Sexual and reproductive health and rights knowledge, perceptions, and experiences of adolescent learners from three South African townships: qualitative findings from the Girls Achieve Power (GAP Year) Trial

Background: Adolescence is a time of psycho-social and physiological changes, with increased associated health risks including vulnerability to pregnancy, HIV, sexually transmitted infections, and gender-based violence. Adolescent learners, from three townships in South Africa, participated in a 44 session, after-school asset-building intervention (GAP Year), over 2 years providing sexual and reproductive health (SRH) education. This paper explores adolescent learners’ SRH, sexual risk and rights knowledge; perceptions about transactional sex; and contraceptive method preferences and decision-making practices. Methods: The intervention was conducted in 13 secondary schools across Khayelitsha, Thembisa, and Soweto, South Africa. A baseline survey collected socio-demographic data prior to the intervention. Overall, 26 focus group discussions (FGDs): 13 male and 13 female learner groups, purposively selected from schools, after completing the intervention (2 years after baseline data collection). Descriptive analyses were conducted on baseline data. Qualitative data were thematically coded, and NVivo was used for data analysis. Results: In total, 194 learners participated in the FGDs. Mean age at baseline was 13.7 years (standard deviation 0.91). Participants acquired SRH and rights knowledge during the GAP Year intervention. Although transactional sex was viewed as risky, some relationships were deemed beneficial and necessary for material gain. Negative healthcare provider attitudes were the main barrier to healthcare service utilisation. There was awareness about the benefits of contraceptives, but some myths about method use. The injectable was the preferred contraceptive method, followed by the implant, with equal preference for condoms and oral pill. Conclusions : An afterschool intervention at school is a viable model for the provision of SRH and rights education to learners. Recommendations include the need for risk reduction strategies in the curriculum, dealing with misconceptions, and the promotion of informed decision making. Endeavours to ensure health services are youth friendly is a priority to limit barriers to accessing these services.


