The levels and factors associated with inconsistent condom use among young males with older, same-sex partners in South Africa

Abstract

Condom use among females with older, male sexual partners is well documented in South Africa. To a far lesser extent, condom use among males with older same-sex partners is known. However, the sexual health of young males is of pivotal importance to social and economic development. The purpose of this study is to determine the association between age of older same-sex partner and the likelihood of condom use among young males in South Africa. In this study, data from Fourth South African National HIV, Behaviour and Health Survey 2012 are used. A sample of 46,790 (N) young (15–24 years old) males with older same-sex partners is analysed to determine the level and sociodemographic factors associated with inconsistent condom use. Descriptive statistics and a binary logistic regression model are used. About 55.51% of you th with older (1–5 years) partners use condoms inconsistently. Further, only 36.08% of young males residing in urban areas with older (>1 year) partners use condoms every time they have sexual intercourse. Also, young males whose partners are 6–10 years (odds ratio: 0.47) and 11 or more years (odds ratio: 0.58) are less likely to use condoms consistently. Finally, the type of relationship with most recent same-sex partner shows that respondents who are in casual relationships (odds ratio: 0.08) are least likely to use condoms. Young males with older, same-sex partners do not use condoms consistently and are therefore at risk of contracting HIV and other STIs. However, HIV-positive youth and those with casual sexual partners are more likely to use condoms, proving that some positive and protective sexual behaviours are being exercised. With these findings in mind, further investigation into why some protective behaviours are possible while others are not needed. In particular, the role of older-sexual partners needs to be investigated.

Keywords:

• HIV/AIDS
• older same-sex relationships
• inconsistent condom use
• regression
• South Africa
• young males

PUBLIC INTEREST STATEMENT

Youth are vulnerable to HIV and other STIs because generally, they do not use condoms consistently. Many studies before have proven this to be true among youth in heterosexual relationships and for youth with older-sexual partners. Our study attempts to investigate if inconsistent condom use is also true for young males in same-sex relationships with older partners. We found that young men are not likely to use condoms consistently with older, same-sex partners but HIV-positive young men are more likely to use condoms consistently. The results of our study show the need for more investigation into same-sex relationships with older partners.

1. Introduction

Not much is known about the condom use practices among young males in age-disparate relationships with same-sex partners in South Africa. Consistent condom use among sexually active youth (15–24 years old) in South Africa is low at 57.9% (Ntshiqa et al., 2018). One study of 741 sexually active adolescents (14–19 years old) from the Soweto township in South Africa found 54.6% of males and 54.2% of females do not use condoms (Closson et al., 2018). This appears higher than percentages reported in other countries, including Burkina Faso, Ghana, Uganda, and Malawi, where 34–57% of sexually active youth (aged 14–19 years) reported using condoms but is lower than countries such as Greece where 89% of youth reportedly use condoms (Koyama, Corliss, & Santelli, 2009). A study of youth (n = 4,853) who perceive their risk of contracting HIV as low, in Cape Town found that only 61% of males and 67% of females were using condoms consistently (Muchiri, Odimegwu, & De Wet, 2017). While it has also been identified that young females (16–24 years old) who are students (odds ratio: 1.65, CI: 1.2–2.3) and who have exposure to HIV communication programmes (odds ratio: 3.1, CI: 1.2–8.6) are more likely to use condoms (Ntshiqa et al., 2018). Research done in the USA also shows that youth who have knowledge of HIV and AIDS and are educated are more likely to use condoms (Gleton, Jahanfar, Inungu, & Latty, 2019; McNeill, George, & Glover, 2017). Other studies on youth globally and in South Africa have identified gender roles and norms, lower level of education and low socioeconomic status to be associated with lack of condom use (Bui et al., 2012; Eggers, Aarø, Bos, Mathews, & de Vries, 2014; Fladseth, Gafos, Newell, & McGrath, 2015; Pinchoff, Boyer, Mutombo, Chowdhuri, & Ngo, 2017). And despite national programmes in the country such as “loveLife” and the Youth-Friendly Services promoting the consistent use of condoms to prevent HIV transmission, recent statistics show an incidence rate of 0.91% of new infections among adults (15+ years old) in 2017 with a youth prevalence rate of 4.64% for the same year (Beksinska, Pillay, Milford, & Smit, 2014; Statistics SA, 2018). From these statistics, however, the percentage distribution for same-sex versus heterosexual couples cannot be determined.

