
|
20%
|
|
Total
|
257
|
100
|
1717
|
100
|
1974
|
100
|
|
HIV Status
|
|
|
|
|
|
Negative
|
186
|
98%
|
1234
|
97%
|
1418
|
97%
|
0.97
|
|
Positive
|
4
|
2%
|
38
|
3%
|
44
|
3%
|
|
Total
|
190
|
100
|
1272
|
100
|
1462
|
100
|
|
Knowledge and psychosocial skills outcomes
HIV Knowledge
Seventy-three percent of the respondents demonstrated correct HIV prevention knowledge measured using a knowledge scale which combined a set of questions on knowledge and perceptions of HIV/AIDS. A continuous HIV knowledge variable was created from 10 questions that looked at HIV knowledge. There was very strong evidence of an association between HIV knowledge and exposure to SBC with more respondents that were exposed to SBC reported having correct knowledge (87%) in comparison to those not exposed to the programme (75%) (p < 0.0001) (see Table). This finding was supported by the qualitative data with SBC members reporting that Soul Buddyz had improved their knowledge of HIV, how to protect themselves from contracting HIV, importance of condom use and identifying myths and misinformation about HIV such as that you can see by looking at someone if they have HIV as illustrated in the quotes below.
“I’ve learnt about sex and not to go too deep – not to have a sexual relationship”. SBC FGD mixed 10–12 year olds, Western Cape, February 2019
“As a girl I learned not to be sexually active and to protect myself from boys (by not having sex or using condoms if I have sex)”. SBC FGD mixed 10–12 year olds, Western Cape, February 2019
“The book (Soul Buddyz Booklet) tells us about looking after ourselves, gender and protecting ourselves from diseases like HIV and AIDS”. SBC FGD, Boys, 10–12 years, Mpumalanga, November 2018
Soul Buddyz had also helped members identify and address myths and incorrect information about HIV and how HIV is acquired. Club members noted that one of the myths they had heard which was corrected through Soul Buddyz was:
“People get AIDS if they walk where someone spit and they have the sickness”. SBC FGD mixed 10–12 years olds, Western Cape, February 2019
HIV stigmatizing attitudes
A total of 20% of teenagers reported to have stigmatizing attitudes measured using a knowledge scale which combined a set of questions on HIV/AIDS stigma. A continuous HIV stigma variable was created from six questions that looked at attitudes towards people living with HIV. There was strong evidence of association in reporting stigmatizing attitudes between those exposed to SBC and those unexposed (p = 0.005) with more respondents exposed to SBC reporting not having stigmatizing attitudes (89%) compared to those not exposed (79%) -(see Table 3). Club members from the qualitative study reported having learnt how people can be affected by stigma and misinformation. See Fig. 2 for a drawing depicting a discussion on HIV and AIDS and going to the hospital.
“Yes, some children are born HIV positive and when other children find out about their status they do not want to play with them. We play with those children because they are like the rest of us. We also play with them because we will also need help one day.” SBC FGD mixed gender 10–12 years, KZN, January 2019.
Figure 2: SBC FGD, boys 10–12 years, Mpumalanga, November 2018
Risky sexual behaviour
Sexual activity and condom use
Some 1.5% respondents reported to have ever had sex. There was an insignificant association between sexual activity measured through “ever having sex” and exposure to SBC p = 0.33 (see Table 3). Some 53% respondents reported to have used a condom the first time they had sex. There were too few observations to measure the association between exposure to SBC and condom use at first sex. Qualitative findings further highlighted that children learned about safe sex and the use of condoms from SBCs.
“If someone has HIV you won’t know just by looking at them. If you can’t get tested then rather just use condoms. Yes, safe sex. They teach us about that in Soul Buddyz” SBC FGD, female 13–14 years, Western Cape, January 2019
Access to biomedical intervention
Medical Male Circumcision
Thirty four percent (34%) of males reported having had MMC performed by a doctor. There was very strong evidence of an association between medical male circumcision and exposure to SBC (p = 0.0005). More males exposed to SBC reported being medically circumcised than those not exposed (55% versus 32%) (see Table 3).
HIV testing and status
A total of 20% of the respondents reported to have been tested for HIV. There was evidence of an association between HIV testing and exposure to SBC (p = 0.02). More adolescents exposed to SBC (27%) reported taking an HIV test compared to those not exposed to SBC (19%). Moreover, some 3% respondents tested positive for HIV – fewer of those exposed to SBC tested negative 2% compared to 3% who were unexposed to SBC but this was not statistically significant (p = 0.97) (see Table 3).