Introduction
Adolescence is a time of dynamic psycho-social, physiological and sexual change, accompanied by an increase in health-related risks 1 . Understanding this for young people in sub-Saharan Africa is particularly important where approximately 3 out of 10 unmarried adolescents have ever had sex 2 , and more than 50% of rural adolescent girls and young women (AGYW) (15-24 years of age) are estimated to have been pregnant before the age of 18 3 . With an increased vulnerability to early and unintended pregnancy, sexually transmitted infections (STIs), gender-based violence (GBV) and human immunodeficiency virus (HIV) 4 , AGYW in sub-Saharan Africa account for one quarter of new HIV infections, even though they comprise only 10% of the population 5 . Adolescent girls (aged 15-19 years) are particularly vulnerable, comprising 80% of all new HIV infections in sub-Saharan Africa, and twice as likely to get HIV than their male counterparts 6 . This is driven by multiple cross cutting socio-economic and structural factors, including harmful gender norms and inequalities, social and cultural norms, transactional and age disparate sex, sexual and intimate partner violence, and poor access to social and health-related services 7-9 . Access to knowledge is also of concern -with HIV-knowledge in sub-Saharan Africa being low 10 , for example only 36.4% and 29.8% of young men and women, respectively, were shown to have a basic knowledge about HIV prevention 3 .
The situation in South Africa mirrors trends in sub-Saharan Africa. Young people, particularly AGYW, are disproportionally affected by HIV: young people (aged 15-24) represented more than a quarter of all new infections in 2017, and AGYW were three times more likely to be infected with HIV than their male counterparts 11 . AGYW have a 1.5% incidence and 26.3% prevalence of HIV infection 12 . A study in rural KwaZulu-Natal in South Africa, demonstrated high STI prevalence in 15-24 year olds -14% of all had a curable STI (chlamydia, gonorrhoea, syphilis or trichomoniasis), and the prevalence of bacterial vaginosis was 41.1% in women aged 15-19 years 13 . GBV rates are high in South Africa, with between 25-40% of women having reported experiencing sexual and/or physical intimate partner violence in their lifetime 14 , and 17% of women receiving HIV treatment reported sexual violence in their lifetime 15 . Due to the underreporting of violence, the true incidence of GBV is likely to be higher 16,17 .
Poverty and unemployment rates are high in South Africa. In the first quarter of 2021, 32.4% of 10.2 million young people (aged 15-24) were not in employment, education or training 18 . Consequently, young people may be motivated to engage in transactional relationships with older partners for financial or material gain 19,20 , known as the "sugar daddy" or "sugar mommy" phenomenon 21,22 , and as "blessers" and "blessees" in South Africa 23,24 . Due to the inequitable power dynamics in these relationships, many young women may be subjected to GBV 21 and may not be able to negotiate condom use 25,26 , putting them at increased risk for unintended pregnancies, STIs and HIV acquisition [20][21][22]27 .
Adolescents face numerous barriers when accessing sexual and reproductive health (SRH) and, more specifically, contraceptive services. Adolescents in sub-Saharan Africa have unmet needs in relation to contraception 28 . The main barriers facing young people accessing contraceptive services include fear of stigma, shame and embarrassment, lack of privacy, provider attitudes and lack of knowledge about services and contraceptive options [29][30][31] . There are also barriers relating to providers being judgemental, and gaps in their contraceptive knowledge, skills and training to respond to the specific needs of adolescents 31-33 . In South Africa, there are several initiatives and programmes to dismantle barriers and standardise the package of adolescent and youth friendly services 34-36 . This is supported by government guidelines which emphasise the importance of accessible and acceptable services for young people 37-40 . However, although these guidelines exist, implementation at facility level has been challenging, and service provision is not always youth friendly 35 .
Adolescent contraceptive use and knowledge is poor in Sub-Saharan Africa 28 . Evidence from demographic health surveys (DHS) in sub-Saharan Africa showed that most adolescents (92.4%) (aged 15-19 years) reported no contraceptive use, although 21.6% reported recent sexual activity 41 . A later review of DHS data reported that only 24.7% of AGYW (15-19 years) in 29 sub-Saharan African countries used modern contraceptives 42 , with the majority preferring injectable contraceptives (39.9%) and oral pills (31.4%) 41 .
In South Africa more specifically, the most recent DHS in 2016 43 showed that 60% of sexually active women were currently using a method of contraception, and 19% of sexually active women had an unmet need for contraception 43 . The 2012 National HIV Prevalence, Incidence, and Behaviour Survey 44 , found that about a third of women aged 15-19 were using modern contraception 45 . Furthermore, 33.5% of women had reported a pregnancy in the last five years, and only a third of these had desired to be pregnant 44 . Specifically, only 10.1% of women 15-19 years, and 20.9% of women aged 20-24 years had desired the pregnancy 44 . Of those women who reported a pregnancy in the past five years, only 12.8% of 15-19 year olds, and 19.7% of 20-24 year olds, had been using contraception 44 . In terms of method preference, injectables are the predominant method used, followed by oral contraceptive pills 45 . The implant was introduced into South Africa in 2014, but only 4% of South African women were using it in 2016 40,46 . Research on adolescents in South Africa (Limpopo Province) demonstrated that most did not have knowledge about the emergency contraceptive, intrauterine device (IUD) or female condom 47 . Contraceptive preferences are shaped by choice, provider bias and training, demand creation, availability, misinformation and side effects 48,49 . Other factors affecting uptake and use are linked to pressure from male partners, fear of parental reaction, poor contraceptive education, and counselling 47 . Therefore, ensuring contraceptive choice is critical for adolescents to have options when making contraceptive use decisions.
Age-appropriate sexuality education is seen as an important intervention for mitigating against risk in young adolescents. Early adolescence is seen as an ideal time to conduct education with a focus on gender and rights, before harmful gender norms become entrenched, enabling improved SRH and non-violent outcomes 50 . School and curriculum based educational interventions have been implemented globally, seeking to improve adolescent SRH outcomes 51 . This has largely been informed by a rigorous evidence review of over 77 randomised controlled trials and systematic reviews conducted by UNESCO 52 , evaluating the impact of Comprehensive Sexuality Education (CSE). The findings clearly demonstrate that CSE produces favourable outcomes, specifically reducing sexual behaviours that put them at increased risk, sexual debut, number of sexual partners whilst improving health seeking behaviour, knowledge and uptake of preventative services such as contraception and condoms. Peer-led interventions have demonstrated moderate effectiveness at improving behavioural outcomes, such as increasing HIV-related knowledge and improving condom usage 53 . The importance of locating CSE within the context of a sex positive approach focussing on sexual pleasure, sexual preferences and expression of sexuality within the context of safety and rights has also been advanced as an important dimension of risk reduction and sexual health promotion strategies 54,55 .
Young adolescence (10-14 years) in particular, is the time of onset of puberty and sexual maturation and is a transitional period that shapes later adolescent and adult health behaviours 50,56,57 , including SRH behaviours which may cause diseases later in life 10 . There is very little research on the SRH of this younger adolescent population, for both males and females. There is a need to focus on this missed and neglected age group to promote optimal SRH in future 10 . Therefore, this research included younger adolescents (from 12 years of age) seeking to further understand their knowledge, experience and needs.