A factor associated with inconsistent condom use is age-disparate relationships (Maughan-Brown, Evans, & George, 2016; Toska, Cluver, Boyes, Pantelic, & Kuo, 2015). Globally, the levels of intergenerational or age-disparate relationships range between 13% and 14% (Harling et al., 2014). Studies have identified that younger partners are often unable to negotiate condom use in these relationships for fear of rejection and violence (Caldwell & Mathews, 2016; Rodger et al., 2016). A qualitative study done in the US found that younger partners experience more intimate partner violence in age-disparate relationships, when they request older partners to use condoms (Raj et al., 2007). While another study from India found that young brides are subject to violence from older grooms and are not able to negotiate condom use either (Santhya & Jejeebhoy, 2007). The consequences of this include the spread of sexually transmitted infections (STIs) including HIV and also unintended pregnancies (Ritchwood et al., 2016; Shisana et al., 2014; Zuma et al., 2016). In South Africa, research has shown that there is a higher likelihood that youth will become HIV positive (odds ratio: 1.53; CI: 0.92–2.54) if their sexual partner is five or more years older (Zuma et al., 2016). Additional research has revealed that poverty, unemployment and consumerism are contributing factors in age-disparate relationships (Kamndaya, Vearey, Thomas, Kabiru, & Kazembe, 2016; Schaefer et al., 2017). However, most of the published research focus on heterosexual couples, more specifically females with older male sexual partners and neglects young males who have older same-sex partners. Yet this is an important area of study because young males share similar environments and challenges including poverty, unemployment, and under-education as females in South Africa and would, therefore, have similar rationales in terms of sexual partner selection (Graham & De Lennoy, 2016; Whitley, Gould, Wright, & Hayden, 2017). For this reason, the purpose of this study is to (1) estimate the level of condom use among young males who engage in age-disparate sexual relationships with same-sex partners and (2) to identify the factors associated with condom use within these relationships.

2. Methods

The study is a cross-sectional analysis of secondary data from the Fourth South African National HIV, Behaviour and Health Survey 2012. The survey is the fourth in the series of national HIV household surveys conducted by a consortium of scientists led by the Human Sciences Research Council (HSRC) (Shisana et al., 2014). The sampling procedure and full methodology carried out by the HSRC researchers and collaborators can be found in the South African National HIV Prevalence, Incidence and Behaviour Survey, 2012 report (Shisana et al., 2014). A subsequent survey was conducted in 2017 but has not yet been made available for analysis by the HSRC. The current dataset was obtained through a written request by the authors to the HSRC for consent to analyse the data for research purposes. The data that were then made available for download were completely anonymised with no identifying information about the participants included. The data used for this study are not the most recent in the series, but given that behavioural practices do not dramatically change over short periods of time, the study is still useful to better understand sexual behaviours among youth.

The nationally representative data collected 38,438 (52,387,332 weighted) respondent’s HIV status, demographic, socioeconomic and behavioural characteristics (Shisana et al., 2014). However, analysis in this study has been restricted to male youth between 15 and 24 years old with completed information on the age and sex of their most recent sexual partner. A weighted sample of 46,790 (1,747 unweighted) young males with older (>1 year), same-sex sexual partners is included in the study.

The main outcome variable of this study is “consistent condom use”. This variable is derived from the survey question: how often do you use a condom with your most recent sexual partner? With response categories of (1) every time; (2) almost every time; (3) sometimes; (4) never. Consistent condom use is considered as response categories (1) and (2). For the adjusted logistic regression model, a binary variable of consistent (1 and 2) and inconsistent (3 and 4) is used.

