Abstract
Young heterosexual men have low uptake of HIV prevention and treatment services and represent an important key population that may require novel strategies. We recruited 1271 heterosexual men, 12 years and older from socializing venues such as “shebeens”, transport hubs, “spaza” shops, and community centers in rural KwaZulu-Natal, South Africa. Participants completed a questionnaire and were tested for HIV serostatus. Generalized estimating equations (GEE) with exchangeable covariance structure estimated factors independently associated with prevalent HIV infection. Median age was 25 years [Interquartile range (IQR) 21–29]. HIV prevalence was 15.5% [95% confidence interval (CI) 11.0–21.9] and increased significantly by age. Factors associated with a higher odds of HIV infection were being 25 years and older [adjusted odds ratio (aOR)=4.82, 95% CI 3.47–6.69; p<0.001), not completing high school (aOR=1.60, 95% CI 1.39–1.85; p<0.001), not using condoms at first sex (aOR=1.43, 95% CI 1.20–1.70; p<0.001), consuming alcohol (aOR=1.63, 95% CI 1.15–2.31; p=0.006) or substances (aOR=1.37, 95% CI 1.31–1.44; p<0.001), and absence of medical circumcision (aOR=2.05, 95% CI 1.71–2.44; p<0.001). Risk was lower among those testing for HIV in last 12 months (aOR= 0.54, 95% CI 0.36–0.80; p=0.002). Greater effort is needed to implement innovative programs within settings that are easily accessible and where heterosexual men are likely to be.
Keywords: KwaZulu-Natal, heterosexual men, HIV prevalence, risk factors
Introduction
In 2018 an estimated 7.9 out of ~57 million South Africans were living with human immunodeficiency virus (HIV). Despite the scale-up of HIV prevention and treatment programs, 270,000 new HIV infections and 110,000 HIV related deaths occurred that year. The predominant mode of HIV acquisition in this region is through heterosexual transmission. The national and regional HIV sero-prevalence surveys show the marked sex-age differences in the burden of HIV [1, 2] with young women acquiring HIV at least 5 to 7 years earlier than their males counterparts, though prevalence increases among men 25 years and older [3, 4]. Several studies have provided empirical evidence that sexual partnering between older men and younger women perpetuates HIV transmission [5, 6]. In the KwaZulu-Natal region of South Africa extensive epidemiologic community defined sampling and phylogenetic analyses of HIV-1 sequences have provided a better understanding of underlying HIV transmission dynamics. Specifically, the findings provide evidence for a cycle of heterosexual HIV transmission in which young women 15–24 years were phylogenetically linked to men who on average were 8.7 years older, whilst women 25 to 40 years were linked to men who on average were 1.1 years older. Majority of these men were unaware of their HIV-positive status, were not on antiretroviral therapy and had high HIV viral load [5]. These data provide HIV risk associated with age disparate sexual partnerships [4, 7] and with different types of sexual networks [5, 6] simultaneously elevating older men’s risk for HIV acquisition. Furthermore, several studies have highlighted behaviors such as early sexual debut, inconsistent condom use, multiple concurrent or sequential sexual partnerships that enable and elevate HIV risk among women, however, the data for men are limited [2, 4, 8, 9].
In most cultures, men have been decision makers in matters related to sexual reproductive health [10] even though their own engagement with healthcare services is poor [11]. In South Africa, masculinity, complex gender norms, and inequalities influence health care engagement and/or unwillingness with HIV testing and care services. Masculinity constructs typically focus on displaying strength, control, being sexually competent, earning an income, and loss of any of these can create fear of loss of masculinity [12]. Current programs are generally designed for women and do not adequately engage men to support their sexual, reproductive health care needs [13]. Furthermore, men are often underrepresented in HIV prevention programs, yet form a crucial link in the spread of HIV infection. Thus, the need to create and increase opportunities to effectively engage and support men and their partners to access programs to address their sexual reproductive health needs including prevention efforts for sexually transmitted infections and HIV [14].
Whilst young heterosexual men are less likely to access sexual reproductive health services [10], it is imperative to engage and undertake research among them to better understand the risk associated behaviors and the epidemiology of HIV, which in turn is crucial to the design and customization of interventions towards achieving HIV epidemic control. Given the numerous studies that have been undertaken to understand individual-level risk taking behaviors driving the epidemic in South Africa, there is evidence that many of these behaviors are influenced and perpetuated around the broader socio-structural features of the individuals’ environment [15]. Certain environments such as socializing venues which may serve alcohol, either under government regulation or discreetly, and involve sex work or transactional sex, and therefore may be associated with risk-taking behaviors [15, 16]. Furthermore, with the high levels of unemployment and poverty [3], many individuals supplement their income through discreet sale of alcohol in low-income communities to socialize and to potentially meet sex partners [16]. Therefore, this study aimed to provide insights into HIV prevalence among heterosexual men at various socializing venues, and factors associated with HIV infection in a high HIV burden setting in rural KwaZulu-Natal, South Africa.
Methods
Study setting
This study was undertaken in rural Vulindlela, a community situated in the uMgungundlovu District of Kwa Zulu-Natal. The area has approximately 85,000 households with a population of about 150,000. Whilst the community has access to basic utility services such as water, sanitation, and electricity, the area remains under-developed and is characterised by a high population HIV prevalence, high levels of unemployment, poverty and limited access to income generating opportunities [3]. Health care is provided predominantly through public sector primary health care clinics and community-based organisations.