Girls Achieve Power (GAP Year) Trial
Girls Achieve Power (GAP Year), a cluster randomised controlled trial (cRCT), was a social asset building intervention focussing on SRH, across 26 schools in Gauteng and the Western Cape Provinces in South Africa 58 . It provided SRH information aligned to the CSE curriculum and sought to contribute to and complement the CSE programme 59 . The schools were in three peri-urban townships: 14 schools in Khayelitsha (Western Cape Province), six schools in Soweto, and six schools in Thembisa (Gauteng Province). Schools were selected in collaboration with the Department of Basic Education using the following inclusion criteria: mixed sex public high schools in three townships; in quintiles 1-3 1 which had not been exposed to any asset building interventions in the past six months. A one-to-one (1:1) random stratification scheme was employed, assigning each school to either intervention or control groups, with 13 schools in each study arm.
The GAP Year Trial tested the effectiveness of a four-pronged ecological intervention: a sports-based peer facilitated afterschool asset-building intervention, a parent/guardian component, linkage to care, and school safety. The intervention is described in further detail in Kutywayo, et al 58 . For the after-school intervention, a two-year curriculum (of 44 sessions in total) was designed and implemented, delivered after school by peer coaches, covering different aspects of SRH and rights, HIV, STIs, contraception, decision making, sex and gender. The trial aim was to reduce school dropout among adolescent girls between grades 8-10 and increase reporting of GBV. Complimenting these outcomes, the intervention sought to improve adolescent girls' agency and safety while shifting gender attitudes and encouraging positive behavioural change among adolescent boys.
For this manuscript, we explore the SRH knowledge and behaviour, including risks faced, of grade 9 and 10 learners, from three townships in the Western Cape and Gauteng Provinces, who participated in this intervention. We focus on four thematic areas: adolescent learners' knowledge about SRH and understanding of sexual risk, risk reduction and rights; perceptions related to transactional sex and age disparate relationships; health seeking behaviour and experiences; and contraceptive method preferences and decision-making practices.

Study design
A cross sectional qualitative study utilising data from single sex focus group discussions (FGDs) explored adolescents' knowledge and experiences on health, SRH (including contraception knowledge and uptake) and behaviours that put them at increased risk. The FGDs were conducted after the two-year GAP Year intervention had been completed. No participant demographic information was collected during the FGDs. FGD participant socio-demographic and sexual behaviour data was extracted from a baseline survey conducted with all GAP Year participants prior to the implementation of the GAP Year intervention [60][61][62] .

Study setting and population
The qualitative study was conducted in all 13 intervention schools in the three South African peri-urban townships, Khayelitsha (7 schools), Soweto (3 schools) and Tembisa (3 schools). 1 Schools receive money from government according to Quintiles. Quintile 1 schools being the poorest receive the highest allocation per learner while Quintile 5 receives the lowest. Quintiles 1-3 are classified as no-fee schools (https://pmg.org.za/committee-meeting/30934/) All grade eight learners, at selected schools were eligible to participate in the baseline survey and the GAP Year intervention, irrespective of sex, age, or race. The study population for the FGDs consisted of grade 9 and 10 male and female learners who were enrolled in the GAP Year Trial and had participated in the two-year GAP year intervention. Stratified purposive sampling 63 was used to select learners for participation in the FGDs. Attendance data was extracted from the GAP Year intervention registers. Only those that had participated in a minimum of 32 of the 44 sessions of the two-year GAP Year intervention were invited to participate. Eligible participants were recruited face to face, by peer coaches, who went to the schools and invited them to participate in the FGDs. The sample size for the FGDs was between seven to ten participants, based on the average for this methodology 64 . One female and one male FGD was conducted for each school in the intervention study arm.
FGDs were conducted either in GAP Year schools or at a local venue in the community. Participants were assigned a unique number, to ensure anonymity when identifying themselves for the audio recording. Prior to starting the FGD, participants were reminded that, due to the nature of FGDs, they were not anonymous, but participants were encouraged to keep what was discussed in the group confidential.