Age difference between the respondent and most recent same-sex partner has been identified through the survey questions: age of respondent (in single years) and what is the approximate age of the most recent partner with whom you have had sex? Through subtracting the age of their partner from the age of the respondent, the age difference was determined. Respondents with same-age or younger same-sex partners were dropped from the analysis (N = 1,179,754) as these would not be age-disparate relationships and therefore would not necessarily have the same power dynamics with regard to condom use. Partner age difference was categorised into “1–5 years older”; “6–10 years older” and “11+ years older”. The reason two five-year intervals were selected is based on literature which shows that partner age difference is often 1–5 years, 6–10 years and 11+ years older (Harling et al., 2014).

Demographic and socioeconomic control variables included in the study are age of the respondent (15–19 or 20–24); population group (African, White, Coloured, Indian/Asian); place of residence (urban or rural); employment status (employed, unemployed or student) and type of relationship (casual, husband, live-in partner, boyfriend, not living with). The population group classifications, despite being developed by the Apartheid government to strengthen their separatist policies, are still used today by citizens and enumerators for various purposes. Of the groups, the most contentious naming is of the Coloured population which refers to persons of mixed race and/or Khoisan heritage (Quintana-Murci et al., 2010). This study uses the four dominant racial groups that were based on the South African Census and posed to participants of this survey (Statistics SA, 2011).

Three sexual and reproductive health control variables have also been included and these survey questions are: Can AIDS be cured? (yes, no, don’t know responses); have you been circumcised? (yes or no); what is your HIV status? (positive or negative). Recent studies in less developed countries identified these variables as having a relationship to condom use (Fladseth et al., 2015; George et al., 2019; Leddy, Chakravarty, Dladla, de Bruyn, & Darbes, 2016). For this reason, these variables were adopted for the current study.

Cross-tabulations and a forward stepwise logistic regression model are used to determine the level and assess the relationship between age difference and inconsistent condom use among young males with older, same-sex partners in South Africa. For all analyses, statistical significance was set at a p-value of less than 0.05.

3. Results

Among 15–19-year olds, 75.93% had a sexual partner 11 or more years older (Table 1). Compared to 20–24-year olds where the majority (62.84%) had a same-sex partner who is 6–10 years older. By population group, 37.23% of African youth had a same-sex partner who is 6–10 years older and 98.13% of White youth had a partner 1–5 years older. More than half of Coloured youth (51.8%) also have a same-sex partner who is between 1 and 5 years older while 68.61% of Indian/Asian young males have a sexual partner who is 11 or more years older. With regards to the place of residence, the majority (68.6%) of youth living in urban areas had a partner 1–5 years older. Among youth in rural areas, 39.67% had a partner 6–10 years older while 49.02% had a partner 11 and more years older. For employment status, more than half (52.28%) of employed youth had a partner 1–5 years older. The percentage of unemployed youth who had a partner 11 and more years older was 32.76% and 18.25% among employed youth. The sexual and reproductive health characteristics of youth showed that 70.81% of youth who said AIDS can be cured had a partner 1–5 years older while 29.19% had a partner 6–10 years older. More than half (59.34%) of youth who have not undergone circumcision had a partner 1–5 years older while 40.23% had a partner 6–10 years older. With regards to HIV status, 35.82% of youth who reported being HIV negative had a partner 1–5 years older, 33.17% had a partner 6–10 years older and 31.01% had a partner 11 and more years older. Among those that reported being HIV positive, more than half (60.01%) had a partner 1–5 years older while 39.99% had a partner 11 and more years older.