Study population, design, and recruitment
Men were recruited through four community-based socializing venues, including: 1) “shebeens” – these are informal alcohol serving establishments also known as taverns that have recently been government regulated and may include car washes, which are not regulated to serve alcohol, but do so discreetly; 2) transport hubs – these are informal areas that are designated hubs where the paratransit system have established collection points to be utilised by minibus taxis to await and collect passengers, typically used by most township residents as a quick and cheap means of commuting to and from home, work, shops and schools. The industry has developed fulfilling a critical role for the provision of rapid transportation and has built a powerful second economy. Being the hub of the transportation system, there is considerable interaction between the providers and users of the taxi services which could potentially lead to risk-taking behaviors; 3) “spaza” shops – these are informal convenience shops that supplement household income and are an important supplier of household goods to the community in the absence of large retail outlets. Against the background of high rates of unemployment in this rural area, many individuals tend to frequent these spaza shops and socialize around these venues; 4) community centers – these venues have been set up to provide a range of training and life-skills opportunities to rural community members and are generally common socializing venues.
At each venue, men were randomly approached and provided with study information in English and/or isiZulu, and those agreeing to continue with study participation were deemed eligible if they were 12 years and older, not attending primary, secondary or high school and had not participated in a similar study in the past year. Participants were provided with study information and written informed consent was obtained for those 18 years and older whilst assent and parental consent for those younger than 18 years of age was obtained.
Study procedures
To ensure privacy, field staff undertook face to face interviews to obtain socio-demographic, behavioral, clinical and HIV testing related information. Questions were embedded on a personal digital assistant (PDA) mobile electronic device by Mobenzi Researcher (Durban, South Africa). To maintain confidentiality, participants were assigned a unique participant identification (PID) number. Using an electronic biometric system linked to the PDA participants provided finger-prints to verify identity and exclude potential duplicate enrolment into the study. These finger-prints were stored in a database separately from the questionnaire and laboratory data. Global positioning system (GPS) obtained co-ordinates of the recruitment site to locate each socializing venue. All participants provided finger-prick blood samples into BD Microtainer® (Becton Dickinson, South Africa) blood collection tubes for laboratory measurements of HIV antibodies.
Laboratory procedures
HIV antibody testing was performed in a central laboratory using the fourth generation HIV enzyme Biomerieux Vironostika Uniform II Antigen/Antibody Microelisa system (BioMérieux, Marcy I’Etoile, France). All reactive samples were confirmed positive with the HIV 1/2 Combi Roche Elecys (Roche Diagnostics, Penzberg, Germany). All indeterminate results were resolved with ADVIA Centaur HIV Antigen/Antibody Combo (CHIV) Assay (Siemens, Tarry Town, NY, USA). HIV test results with PIDs and corresponding barcodes were sent to the local primary health care clinics for management of participants accessing laboratory results. Participants were provided with clinic referral cards to access their HIV test results and to receive follow-up care.
Study approvals
This study was approved by the University of KwaZulu-Natal Biomedical Research Ethics Committee (BREC, REF # BF001/16) and additional approval was obtained from the University of Pennsylvania, Institution Review Board (REF # 825071). Permissions to access the local community and socializing venues were obtained from the local authority and venue owners or representatives.
Statistical analysis
Questionnaire data and HIV test results were merged using a unique linking number. The demographic, behavioral and clinical characteristics were summarized using descriptive summary measures, expressed as proportions for categorical variables and central measures of tendency for medians with interquartile range (IQR) for continuous variables. We considered men to be more similar within each socializing venue and therefore treated each as an independent cluster. HIV prevalence in each venue was calculated as the number of participants who were HIV positive, divided by the total number of enrolled participants from the venue. The Wilson Score Test was used to calculate confidence interval for crude HIV prevalence. The Cochran–Armitage chi-square test was used to assess for linear trends in HIV prevalence across age group within each venue. The cluster-adjusted HIV prevalence was calculated as a geometric mean of cluster specific prevalence and the confidence intervals were based on t-distribution. This was achieved by taking the logarithmic of cluster-specific prevalence to calculate the arithmetic mean and the 95% confidence interval (95% CI), where the anti-logarithm provided the geometric mean and the 95% CI. The advantage of the geometric mean is that results are less sensitive to extreme observations.
The full dataset was used to assess the independent factors associated with HIV infection among heterosexual men. To account for clustering within each venue, generalized estimating equations (GEE) with exchangeable covariance structures were used to estimate the magnitude of the association between HIV status and socio demographic, behavioral, and clinical characteristics, and to derive the unadjusted and adjusted odds ratio (OR) with corresponding 95% CI. Quasi-likelihood information under the independence model criterion (QIC) was used for determining the best fitting correlation structure. Inclusion of variables in multivariable models was based on subject knowledge and biological plausibility. However, some variables were excluded in the model to avoid multicollinearity. Statistical analyses were performed using Base SAS (SAS Institute, Cary, North Carolina) version 9.4.
Results
Participants characteristics
Between 06 September 2017 and 01 June 2018, a total of 1271 men were enrolled and had blood samples collected. Of these, 489 (38.5%) were enrolled from shebeens, 451 (35.5%) from transport hubs, 213 (16.8%) from spaza shops and 118 (9.3%) from community centers. All participants were analysed for HIV outcome, whilst 1242 (97.7%) completed the questionnaire and were analysed for socio-demographic, behavioral and clinical outcomes. Table 1 shows the characteristics of enrolled participants overall and by socializing venue. Median age of the participants was 25 (interquartile range (IQR) 21–29) years. Nearly half (43.3%) reported not completing high school, 56.7% were unemployed, and only 10.6% were currently studying in a tertiary institution. Nearly all (98.4%) reported their relationship status to be single. About one-third (29.8%) reported their first sex occurred before age 16 years and about a third (32.7%) reported condom use. The median number of lifetime sex partners reported was 3 (IQR 2–4). About two thirds (59.5%) reported consuming alcohol, whilst just over a third (36.4%) reported substance use. Overall, 87.4% ever had an HIV test, whilst 59.7% had tested for HIV within the last 12 months. Just over half (57.3%) reported being medically circumcised.