Data collection
Collection of demographics, health seeking and behavioural data. A baseline survey was conducted as part of the GAP Year trial with all GAP Year participants (see Extended data 65 ). Data for the first section (demographics, knowledge and attitudes) was collected by trained fieldworkers and captured directly onto an android tablet, formatted with the Research Electronic Data Capture (REDCap, RRID:SCR_003445) system 66 , lasting 45 minutes to 1 hour. The second, a behavioural section, lasted 20-30 minutes and was administered using the audio computer assisted self-interviewing (ACASI) method, seeking to reduce social desirability, because of the sensitive nature of the questions. For the purposes of this manuscript, key variables from the baseline survey have been extracted for the FGD participants, to explore their background and demographic characteristics and contextualise their qualitative discussions.

Collection of FGD data.
A total of 26 FGDs were conducted, with 13 male and 13 female groups, respectively. Participants were allocated to a male/female group, based on their self-identified sex, indicated on their baseline survey. Semi-structured FGD guides collected data based on the content of the GAP Year intervention facilitators' manuals (see Extended data 67 ). FGDs were conducted in isiXhosa, isiZulu, Sesotho or Sepedi, the different local languages, as was preferred by the participants. They were between 45 minutes and an hour in duration. Data saturation was reached, and no more FGDs were conducted after these.
FGDs were conducted in 2019 and 2020 and were facilitated by one experienced researcher (facilitator) and assistants. The facilitators were Black South Africans with master's degrees.
The assistants were coach mentors and assisted the facilitator with logistics and note taking. Facilitators and assistants were trained in research ethics, qualitative research methods and the study protocol, and were proficient in the local languages spoken at the study sites. Female facilitators and assistants conducted the female learner FGDs, and the male learner FGDs had a combination/variation of male and female facilitators and assistants. There were no existing relationships between facilitators and participants prior to data collection. Learners were informed of the aims of the data collection and that the researchers were part of the research team. Immediately following each FGD, the researcher and assistant completed field notes and observations of the FGD.

Data management and analysis
The completed baseline surveys were stored on encrypted password-protected tablets and the synced data was stored on secured organisational servers. All data from the REDCap 66 and ACASI systems were exported into Stata 17 (RRID:SCR_012763) 68 , and descriptive analyses were conducted to describe the socio-demographic variables of the FGD participants.
The FGDs were transcribed and translated directly into English by the researcher and checked by the research assistant for accuracy. Three coders (a researcher, a research assistant, and an external consultant) generated codes for male and female learners iteratively, based on input from the questions in the guides as well as from emergent themes from the data. Data were double coded to ensure reliability of coding. Data coding and analysis was facilitated using NVivo 12 (QSR International, RRID:SCR_014802) software 69 . A deductive approach to data analysis was used in this study.

Ethical approval and considerations
The University of the Witwatersrand's Human Research Ethics Committee (HREC) approved the GAP Year trial (#M160940) in October 2016. FGDs formed part of the trial and were included in the ethics approval from the onset. The provincial research committees at the Western Cape and Gauteng Department of Basic Education, and each participating school also provided written approval. Parents or guardians provided written informed consent and written assent was provided by participating learners for participation in the FGD and for the FGD to be audio recorded. The participating schools, parents and learners were fully informed about the voluntary nature of participation in the study, and of confidentiality of data management.