Table 1. Percentage distribution of the characteristics of young males by age difference of most recent same-sex partner

Characteristics		Male Partner
	1–5 yrs (N = 8,859)		6–10 yrs (N = 25,497)		11+ yrs (N = 12,012)
							Demographic and socioeconomic	N	%	N	%	N	%
Age group*
15–19	915,304	19.74	200,773	4.33	3,520,722	75.93
20–24	907,422	19.57	2,913,765	62.84	815,149	17.58
Population group*
African	1,669,248	36	1,726,281	37.23	1,241,271	26.77
White	4,550,092	98.13	37,094	0.8	50,077	1.08
Coloured	2,401,862	51.8	2,234,938	48.2	0**	0
Indian/Asian	1,455,492	31.39	0**	0	3,181,308	68.61
Place of residence*
Urban	3,183,627	68.66	1,376,202	29.68	76,971	1.66
Rural	524,886	11.32	1,839,419	39.67	2,272,959	49.02
Employment status*
Employed	2,424,119	52.28	693,665	14.96	1,519,016	32.76
Unemployed	1,862,139	40.16	1,928,445	41.59	846,216	18.25
Student	253,169	5.46	4,383,631	94.54	0**	0
Sexual and reproductive health
AIDS can be cured*
Yes	3,283,318	70.81	1,353,482	29.19	0**	0
No	1,791,660	38.64	1,487,949	32.09	1,357,191	29.27
Don’t know	982,538	21.19	3,117,784	67.24	536,478	11.57
Circumcision status*
Yes	4,073,892	87.86	562,908	12.14	0**	0
No	2,751,477	59.34	1,865,385	40.23	19,938	0.43
HIV status*
Negative	1,660,902	35.82	1,538,027	33.17	1,437,872	31.01
Positive	2,782,544	60.01	0**	0	1,854,256	39.99

*p-value <0.05; **no responses from participants.

Figure 1 shows that among young men who were in casual relationships, all (100%) had a partner 1–5 years older. Among those with husbands, more than 90% of the husbands are 11+ years older and among those with a live-in partner about 50% of the partners were 6–10 years old and just less than half were 11+ years old. For those with boyfriends who are not cohabiting, over 60% of the partners were 6–10 years older than the respondents.

Figure 1. Percentage distribution of type of relationship by partner age difference among young males with older, same-sex partners in South Africa

Compared to 20–24-year olds (39.98%), 92.38% of 15–19-year olds said they never or sometimes (inconsistent) use a condom (Table 2). By race, more than half (55.31%) of Africans said they use a condom consistently while the percentage of those that use a condom inconsistently was 44.69%. The majority (87.21%) of the Coloured population group use condoms inconsistently, and 12.79% reported using a condom consistently. With regards to the place of residence, 63.33% of youth living in urban areas said they never or sometimes (inconsistent) use a condom compared to 41.15% among youth in rural areas. The frequency of consistent condom use was 58.85% among adults in rural areas compared to 36.67% in urban areas. By employment status, 85.72% of employed adults and 94.48% of unemployed adults, respectively, report inconsistent condom use. The majority (96.16%) of those that reported using a condom consistently were students compared to 14.28% among youth who are employed and 5.52% among unemployed youth. All youth (100%) that said AIDS can be cured use a condom consistently. More than half (53.03%) of youth who said AIDS cannot be cured use a condom consistently while 46.97% use a condom inconsistently. Among circumcised youth, 100% use a condom consistently. The majority (95.83%) of youth who never use a condom were those that were not circumcised. With regards to HIV status, more than half (54.97%) of HIV-negative youth use a condom consistently compared to 42.30% among HIV-positive youth. The percentage of HIV-positive youth who use condoms inconsistently was 57.70% compared to 45.03% among HIV-negative youth.