Table 1.
Socializing venue-based characteristics of enrolled heterosexual men in rural KwaZulu-Natal, South Africa
| Characteristic | Shebeens | Transport hubs | Spaza shops | Community centres | Total | |
|---|---|---|---|---|---|---|
| (n=489) | (n=451) | (n=213) | (n=118) | Unadjusted | Adjusteda | |
| Socio demographic | ||||||
| Age (median, IQR ) | 24 (20 – 28) | 26 (22–30) | 23 (20–28) | 25 (22–30) | 25 (21–29) | |
| Age group in years (n, %) | ||||||
| 14–19 | 92 (18.8) | 26 (5.8) | 35 (16.4) | 4 (3.4) | 157 (12.4) | 157 (8.8) |
| 20–24 | 180 (36.8) | 146 (32.4) | 93 (43.7) | 46 (39.0) | 465 (36.6) | 465 (37.8) |
| 25–29 | 118 (24.1) | 149 (33.0) | 46 (21.6) | 37 (31.4) | 350 (27.5) | 350 (27.1) |
| 30–50 | 99 (20.2) | 130 (28.8) | 39 (18.3) | 31 (26.3) | 299 (23.5) | 299 (23) |
| Education level (n, %)a | ||||||
| Incomplete high school | 207 (43.4) | 218 (49.5) | 103 (48.6) | 38 (33.6) | 566 (45.6) | 566 (43.3) |
| Completed high school | 270 (56.6) | 222 (50.5) | 109 (51.4) | 75 (66.4) | 676 (54.4) | 676 (55.9) |
| Employment status (n, %)a | ||||||
| Student at tertiary institution | 38 (8.0) | 27 (6.1) | 17 (8.0) | 37 (32.7) | 119 (9.6) | 119 (10.6) |
| Unemployed | 314 (65.8) | 253 (57.5) | 152 (71.7) | 43 (38.1) | 762 (61.4) | 762 (56.7) |
| Employed | 125 (26.2) | 160 (36.4) | 43 (20.3) | 33 (29.2) | 361 (29.1) | 361 (27.4) |
| Relationship status (n, %)a | ||||||
| Never married | 468 (98.1) | 429 (97.5) | 210 (99.1) | 112 (99.1) | 1219 (98.1) | 1219 (98.4) |
| Ever married | 9 (1.9) | 11 (2.5) | 2 (0.9) | 1 (0.9) | 23 (1.9) | 23 (1.4) |
| Duration of residing in the community (n, %)a | ||||||
| ≥ 5 years | 452 (94.8) | 406 (92.3) | 203 (95.8) | 87 (77) | 1148 (92.4) | 1148 (89.6) |
| < 5 years | 25 (5.2) | 34 (7.7) | 9 (4.2) | 26 (23) | 94 (7.6) | 94 (7.9) |
|
| ||||||
| Behavioural | ||||||
| Ever had sex (n, %)a,b | ||||||
| Yes | 383 (80.3) | 400 (90.9) | 178 (84.0) | 93 (82.3) | 1054 (84.9) | 1054 (84.3) |
| No | 94 (19.7) | 40 (9.1) | 34 (16.0) | 20 (17.7) | 188 (15.1) | 188 (15.0) |
| Age at first sex (n, %)a | ||||||
| Never had sex | 95 (19.9) | 40 (9.1) | 35 (16.5) | 20 (17.7) | 190 (15.3) | 190 (15.2) |
| <16 years | 120 (25.2) | 164 (37.3) | 54 (25.5) | 37 (32.7) | 375 (30.2) | 375 (29.8) |
| ≥16 years | 250 (52.4) | 227 (51.6) | 109 (51.4) | 52 (46) | 638 (51.4) | 638 (50.3) |
| Refused | 12 (2.5) | 9 (2) | 14 (6.6) | 4 (3.5) | 39 (3.1) | 39 (3.3) |
| Condom use at first sex (n, %)a, e | ||||||
| Yes | 124 (32.6) | 100 (25.2) | 64 (36.6) | 35 (38.0) | 323 (30.9) | 323 (32.7) |
| No | 256 (67.4) | 297 (74.8) | 111 (63.4) | 57 (62.0) | 721 (69.1) | 721 (66.7) |
| Lifetime number of sex partners (median, IQR)a | 3 (2–4) | 4 (3–4) | 3 (2–4) | 3 (2–4) | 3 (2–4) | |
| Current number of sex partners (median, IQR)a | 1 (1–2) | 1 (1–2) | 1 (1–2) | 1 (1–2) | 1 (1–2) | |
| Ever consumed alcohol (n, %)a | ||||||
| Yes | 293 (61.4) | 274 (62.3) | 122 (57.5) | 64 (56.6) | 753 (60.6) | 753 (59.5) |
| No | 184 (38.6) | 166 (37.7) | 90 (42.5) | 49 (43.4) | 489 (39.4) | 489 (40.5) |
| Any Substance use (n, %)a,c | ||||||
| Yes | 123 (25.9) | 216 (49.4) | 86 (40.6) | 38 (33.9) | 463 (37.5) | 463 (36.4) |
| No | 351 (74.1) | 221 (50.6) | 126 (59.4) | 74 (66.1) | 772 (62.5) | 772 (61.9) |
|
| ||||||
| HIV Testing history | ||||||
| Ever had an HIV test (n, %)a | 394 (82.8) | 376 (88.3) | 182 (85.8) | 105 (92.9) | 1057 (86.1) | 1057 (87.4) |
| Last HIV test done (n, %)a | ||||||
| >12 months | 189 (48.0) | 160 (42.6) | 78 (42.9) | 27 (25.7) | 454 (43.0) | 454 (38.7) |
| ≤12 months | 205 (52.0) | 216 (57.4) | 104 (57.1) | 78 (74.3) | 603 (57.0) | 603 (59.7) |
|
| ||||||
| Biological | ||||||
| Ever had any genital symptoms (n, %)a,f | ||||||
| Yes | 78 (16.4) | 111 (25.2) | 33 (15.6) | 20 (17.7) | 242 (19.5) | 242 (18.4) |
| No | 399 (83.6) | 329 (74.8) | 179 (84.4) | 93 (82.3) | 1000 (80.5) | 1000 (81.2) |
| Medically circumcised (n, %)a,d | 276 (58.1) | 251 (57.0) | 108 (50.9) | 75 (66.4) | 710 (57.3) | 710 (57.8) |
=excludes 29 men who did not respond to questionnaire data
=Missing data for 33 observations
=Missing data for 36 observations
=Missing data for 31
=excludes those reporting never having had sexual intercourse
= reported any symptoms such as genital discharge and/ or ulceration
=adjusted for recruitment sites; IQR=Interquartile Range; Percentages exclude missing data
Participant characteristics according to socializing venues