Demographics
Demographic details were collected at baseline, approximately two years prior to conducting the FGDs (Table 1), therefore all details presented in the table are representative of baseline, prior to the intervention and two years prior to the FGDs. Of the 194 who participated in the FGDs, 38.2% (n=47) were female, 61.8% (n=76) male, and there were missing data for 71 participants 67 . In total, 78% (n=96) were aged 12-14 years, with the mean age being 13.7 years at baseline (acknowledging the natural aging of the cohort for this study). The majority were Black African (99.2%, n=122). At baseline, 22.8% (n=28) had ever had sex: of those, 89.3% (n=25) had had their sexual debut at 14 years or under. At baseline, 20.3% had ever used contraception (n=25).
FGD results are grouped into the following four thematic areas which will be unpacked in turn: adolescent learners' knowledge about SRH and understanding of sexual risk, risk reduction and rights; perceptions related to transactional sex and age disparate relationships; health seeking behaviour and experiences; and contraceptive method preferences and decision-making practices.

Consequences of age disparate and transactional sexual relationships.
In spite of the perceived benefits of having a blesser/sugar mama (described above), adolescent males and females also discussed various negative consequences of being in these types of relationships. The most common concerns were related to the unequal power in these relationships, where adolescents could be controlled and forced to do things they didn't want to do. Female participants were asked if they accessed SRH services. Although some noted that they were not sexually active, and did not access SRH services, others went for HIV testing even if they believed they were not at risk of acquiring it from sexual activity. The majority of participants however, had negative experiences, resulting in barriers to them accessing healthcare services. Some females from all three townships described how the attitudes of healthcare providers actually prevented them from accessing services they may need. Contraceptive method preferences, perceptions, and decision-making behaviour Contraceptive method preferences. Preferences for contraceptive methods were closely related to the perceived advantages and disadvantages of the different methods. Female participants were asked about their own and their peers' preferred contraceptive methods. Responses were general, with many females talking about community preferences rather than their own preferences for contraceptive methods.
Participants from all groups described a preference for injectable contraceptives. This was followed by the implant, and next was an equal preference for condoms and oral pills. This was followed by the IUD and the patch and abstinence. There was minimal awareness of emergency contraception.

Perceived benefits of different contraceptive methods.
The perceived benefits of various contraceptive methods were closely linked to method preferences. Females favoured long-acting methods (implant and IUD) and/or injectable contraceptives because using these methods did not require frequent visits to a healthcare facility. An added advantage of the injectable contraceptive was that one didn't need to remember to take it daily.

P11: "I think the injection (is preferred) because you can forget to take your pills and fall pregnant". (Soweto, Group 1, females)
One participant felt that the lack of menses (and no period pains) experienced as a side effect of the implant was beneficial.

P03: "[…] the implant because I don't want to go on my periods, so my mother says that it depends on your body and the side effects are not the same, it is only for people who have grown and don't want to have periods". F: "Okay, why don't you want periods?" P03: "Period pains eish (sigh)". (Soweto, Group 3, females)
Condoms were preferred by a few females who suggested that they were easier to access, and that they did not need to go to healthcare facilities (where they were not treated well) to access them. In addition, the choice of flavours was described as an advantage of condom use. Perceived barriers to contraceptive use. Some of the described barriers to contraceptive use were linked to method side effects and others to myths and misconceptions about contraceptive methods.
The most common disadvantage of a contraceptive method, described by adolescent females, was that people could forget to take their oral contraceptive pill. In addition, participants from one group from Soweto mentioned that if emergency contraception was not taken timeously, it would be ineffective. There were concerns about the injectable -some noted that people may fear injections, others suggested that side effects, such as weight gain associated with use, were a concern.

PID unknown: "The three months (injectable contraceptive) makes you fat". (Tembisa, Group 3, females)
The side effects related to menses -excessive bleeding or amenorrhea, were viewed as disadvantages of the injectable contraceptive or oral pill by some. One participant from Soweto had concerns that lack of menses when using an injectable contraceptive may impact on growth and development. Furthermore, females in one group from Tembisa also raised concerns of future infertility from using the injectable or oral contraceptive.
P09: "I think some of them they have bad, uh they have disadvantage because like maybe for who are getting injected or those taking pills it could stop them from making children in future.
[…] Yah so when they face the problem like they won't have children because of the things they were using". (Tembisa, Group 2, females) The implant was perceived to be dangerous to use by some participants from Khayelitsha -with the rumour that it could be stolen from their arms, as it was believed that people use the implant for drugs.
PID unknown: "The skollies (bad people) smoke implant. So, they mug you off it". PID unknown: "They pull it out of your arm because it is visible, so they cut you and then remove it from your arm". (Khayelitsha, Group 2, females) In addition, females from a group in Soweto had concerns that the implant was not an effective contraceptive method (possibly related to drug interactions), and had false beliefs that it could rust inside one's arm. The IUD was believed, by one participant from Khayelitsha, to take up space inside the vagina. Condom use was also described as a challenge by participants from Soweto who said that people prefer condomless sex.