Table 2. Percentage distribution of the characteristics of the respondents with older same-sex partners (>1 year) by regularity of condom use

Characteristics		Frequency of condom use
	Consistent (N = 24,664)		Inconsistent (N = 22,127)
Demographic	N	%	N	%
Age group*
15–19	341,307	7.62	4,137,793	92.38
20–24	2,688,356	60.02	1,790,744	39.98
Race*
African	2,477,390	55.31	2,001,710	44.69
White	4,479,100	100	0**	0
Coloured	572,877	12.79	3,906,223	87.21
Indian/Asian	383,859	8.57	4,095,241	91.43
Place of residence*
Urban	1,642,486	36.67	2,836,614	63.33
Rural	2,635,950	58.85	1,843,150	41.15
Employment status*
Employed	639,615	14.28	3,839,485	85.72
Unemployed	247,246	5.52	4,231,854	94.48
Student	4,307,103	96.16	171,997	3.84
Sexual and reproductive health
AIDS can be cured*
Yes	0**	0	4,479,100	100
No	2,375,267	53.03	2,103,833	46.97
Circumcision status*
Yes	4,479,100	100	0**	0
No	186,778	4.17	4,292,322	95.83
HIV status*
Negative	2,462,161	54.97	2,016,939	45.03
Positive	1,894,659	42.3	2,584,441	57.7
Partner Age difference*
1–5 years	1,992,752	44.49	2,486,348	55.51
6–10 years	3,542,072	79.08	937,028	20.92
11+ years	321,151	7.17	4,157,949	92.83

*denotes p-value <0.05; **no responses from participants.

More than two-thirds (79.08%) of youth with a partner 6–10 years older reported that they use a condom consistently. Among youth with a partner 1–5 years older, 44.49%, and those with a partner 11 or more years older reported 7.17% consistent condom use. Among youth with a partner 1–5 years older, 55.51% said they use a condom inconsistently. Finally, 20.92% of youth with a partner 6–10 years reported that they use condoms inconsistently compared to 92.83% of youth with a partner 11 or more years older.

Table 3 shows the results of the adjusted logistic regression model showing odds ratios. Young men with same-sex partners who are 6–10 years older (OR = 0.47) and 11 or more years older (OR = 0.58) are less likely to use a condom consistently. Youth aged 20–24 years were 1.57 times more likely to use a condom compared to those aged 15–19 years [CI: 1.53–1.60]. Compared to Africans, youth of all other races were more likely to use a condom. White people were 7.09 times more likely to use a condom [CI: 6.83–7.35], coloured people 2.14 times more likely [CI: 2.07–2.20] and Indian/Asians 1.52 times more likely to use a condom [CI: 1.42–1.63]. By employment status, unemployed youth were 1.61 [CI: 1.57–1.65] and students 1.24 [CI: 1.24–1.27] times more likely to use a condom compared to employed youth. Youth who were living with their sex partner were 0.26 times less likely to use a condom compared to youth who reported having a husband [CI: 0.23–0.28].

Table 3. Odds of irregular condom use by respondent characteristics (adjusted odds ratios)

Characteristics
Partner Age difference	Odds Ratio	95% Conf.	Interval
1–5 years	RC
6–10 years	0.47	0.4589	0.6120
11 + years	0.58	0.5535	0.6982
Demographic and socioeconomic
Age group
15–19	RC
20–24	1.57	1.5389	1.6072
Race
African	RC
White	7.09	6.8351	7.3542
Coloured	2.14	2.0748	2.2022
Indian/Asian	1.52	1.4235	1.6325
Place of residence
Urban	RC
Rural	1.48	1.4502	1.5073
Employment status
Employed	RC
Unemployed	1.61	1.5713	1.6502
Student	1.24	1.2122	1.2781
Relationship with most recent same-sex partner
Husband	RC
Live-in partner	0.26	0.2352	0.2808
Boyfriend not living with	0.15	0.1357	0.1592
Casual partner	0.08	0.0722	0.0853
Sexual and Reproductive Health
AIDS can be cured
Yes	RC
No	2.52	2.4449	2.5918
Circumcision status
Yes	RC
No	2.67	2.6205	2.7248
HIV status
Negative	RC
Positive	7.11	6.8054	7.4202

RC = Reference Category.