As shown in table 1, several differences in the socio-demographic and behavioral characteristics were observed among men recruited from different socializing venues. Men from shebeens and spaza shops were younger compared to men from transport hubs and from community centers. A higher proportion of men from community centers had completed high school education compared to men recruited from other venues, whilst the highest proportion of employed men were from spaza shops. There were no differences in the relationship status across the socializing venues as more than 95% reported being never married. More men from transport hubs and spaza shops compared to shebeens and community centers reported ever having had sex, whilst a higher proportion of men from transport hubs and community centers had first sex at age <16 years. Condom use at first sex was the lowest among men from community centers compared to the other socializing venues. Alcohol consumption was similar for men across all venues, whilst substance use was higher among men from transport hubs. A higher proportion of men from transport hubs versus the other sites reported to have presence of genital symptoms at least once in their lifetime. The lowest proportion of men from shebeens had ever had an HIV test and were less likely to receive an HIV test result in the past 12 months. The proportion of medically circumcised men from community centers was slightly higher compared to other sites.
HIV prevalence
Table 2 shows the HIV prevalence, both crude and adjusted by venue and stratified by age. Overall HIV prevalence was 15.5% (95% CI 11.0–21.9); among men recruited from transport hubs, prevalence was 20.2%, in spaza shops, it was 16.0%, in shebeens it was 15.1%, and in community centers it was 11.9%. Age-stratified HIV prevalence was 4.2% (95% CI 0.7–23) among men 15–19 years, 5.6% (95% CI 1.9–16.5) among men 20–24 years, 16.7% (95% CI 8.3–33.9) among men 25–29 years, and 32.8% (95% CI 17.8–60.4) among men 30 years and older. HIV prevalence increased significantly by age among men in shebeens (p<0.001), transport hubs (p<0.001), spaza shops (p<0.001) and community centers (p=0.015).
Table 2.
HIV prevalence overall, by socializing venues and stratified by age group
| Socializing venues | Total | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||||
| Shebeens | Transport hubs | Spaza shops | Community centres | Unadjusted | Adjusteda | ||||||
| n/N | % (95%CI) | n/N | % (95%CI) | n/N | % (95%CI) | n/N | % (95%CI) | n/N | % (95%CI) | % (95%CI) | |
|
| |||||||||||
| Overall | 74/489 | 15.1 (12.2–18.6) | 91/451 | 20.2 (16.7–24.1) | 34/213 | 16.0 (11.7–21.5) | 14/118 | 11.9 (7.2–18.9) | 213/1271 | 16.7 (14.8–18.9) | 15.5 (11.0–21.9) |
| Age specific | |||||||||||
| 14–19 | 2/92 | 2.2 (0.6–7.6) | 1/26 | 3.8 (0.7–18.9) | 3/35 | 8.6 (3.0–22.4) | 0/4 | - | 6/157 | 3.8 (1.8–8.1) | 4.2 (0.7–23) |
| 20–24 | 10/180 | 5.6 (3.0–9.9) | 14/146 | 9.6 (5.8–15.5) | 8/93 | 8.6 (4.4–16.1) | 1/46 | 2.2 (0.4–11.3) | 33/465 | 7.1 (5.1–9.8) | 5.6 (1.9–16.5) |
| 25–29 | 27/118 | 22.9 (16.2–31.2) | 31/149 | 20.8 (15.1–28.0) | 4/46 | 8.7 (3.4–20.3) | 7/37 | 18.9 (9.5–34.2) | 69/350 | 19.7 (15.9–24.2) | 16.7 (8.3–33.9) |
| 30–50 | 35/99 | 35.4 (26.6–45.2) | 45/130 | 34.6 (27.0–43.1) | 19/39 | 48.7 (33.9–63.8) | 6/31 | 19.4 (9.2–36.3) | 105/299 | 35.1 (29.9–40.7) | 32.8 (17.8–60.4) |
|
| |||||||||||
| P value assessing trend b | P<0.001 | P<0.001 | P<0.001 | P=0.015 | |||||||
= adjusted for socializing venues
= Cochran–Armitage χ2 test for linear trend in HIV prevalence
Factors associated with HIV infection
Table 3 shows the factors associated with HIV infection. In multivariable analyses, factors associated with higher odds of HIV infection were being 25 years and older (aOR=4.82, 95% CI 3.47–6.69; p<0.001), not completing high school (aOR=1.60, 95% CI 1.39–1.85; p<0.001), not using condoms at first sex (aOR=1.43, 95% CI 1.20–1.70; p<0.001), ever having consumed alcohol (aOR=1.63, 95% CI 1.15–2.31; p=0.006), ever using substances (aOR=1.37, 95% CI 1.31–1.44; p<0.001), and those with intact foreskin (aOR=2.05, 95% CI 1.72–2.44; p<0.001). In contrast, having received an HIV test in last 12 months (aOR= 0.54, 95% CI 0.36–0.80; p =0.002) was associated with a lower HIV prevalence. HIV prevalence for different sociodemographic, behavioral and clinical characteristics for each of the socializing venues is shown in Supplementary table 1 (Supplemental Digital Content 1).