Multivariate analysis findings: The relationship between risky sexual behaviours, HIV status, MMC and exposure to SBC
Table 4
Adjusted odds ratio for behavioural and biomedical outcomes in exposure to SBC estimated by logistic regression
|
Univariate regression OR (95% CI)
|
N
|
Exposure to SBC
|
Unadjusted Odds ratio
|
Adjusted Odds ratio’
|
95% CI
|
P-value
|
Sample size
|
Confounding variables
|
|
|
|
|
|
|
|
|
|
Correct HIV prevention knowledge
|
|
|
|
|
|
|
2.35* (1.49–3.68)
|
2.21
|
1.36–3.57
|
0.001
|
2 079
|
Age
|
|
2079
|
|
|
|
|
|
|
13–14 years
|
1.66*** (1.17–2.35)
|
|
|
HIV stigmatizing attitudes
|
|
|
10–12 years
|
0.60*** (0.43–0.85)
|
|
0.47*** (0.27–0.81)
|
0.54
|
0.31–0.93
|
0.025
|
2 056
|
Province
|
|
2079
|
0.98**
(0.91–1.04)
|
|
Medical Male Circumcision
|
|
|
Eastern Cape
|
0.93** (0.39–2.19)
|
|
2.59* (1.49–4.52)
|
2.38
|
1.29–4.40
|
0.006
|
1 004
|
Free State
|
1.66** (0.75–3.69)
|
|
|
|
|
|
|
Gauteng
|
2.32" (1.13–4.74)
|
|
|
|
|
|
|
KwaZulu-Natal
|
1.32** (0.68–2.59)
|
|
|
|
|
|
|
Limpopo
|
1.24** (0.55–2.78)
|
|
|
Recent HIV test
|
|
|
Mpumalanga
|
0.96** (0.41–2.25)
|
|
1.57**(1.07–2.32)
|
1.48
|
0.97–2.25
|
0.06
|
1 974
|
North West
|
1.96** (0.95–4.06)
|
|
|
|
|
|
|
Northern Cape
|
1.33** (0.63–2.83)
|
|
|
|
|
|
|
Western Cape
|
1.-8**(0.46–2.56)
|
|
|
|
|
|
|
Grade
|
|
2044
|
1.22*** (1.06–1.41)
|
|
Ever had sex
|
|
|
|
Grade 1
|
1.04** (0.19–5.55)
|
|
1.82** (0.53–6.31)
|
1.44
|
0.42–4.88
|
0.56
|
1 916
|
Grade 2
|
0.36** (0.49–2.69)
|
|
|
|
|
|
|
Grade 3
|
0.18" (0.03–1.01)
|
|
|
|
|
|
|
Grade 4
|
0.43** (0.08–2.35)
|
|
|
|
|
|
|
Grade 5
|
0.50** (0.09–2.72)
|
|
|
Ever consuming alcohol
|
|
|
Grade 6
|
0.49** (0.09–2.61)
|
|
1.67** (0.92–3.05)
|
1.47
|
0.79–2.71
|
0.22
|
2 078
|
Grade 7
|
0.96** (0.18–5.09)
|
|
|
|
|
|
|
Grade 8
|
0.75** (0.12–4.82)
|
|
|
|
|
|
|
Exposure to other media
|
|
|
|
|
HIV Status
|
|
|
|
Yes
|
4.05*** (1.61–10.19)
|
2079
|
|
0.98** (0.31–3.10)
|
0.95
|
0.29–3.11
|
0.93
|
1 462
|
No
|
0.25*** (0.09–0.62)
|
|
|
|
|
|
|
|
"p < 0.05
* p = 0.001
**p > 0.05
***p < 0.01
|
|
|
|
’controlling for all confounders except for MMC that excluded gender (age, province, exposure to media)
|
The multivariate regression analysis models presented in Table 4 reveals that after controlling for age, sex, level of education, province and exposure to other media, SBC exposure was significantly associated with medical circumcision (AOR 2.38; 95%CI 1.29–4.40, p = 0.006), HIV knowledge (AOR 2.21; 95%CI 1.36–3.57, p < 0.001) and less stigmatising attitudes (AOR 0.54; 95%CI 0.31–0.93, p = 0.006). Addressing selection bias in trying to estimate the effect of the intervention, we conducted a PSM (see Table 5) – the PSM findings are consistent with the regression findings showing that the probability of having stigma is -0.09 percentage point (0.9% point) reduced; and a 0.26% point increase in doing MMC and 0.102% point increase in having correct HIV prevention knowledge, in the exposed as opposed to those unexposed.
Table 5
Propensity score matching treatment effect of exposure to SBC for behavioural and biomedical outcomes
Average treatment effect
|
|
% frequency
|
Coefficient
|
95% CI
|
P value
|
Correct HIV prevention knowledge
|
SBC exposure
|
15%
|
.102
|
0.101–0.103
|
< 0.0001
|
|
unexposed
|
85%
|
|
|
|
Medical Male Circumcision
|
SBC exposure
|
16%
|
.229
|
0.226–0.231
|
< 0.0001
|
|
unexposed
|
84%
|
|
|
|
HIV stigmatizing attitudes
|
SBC exposure
|
7%
|
− .092
|
(-)0.093 - (-)0.091
|
< 0.0001
|
|
unexposed
|
93%
|
|
|
|
Recent HIV test
|
SBC exposure
|
17%
|
.056
|
0.054–0.057
|
< 0.0001
|
|
unexposed
|
83%
|
|
|
|
HIV Status (positive)
|
SBC exposure
|
13%
|
− .0040296
|
(-)0.004 - (-)0.003
|
< 0.0001
|
|
unexposed
|
87%
|
|
|
|
Ever had sex
|
SBC exposure
|
21%
|
.008
|
0.007–0.008
|
< 0.0001
|
|
unexposed
|
79%
|
|
|
|
Ever consuming alcohol
|
SBC exposure
|
19%
|
.011
|
0.010–0.012
|
< 0.0001
|
|
unexposed
|
81%
|
|
|
|
Adjusting for age, sex, province and exposure to other media, adolescents exposed to SBC have a positive but insignificant association with HIV testing (AOR 1.48 95%CI 0.97–2.25 p = 0.06) and a negative HIV test outcome with 5% decreased (less) odds of being HIV positive compared to those not exposed (AOR 0.95 95%CI 0.29–3.11, p = 0.93).
While, teenagers exposed to SBC have 1.44 odds of having had sex compared to those not exposed, this was statistically insignificant (AOR 1.44 95%CI 0.42–4.88; p = 0.56). There were too few observations to draw any significant results for condom use at first sex among those exposed to SBC compared to those unexposed. In summary, this analysis has demonstrated that being a Soul Buddyz Club member has a positive effect on biomedical HIV prevention uptake specifically MMC and HIV testing, HIV prevention knowledge and HIV related stigma outcomes. Qualitative findings support some of these findings with SBC members indicating having learnt life skills related to HIV prevention through participation in the clubs.