Youth, whose sex partner was a boyfriend they were not living with, were also less likely (OR: 0.15, CI: 0.1357-0.1592) to use a condom compared to those who were married with a husband [CI: 0.13–0.15]. Youth who had a casual sex partner were 0.08 times less likely to use a condom compared to youth who had a husband [CI: 0.07–0.08]. Compared to youth who believe that AIDS can be cured, those that thought AIDS has no cure were 2.52 times more likely to use a condom [CI: 2.44–2.59]. Circumcised youth were 2.67 times more likely to use a condom compared to youth who were not circumcised [CI: 2.62–2.72]. Youth who were living with HIV were 7.11 times more likely to use a condom compared to HIV-negative youth [CI: 6.80–7.42].

4. Discussion

The aim of this study was to identify the factors associated with inconsistent condom use among young males with older, same-sex partners in South Africa. Inconsistent condom use is a risk factor for the transmission of HIV and other STIs (Madiba & Ngwenya, 2017). Youth, compared to older age groups, are more vulnerable to infection because their use of condoms is more inconsistent (Protogerou, Johnson & Hagger, 2017; Protogerou, Johnson, & Hagger, 2018). This study addresses two apparent gaps in existing literature. First, there is a scarcity of nationally representative research on condom use in same-sex partnerships in South Africa. Second, while research on condom use in age-disparate relationships has been done in the country, these studies have predominantly focused on the plight of younger females with older male partners.

It is not likely that young males will use condoms with older same-sex partners. This result is similar to a number of studies on youth in age-disparate relationships where power dynamics favour the older partner (Newcomb, Ryan, Garofalo, & Mustanski, 2014; Pascoe et al., 2015). In one study, it was found that young females were at a higher risk of HIV infection with partners six or more years older and that the use of condoms was low (Pascoe et al., 2015). A second study on young males in age-disparate relationships with male partners in the US showed that having an older partner was positively associated with total no condom use which is for both insertive and penetrative sex (Newcomb et al., 2014). These studies, among others, suggest that younger partners in sexual relationships, regardless of the sex of the partner, are less likely to practice safe sex. Among the reasons offered to explain this phenomenon is the financial dependence of younger persons on older partners, particularly if the younger person is an adolescent (15–19 years old) (Boyer et al., 2017; Mabaso et al., 2018; Toska, Pantelic, Cluver, Meinck, & Keck, 2015). Other reasons include power dynamics whereby older persons purposefully select youth for their inability to question the authority of their older partners (Antfolk et al., 2015; Longfield, Glick, Waithaka, & Berman, 2004).

Youth reportedly using a condom every time is high but so is the reporting of never using a condom. This polarised approach to condom use is problematic because condom use should not be an “always or never” situation. Sexually active youth should, ideally, be using condoms consistently which is at all or almost all sexual encounters. However, even those who use condoms “sometimes” are more protected from HIV and other STIs compared to those who “never” use condoms. A possible reason for this dichotomy could be related to whether or not the partners are in a monogamous, stable relationship. The research which supports this possibility found that youth who report having casual relationships are more likely to use condoms than those who are in “steady”, cohabiting or married relationships (Bauman & Berman, 2005; Rutakumwa, Mbonye, Kiwanuka, Bagiire, & Seeley, 2015). The use of condoms in these contexts, therefore, is related to how long and well partners know each other and the level of trust within these relationships.

Moreover, this study also found some of the protective or positive sexual behaviours practiced by youth with older, same-sex partners. Young males in the study who are HIV positive are more likely to use condoms consistently. This shows that HIV-positive youth are not careless with regard to spreading the disease to their sexual partners. This is also similar to the results of a study done in the Eastern Cape province of South Africa, where it was found that HIV-positive youth who were able to disclose their status to partners were more likely to practice safe-sex (Toska, Cluver, Hodes, & Kidia, 2015). In addition, youth who believe there is no cure for AIDS are also more likely to use condoms consistently. That is, youth who have knowledge of the disease are practicing protective behaviours. Research from the US has also found that youth with accurate knowledge of HIV and AIDS do attempt to practice safe-sex (Dariotis & Johnson, 2015). However, there is also literature which contends that knowledge does not directly translate into safe behavioural practices (Davis, Campbell, Hildon, Hobbs, & Michie, 2015; Kelly & Barker, 2016; Sniehotta, Presseau, & Araújo-Soares, 2014). Results from this study show that an association between AIDS knowledge and protective behaviour does exist; however, causality could not be determined and confounders were not identified.