Table 3.
Generalized Estimating Equation analyses for factors associated with HIV prevalence among heterosexual men in rural KwaZulu-Natal, South Africa
| Characteristics | Unadjusted OR 95%CI | P value | Adjusted OR 95%CI | P value |
|---|---|---|---|---|
|
| ||||
| Age group | ||||
| 25–50 years | 5.43 (3.69–7.99) | <0.001 | 4.82 (3.47–6.69) | <0.001 |
| 14–24 years | Ref | |||
| Education level | ||||
| Incomplete high school | 1.68 (1.31–2.15) | <0.001 | 1.60 (1.39–1.85) | <0.001 |
| Completed high school | Ref | |||
| Employment status | ||||
| Student at tertiary institution | 0.27 (0.19–0.38) | <0.001 | ||
| Unemployed | 0.71 (0.52–0.98) | 0.037 | ||
| Employed | Ref | |||
| Relationship status | ||||
| Never married | 0.42 (0.17–1.05) | 0.063 | 0.86 (0.41–1.79) | 0.688 |
| Ever married | Ref | |||
| Duration of residing in the community | ||||
| ≥5 years | 1.94 (0.61–6.10) | 0.259 | ||
| <5 years | Ref | |||
| Ever had sex | ||||
| Yes | 1.78 (1.08–2.92) | 0.023 | ||
| No | Ref | |||
| Age at first sex | ||||
| <16 years | 1.25 (0.95–1.64) | 0.113 | ||
| Never had sex | 0.61 (0.37–1.00) | 0.051 | ||
| Refused | 0.98 (0.44–2.19) | 0.963 | ||
| ≥16 years | Ref | |||
| Condom use at first sex | ||||
| Never had sex | 0.87 (0.57–1.33) | 0.525 | 0.80 (0.49–1.32) | 0.382 |
| No | 1.84 (1.50–2.26) | <0.001 | 1.43 (1.20–1.70) | <0.001 |
| Yes | Ref | |||
| Lifetime number of sex partners | ||||
| Between 2–5 partners | 1.24 (0.92–1.66) | 0.156 | ||
| More than 5 partners | 1.90 (1.67–2.16) | <0.001 | ||
| Don’t know/remember | 1.78 (1.48–2.14) | <0.001 | ||
| Never had sex | 0.85 (0.54–1.34) | 0.486 | ||
| One partners | Ref | |||
| Current number of sex partners | ||||
| 1 partner | 1.77 (1.07–2.90) | 0.025 | ||
| 2 or more partners | 1.80 (1.26–2.57) | 0.001 | ||
| Never had sex | 0.95 (0.46–1.97) | 0.892 | ||
| 0 Partner | Ref | |||
| Ever consumed alcohol | ||||
| Yes | 1.67 (1.27–2.18) | <0.001 | 1.63 (1.15–2.31)) | 0.006 |
| No | Ref | |||
| Any substance use | ||||
| Yes | 1.22 (1.04–1.44) | 0.013 | 1.37 (1.31–1.44) | <0.001 |
| No | Ref | |||
| Ever tested for HIV | ||||
| No | 1.80 (1.18–2.76) | 0.007 | ||
| Yes | ||||
| Last HIV test done | ||||
| ≤12 months | 0.41 (0.28–0.61) | <0.001 | 0.54 (0.36–0.80) | 0.002 |
| >12 months | 0.84 (0.57–1.26) | 0.403 | 0.97 (0.67–1.39) | 0.863 |
| Never tested for HIV | Ref | |||
| Ever had any genital symptoms | ||||
| Yes | 1.92 (1.21–3.04) | 0.006 | ||
| No | Ref | |||
| Medically circumcised | ||||
| No | 2.98 (2.25–3.94) | <0.001 | 2.05 (1.72–2.44) | <0.001 |
| Yes | Ref | |||
Variables not included in the adjusted model because of collinearity were employment status, ever tested for HIV, ever had sex, age at first sex, lifetime number of sex partners, current number of sex partners.
ref=Reference category; CI= Confidence Interval
Discussion
Despite the progressive scale-up of HIV prevention and treatment programs, the South African epidemic is characterised by persistently high HIV incidence rates [4, 7, 17]. It is possible that behavioral risk compensation sustain these high incidence rates and threaten the potential benefits of HIV prevention strategies [18–20]. As socializing venues appeal to and attract specific groups of individuals defined by varying characteristics or interests, we sought to provide insights into HIV risk among young heterosexual men recruited from these socializing venues, many of whom might not have been home at the time of the household survey [3].