PID unknown: "Others don't want a condom they want skin to skin". (Soweto, Group 3, females)
Contraceptive decision making. Females were asked who makes the decision for young females to start using contraception. The majority of them noted that it was either their parents (usually specifying their mother), or themselves that made the decision.

Discussion
Adolescents in this study demonstrated that they had acquired SRH knowledge during the GAP Year intervention, including information on SRHR, risk reduction and where to obtain services. They also learnt about HIV and STI acquisition and prevention, dual protection, as well as about various contraceptive methods.
Our findings confirm that transactional sexual and age disparate relationships were occurring in these communities. As has been reported elsewhere, adolescents engage in these relationships for material gain -sometimes for survival, especially in poor communities, and sometimes for improved lifestyles 19,20,27 . In this study, it was also reported that in some cases adolescents engage in these relationships to seek love and affirmation which they do not get at home. This reinforces the blurred line between transactional sex in exchange for money/goods, and relationships based on need, intimacy and love 27 . Despite the perceived benefits of these relationships, adolescents in the FGDs had an awareness that they could also have negative consequences, where power imbalances result in gender inequalities, and the potential for risk of GBV. Lack of power was also seen to reduce capacity to negotiate safer sex practices. These negative factors can result in increased sexual health risk -including pregnancies, HIV and STI transmission. However, despite the perceived risks of these transactional relationships, participants described some level of acceptability and support, also noted elsewhere 70 , especially in the context of poverty, and where such relationships lead to improved living conditions for an individual and their family 71 . It is therefore also important to be cognisant of the contextual factors framing transactional sex and social norms whereby it is acceptable and sometimes sanctioned as a viable source of income and means for procuring both essentials such as school fees and food, and non-essentials such as lifestyle-related commodities 27 . The awareness of age disparate relationships was not explored, and this is of concern, given the association with increased risk of HIV 8 . Of particular importance for this paper, is the perspective that interventions such as educational programmes need to be cautious not to judge and cast all transactional sex as inherently wrong, but rather strike the balance between acknowledging the potential motivations and benefits on the one hand, and on the other, discussing strategies to mitigate risks and protect themselves 27,72 Adolescent access to SRH services is a challenge globally, and more specifically in sub-Saharan Africa 29-31 . Similarly, most adolescents in this study reported barriers to accessing SRH services, largely related to provider attitudes and fear of stigma. A few even chose not to access these services because of barriers, putting them at increased risk of unintended pregnancies, STIs and HIV acquisition. Reassuringly, some participants described positive experiences accessing SRH services at healthcare facilities, largely related to strategies they had developed to improve their experiences, for example choosing to attend a healthcare facility with a reputation for being youth friendly, or accessing services with a parent, to avoid judgmental attitudes. A few participants described making use of traditional healers or home remedies for health-related concerns.
Preference, choice and uptake of contraceptive methods has been linked to access to and knowledge about methods 49,73 . Perceived benefits and challenges with contraceptive use are impacted on by personal situations and understandings of use, further compounded by myths and misinformation around particular methods. The GAP Year intervention provided information on male and female condoms, the injection, emergency contraception, the implant and the IUD. Participants demonstrated an understanding of the benefits of long-acting reversible contraception (LARC), and also mentioned emergency contraception.
The most preferred contraceptive method in this study was the injectable, reported in other research in South Africa and sub-Saharan Africa 41,43,74 . Young females in this study also had positive perceptions about LARCs (including both the IUD and implant). The reasons for this were similar to those associated with using injectable contraceptives -the fact that they do not need to remember to use it daily, and also because of the less frequent need to access healthcare services when using these methods, also found in research with implant users 46 . This is important as South Africa's method mix is still dominated by injectables, and awareness and use of long acting methods is low 45 . The promotion and awareness of LARCs is emphasised in the DOH guidelines 75 , as well as the promotion of informed choice, including implants and IUDs. LARCs have a higher efficacy and continuation rate 45, 76 and are suitable for use by adolescents 77 . Adolescent preferences for LARCs and injectables could also have implications for long-acting HIV prevention methods, including injectable PrEP and vaginal rings -where less frequent access to healthcare services is required. Similarly, several participants noted a challenge with the oral contraceptive pill in terms of remembering to take it daily, and this too needs to be taken into account with daily oral PrEP. Individual product attributes should be considered together with these perceived advantages when designing information sessions and promotional activities for long-acting HIV prevention products.