5. Limitations, recommendations and strengths

The current study is subject to a few limitations. First, we could not determine the accessibility of condoms for youth who never used condoms. A study which addresses the availability of condom use is recommended to better explain a possible reason for inconsistent condom use. Second, the study data are not the most recent. A similar study using the subsequent dataset in the series is suggested, when these data are made available, to examine if there has been any change to condom use behaviours more recently. Third, due to the cross-sectional nature of the data, confounding between the HIV knowledge questions in the survey could not be controlled for. A longitudinal study of a cohort of young males with older, same-sex partners would be ideal to determine this and other confounders, and to determine causality. Also, this study did not address the older, same-sex partners. Future research should include a qualitative study with the older-sexual partners to investigate their role or contribution to the inconsistent condom use. That is, research which pinpoints the locus of sexual power is needed since it could not be determined in this study. Also, more research on the characteristics of older same-sex partners and of the couples’ characteristics is needed to include older partners in youth programmes. Since age-disparate partnerships are common, making the couple, that is, both partners, the target of intervention would prove more useful in reducing unsafe sexual practices than the inclusion of only the younger partner. Finally, while same-sex unions are legal in South Africa, these relationships are not entirely socially acceptable, as a result, there is a possibility that the respondents are subject to social desirability bias in their responses.

There are also several strengths to the study. First, young males with same-sex partners are an under-researched population in developing countries. In African countries, in particular, where same-sex partnerships remain highly discriminated against, this population remains highly vulnerable to various sexual and reproductive health concerns including inability to access health care and information as well as the risk of contracting HIV and other STIs. This study, therefore, assists in addressing the sexual health needs of young males in same-sex partnerships in South Africa. Second, the study is nationally representative with a large study sample which enabled the use of inferential statistical methods to better understand the condom use of young males in the country. This is in comparison to the few qualitative studies that were found which use fewer males and are not representative of the country (Lesch, Brits, & Naidoo, 2017; Scott, 2018).

6. Conclusion

Condom use among young males with older, same-sex partners is not consistent. This is important because the sexual health of young males (and females) will determine the levels of HIV incidence in the population and is an indicator of expected adult health outcomes. The distorted power dynamics of having an older-sexual partner explains why safe-sex negotiations could be difficult for younger partners. However, the protective behaviours among HIV-positive youth and those whose most recent partner was casual shows that independent of age-disparate relationship dynamics, young males do exhibit some agency in sexual behaviour decision-making. Policies such as the Adolescent and Youth Health Policy 2016–2020, which aims to reduce the burden of diseases, including HIV and other STIs in young people, should be cognisant of these results because young men who have sex with men are a sub-population of youth who are at risk of unsafe sexual practices. Programmes including the Youth-Friendly Services within the public healthcare system should incorporate sexual-preference sensitivity programmes to cater specifically for the needs of young males with older, same-sex partners in the country.

Additional information

Funding

This work was supported by the South African National Research Foundation [N/A].

Notes on contributors

Nicole De Wet-Billings

Nicole De Wet-Billings, PhD, is an Associate Professor the Demography and Population Studies department at the University of the Witwatersrand. Her research interests include adolescent health and mortality outcomes in South Africa. Brendon K. Billings, PhD, is a Lecturer and Curator of collections in the School of Anatomical Sciences at the University of the Witwatersrand. Among his research interests include clinical anatomy, with a main research interest in comparative neuroscience.