Although HIV prevalence of 15.5% was lower than the prevalence reported for men from this region [3], our findings of the almost five fold higher HIV prevalence among men 25 years and older compared to younger men in the 15–24 year age group suggest that prevalence increases rapidly approximating a measure of incident HIV infection [7]. Whilst we hypothesized that socializing venues may predispose individuals to HIV risk, the study found no significant differences in HIV prevalence across these venues, though prevalence was higher amongst men from shebeens, transport hubs and spaza.
Whilst men from shebeens were younger, had not completed high school and were unemployed; overall, about 50% of men had not completed high school education raising concerns on the high rates of premature attrition from schools which potentially perpetuates cycles of poverty leading to risk taking behaviors. Whilst studies have shown the importance of education and that keeping young girls in schools has been protective against HIV infection among young South African women [21]; similarly our findings highlight the importance of ensuring that boys be encouraged to complete their high school education to reduce their risk of HIV infection. It is also possible that adolescent boys who leave school prematurely may have missed out on receiving appropriate information through the school’s life orientation program, and therefore have limited knowledge on HIV and HIV related risk behaviors. About a third of men experienced their first sex by age 16 years, yet only a third of these men reported condom use during this encounter which ranged from around 10% among men from community centers to around 38% among men in shebeens highlighting the lack of safer sex practices. More importantly, those reporting condom-use at first sex had higher odds of being HIV positive. It is well recognised that condom use at first sex is a strong predictor of future condom use, whether or not the partner is likely to be the person with whom the individual first had sex implying that young men who begin their sexual lives safely tend to remain safe [22] and are also less likely to report more sexual partners, are more likely to engage in subsequent protective behaviors, and experience fewer sexually transmitted infections than those individuals who do not use condoms at their sexual debut [23].
Whilst about two thirds of the men reported consuming alcohol, it was somewhat contradictory that just over a third of men from shebeens reported never having consumed alcohol. This contradiction might be due to the negative connotation associated with alcohol consumption and therefore likely to be a reporting bias. Additionally, the significant association of alcohol consumption and substance use with HIV positive status raises concerns of behavioral disinhibition increasing risk. According to the World Health Organization, South Africa has the highest reported alcohol consumption in Africa [24], linked to numerous health risks including lower levels of educational attainment [25] and behaviors that contribute to HIV acquisition [26, 27]. Whilst unemployment was not associated with HIV, the overall high rates of unemployment, in excess of 50% highlights the importance of comprehensive social, economic and structural interventions to improve health overall and minimize risk of HIV [28].
It was encouraging that at almost 90% of men had received an HIV test; and furthermore, those who received the test within the last 12 months had lower odds of being HIV positive compared to those who had tested more than 12 months ago. These findings demonstrate the protective benefits of frequent testing that could moderate or reduce sexual risk behavior even when results are negative, improve health seeking behaviors which have been associated with lower odds of HIV infection and potentially reduce HIV transmission [29, 30]. Even with the proven efficacy of medical male circumcision in preventing HIV acquisition in men, just over half of the men in the study were medically circumcised and the lower prevalence among these men compared to those not medically circumcised further confirms the protective benefits for the surgical removal of the foreskin. The modest coverage of medical circumcision has been successful through the current HIV prevention efforts [31], specifically the implementation of the universal test and treat programs geared towards managing HIV and to reduce HIV incidence [32], However, these findings further highlight the urgent need to scale-up medical male circumcision to benefit women as sexual partners [17, 33, 34] and more importantly to sustain the declines in HIV prevalence in the longer term for public health benefit [35].
Our study has several strengths and limitations. Although we recruited a large sample from socializing venues, it is possible that men who might have known their HIV positive status may not have participated and therefore the study sample may not be representative of all young heterosexual men from rural communities in KwaZulu-Natal, South Africa. Nevertheless, it is important to reach key populations to understand HIV burden and related risk factors to interrupt the cycle of HIV transmission. Given the cross-sectional design of our study, our findings on incomplete high school education and alcohol use limits the ability to conclude temporal association of these factors and HIV. It is possible that participant responses may be subject to recall bias and social desirability bias as high levels of stigma and discrimination in the community may have influenced individuals’ decisions to reliably disclose HIV testing patterns, HIV status and sexual behavioral information. However, the HIV status of participants was not based on self-report but rather on laboratory measurements with assays having high sensitivity and specificity. Whilst we have addressed selected demographic and behavioral factors, a limitation of our study was the absence of data on social, structural and political factors which are equally important to address at the individual and community levels as these could fundamentally shape and influence patterns of HIV risk behaviors [36–38].
Conclusions
In conclusion, not all men share the same risk and risk differs across age groups, therefore a nuanced understanding of factors that provide a life course perspective of HIV risk is important. The higher odds of HIV associated with being 25 years and older, not having completed high school, consuming alcohol and other substance use, and not being medically circumcised highlights the need for intensified efforts to reach younger men to minimize future risk taking behaviors that contribute to HIV infection. Thus, comprehensive novel interventions are critical to engage and encourage heterosexual men to complete high school, encourage frequent HIV testing and provide information and education on the risks associated with alcohol consumption. To improve delivery, greater effort is needed to implement innovative programs within settings that are easily accessible and where heterosexual men are likely to be.
Supplementary Material
Acknowledgements
The authors sincerely thank all the participants for their time and commitment to study participation. To all the traditional and municipal leadership for their support with the study. The Provincial Department of Health and uMgungundlovu District office, the local primary health care clinics for management of follow-up care and support for participants referrals. To all the study staff, laboratory and primary health care clinic staff for study related procedures and follow-up care. Our sincere thanks to Mr Mesuli Mhlongo for his invaluable assistance.
Conflicts of Interest and Source of Funding:
The authors declare that they have no conflicts of interest to disclose. The study was made possible through The Eunice Kennedy Shriver National Institute Of Child Health & Human Development (NICHD), Office of the Director and National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health: R01HD083343 (Multi-PI: Kharsany and Kohler). Further funding support was provided to N. Ntombela from the University Capacity Development Programme from the College of Health Sciences by the University of KwaZulu-Natal. REF:207525096. The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the funding agencies. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the University of KwaZulu-Natal.
Abbreviations
- aOR
adjusted odds ratio
- BREC
University of KwaZulu-Natal Biomedical Research Ethics Committee
- CI
confidence interval
- GEE
Generalized estimating equations
- GPS
Global positioning system
- HIV
Human immunodeficiency virus
- IQR
Interquartile range
- PDA
personal digital assistant
- PID
participant identification number
- QIC
Quasi-likelihood information under the independence model criterion
Footnotes
Data sharing statement
Available upon request to the corresponding author
Electronic supplementary material
Supplementary file l.doc is available for this article.
References
- 1.Joint United Nations Programme on HIV/AIDS (UNAIDS): South Africa: UNAIDS. 2018. (https://www.unaids.org/en/regionscountries/countries/southafrica): Date accessed 10 October 2019. [Google Scholar]
- 2.Shisana O, Rehle T, Simbayi LC, Zuma K, Jooste S, N. Z, Labadarios D, Onoya D: South African National HIV Prevalence, Incidence and Behaviour Survey, 2012. 2014. (http://www.hsrc.ac.za/en/research-outputs/view/6871): Date accessed 17 June 2020. [DOI] [PubMed]
- 3.Kharsany AB, Cawood C, Khanyile D, Lewis L, Grobler A, Puren A, Govender K, George G, Beckett S, Samsunder N: Community-based HIV prevalence in KwaZulu-Natal, South Africa: results of a cross-sectional household survey. The Lancet HIV 2018, 5(8):e427–e437. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Simbayi L, Zuma K, Zungu N, Moyo S, Marinda E, Jooste S, Mabaso M, Ramlagan S, North A, van Zyl J et al. : South African National HIV Prevalence, Incidence, Behaviour And Communication Survey, 2017 HSRC Press, Cape Town: 2019, Available at: https://www.hsrcpress.ac.za/books/south-african-national-hiv-prevalence-incidence-behaviour-and-communication-survey-2017: Accessed 15 May 2019. [Google Scholar]
- 5.de Oliveira T, Kharsany AB, Graf T, Cawood C, Khanyile D, Grobler A, Puren A, Madurai S, Baxter C, Karim QA et al. : Transmission networks and risk of HIV infection in KwaZulu-Natal, South Africa: a community-wide phylogenetic study. Lancet HIV 2017, 4(1):e41–e50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Akullian A, Bershteyn A, Klein D, Vandormael A, Bärnighausen T, Tanser F: Sexual partnership age pairings and risk of HIV acquisition in rural South Africa. AIDS 2017, 31(12):1755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kharsany ABM, Cawood C, Lewis L, Yende-Zuma N, Khanyile D, Puren A, Madurai S, Baxter C, George G, Govender K et al. : Trends in HIV Prevention, Treatment, and Incidence in a Hyperendemic Area of KwaZulu-Natal, South Africa. JAMA Netw Open 2019, 2(11):e1914378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Venkatesh KK, Madiba P, De Bruyn G, Lurie MN, Coates TJ, Gray GE: Who gets tested for HIV in a South African urban township? Implications for test and treat and gender-based prevention interventions. J Acquir Immune Defic Syndr 2011, 56(2):151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Beauclair R, Hens N, Delva W: Concurrent partnerships in Cape Town, South Africa: race and sex differences in prevalence and duration of overlap. J Int AIDS Soc 2015, 18:19372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Joint United Nations Programme on HIV/AIDS (UNAIDS). Working with men for HIV prevention and care. 2001. (http://data.unaids.org/publications/irc-pub02/jc543-workingwithmen_en.pdf): Date accessed 12 November 2019.
- 11.Sisonke Gender Justice Network: Building male involvement in SRHR: A basic model for male involvement in Sexual and Reproductive Health and Rights. 2012(http://menengage.org/wp-content/uploads/2014/01/Sonke-Gender-Justice-Model-for-Male-Involvement-in-SRHR.pdf): Date accessed 1 July 2020.
- 12.Chikovore J, Gillespie N, McGrath N, Orne-Gliemann J, Zuma T, Group ATS: Men, masculinity, and engagement with treatment as prevention in KwaZulu-Natal, South Africa. AIDS Care 2016, 28 Suppl 3:74–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Joint United Nations Programme on HIV/AIDS (UNAIDS): Male engagement in the HIV response —a Platform for Action. 2016. (https://www.unaids.org/sites/default/files/media_asset/JC2848_en.pdf): Date accessed 17 October 2019.
- 14.Stern E, Rau A, Cooper D: Sexual and reproductive health perceptions and practices as revealed in the sexual history narratives of South African men living in a time of HIV/AIDS. SAHARA-J: 2014, 11(1):233–244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Young LE, Michaels S, Jonas A, Khanna AS, Skaathun B, Morgan E, Schneider JA, uConnect Study T: Sex Behaviors as Social Cues Motivating Social Venue Patronage Among Young Black Men Who Have Sex with Men. AIDS Behav 2017, 21(10):2924–2934. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Velloza J, Watt MH, Abler L, Skinner D, Kalichman SC, Dennis AC, Sikkema KJ: HIV-Risk Behaviors and Social Support Among Men and Women Attending Alcohol-Serving Venues in South Africa: Implications for HIV Prevention. AIDS Behav 2017, 21(Suppl 2):144–154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Abdool Karim Q, Kharsany AB, Frohlich JA, Werner L, Mashego M, Mlotshwa M, Madlala BT, Ntombela F, Abdool Karim SS: Stabilizing HIV prevalence masks high HIV incidence rates amongst rural and urban women in KwaZulu-Natal, South Africa. Int J Epidemiol 2010, 40(4):922–930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Eaton LA, Kalichman S: Risk compensation in HIV prevention: implications for vaccines, microbicides, and other biomedical HIV prevention technologies. Curr HIV/AIDS Rep 2007, 4(4):165–172. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Kalichman SC, Cherry C, Amaral CM, Swetzes C, Eaton L, Macy R, Grebler T, Kalichman MO: Adherence to antiretroviral therapy and HIV transmission risks: implications for test-and-treat approaches to HIV prevention. AIDS Patient Care STDS 2010, 24(5):271–277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kalichman S, Mathews C, Kalichman M, Eaton LA, Nkoko K: Male circumcision for HIV prevention: Awareness, risk compensation, and risk perceptions among South African women. Glob Public Health 2018, 13(11):1682–1690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pettifor AE, Levandowski BA, Macphail C, Padian NS, Cohen MS, Rees HV: Keep them in school: the importance of education as a protective factor against HIV infection among young South African women. Int J Epidemiol 2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Hendriksen ES, Pettifor A, Lee S-J, Coates TJ, Rees HV: Predictors of condom use among young adults in South Africa: the Reproductive Health and HIV Research Unit National Youth Survey. Am J Public Health 2007, 97(7):1241–1248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Shafii T, Stovel K, Holmes K: Association between condom use at sexual debut and subsequent sexual trajectories: a longitudinal study using biomarkers. Am J Public Health 2007, 97(6):1090–1095. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.World Health Organization (WHO): Global status report on alcohol and health 2018. 2019. (https://www.who.int/substance_abuse/publications/global_alcohol_report/en/): Date accessed 17 June 2020.
- 25.Peltzer K, Davids A, Njuho P: Alcohol use and problem drinking in South Africa: findings from a national population-based survey. Afr J Psychiatry 2011, 14(1). [DOI] [PubMed] [Google Scholar]
- 26.Fritz K, Morojele N, Kalichman S: Alcohol: the forgotten drug in HIV/AIDS. The Lancet 2010, 376(9739):398–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Panday S, Makiwane M, Ranchod C, Letsoalo T: Teenage pregnancy in South Africa - with a specific focus on school-going learners. 2009. (http://www.hsrc.ac.za/en/research-data/view/4542): Date accessed 17 June 2020.
- 28.Setlalentoa B, Pisa P, Thekisho G, Ryke E, Loots Du T: The social aspects of alcohol misuse/abuse in South Africa. South Afr J Clin Nutr 2010, 23(sup2):11–15. [Google Scholar]
- 29.Liu Y, Qian HZ, Ruan Y, Wu P, Osborn CY, Jia Y, Yin L, Lu H, He X, Shao Y et al. : Frequent HIV Testing: Impact on HIV Risk Among Chinese Men Who Have Sex with Men. J Acquir Immune Defic Syndr 2016, 72(4):452–461. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Scott-Sheldon LA, Carey MP, Carey KB, Cain D, Simbayi LC, Mehlomakhulu V, Kalichman SC: HIV testing is associated with increased knowledge and reductions in sexual risk behaviours among men in Cape Town, South Africa. Afr J AIDS Res 2013, 12(4):195–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Montague C, Ngcobo N, Mahlase G, Frohlich J, Pillay C, Yende-Zuma N, Humphries H, Dellar R, Naidoo K, Karim QA: Implementation of adolescent-friendly voluntary medical male circumcision using a school based recruitment program in rural KwaZulu-Natal, South Africa. PLoS ONE 2014, 9(5):e96468. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.South African National Department of Health.: Implementation of the Universal Test and Treat Strategy for HIV Positive patients and differentiated care for stable patients 2016, Available at: http://www.sahivsoc.org/Files/22%208%2016%20Circular%20UTT%20%20%20Decongestion%20CCMT%20Directorate%20(2).pdf: Accessed 24 February 2019.
- 33.Kharsany AB, Frohlich JA, Yende-Zuma N, Mahlase G, Samsunder N, Dellar RC, Zuma-Mkhonza M, Karim SSA, Karim QA: Trends in HIV prevalence in pregnant women in rural South Africa. J Acquir Immune Defic Syndr 2015, 70(3):289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hankins C: Male circumcision: implications for women as sexual partners and parents. Reprod Health Matters 2007, 15(29):62–67. [DOI] [PubMed] [Google Scholar]
- 35.Nagelkerke NJ, Moses S, de Vlas SJ, Bailey RC: Modelling the public health impact of male circumcision for HIV prevention in high prevalence areas in Africa. BMC Infect Dis 2007, 7(1):16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Seeley J, Watts CH, Kippax S, Russell S, Heise L, Whiteside A: Addressing the structural drivers of HIV: a luxury or necessity for programmes? J Int AIDS Soc 2012, 15 Suppl 1:1–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Cluver LD, Orkin FM, Meinck F, Boyes ME, Sherr L: Structural drivers and social protection: mechanisms of HIV risk and HIV prevention for South African adolescents. J Int AIDS Soc 2016, 19(1):20646. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Auerbach JD, Parkhurst JO, Caceres CF: Addressing social drivers of HIV/AIDS for the long-term response: conceptual and methodological considerations. Glob Public Health 2011, 6 Suppl 3:S293–309. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.