Individuals also expressed concerns about some methods, and many of these were related to side effects (real or perceived) and misconceptions, especially regarding the implant, in particular the belief that it is stolen from the arm and used for drugs, the misconception that it can rust in one's arm, stunted growth, infertility, and low efficacy. Such concerns and misconceptions have been noted elsewhere 46 and can discredit the use of methods. It is therefore important that rumours and misconceptions are addressed in order that they do not undermine contraceptive programmes and method choice 48, 78 .
Decision making about contraceptive method uptake and choice was largely made together with parents (mothers in particular), or by the adolescent females themselves. This could be linked to the fact that the GAP Year intervention engaged parents, encouraging parent-child communication. Given that evidence shows that parents experience challenges discussing SRH issues with their children 71 , more needs to be done in future programmes focusing on parents/guardians and how to communicate effectively about sex. Only a few young females reported that their partners played a role in their contraceptive choice and use, this differs from research which shows that male partners influence contraceptive uptake and use in other areas in South Africa 79 .
The GAP Year intervention provided these learners with important SRHR information and strategies to reduce sexual risk. One benefit that participants noted was that they learnt how to use both male and female condoms. This was covered in the Year 2 curriculum but then reinforced during the graduation events, where a local adolescent SRH organisation repeated these practical demonstrations. The importance of not only promoting condom use and dual protection in HIV/SRH prevention programmes, but also providing practical guidance on the use thereof is important, and sometimes neglected 80,81 . However, it is also important to note that knowledge about condoms does not always translate to being able to, or wanting to, practice safer sex practices. This underscores the need to discuss risk, condom use and HIV prevention within the context issues related to sexual pleasure, intimacy, gendered power, and sexuality, in other words, unpacking the complex factors that influence condom use 82 .
There are several recommendations that can be made based on the findings from this study. Adolescents need to be provided with education and information on SRH rights and contraceptive methods, relative benefits of each, mechanisms of action, and side effects, as well as on service delivery points. Furthermore, these programmes need to address myths and misinformation and replace them with facts to ensure informed decision making and choice, and to strengthen uptake and correct method use. the viability of improving access to contraception at schools needs to be further researched and explored 33 . Increased dose of programmes/interventions and reinforcement of information is necessary to ensure that myths and misinformation are not perpetuated. Practical demonstration and discussions about condom use in the context of young people's sexual lives also needs to be included. Although the afternoon sessions focussed on SRH within the context of gender disparities, gender-based violence and SRH rights, closer attention to locating SRH within a holistic framework of sexuality education is recommended. This needs to be included in the training of facilitators conducting the sessions and allows for the shifting of sex from the realm of problems and risk to a more sex positive approach, where SRH is discussed within the context of safe sexual pleasure, sexual relationships and self-expression 54,83 .
This can be done via school-based education programmes 29 like the GAP Year intervention which complements the current CSE curriculum in schools and reinforces health-seeking behavior and use of peer mentors. To this end, linking school-based educational programmes with services responsive and sensitive to the needs of young people is vital, coupled with healthcare provider training and education to address knowledge gaps and judgmental attitudes 10,33,84 . In addition, providing adolescent health services at a community level, at health education events, could increase uptake and acceptability of services. Cash transfer systems and other structural interventions have been demonstrated to reduce HIV risk behaviours such as transactional sex in young girls, and should be further considered 85,86 . Finally, as has been suggested elsewhere 50 , using evidence from studies on adolescents is important for informing future policies and programmes that are effective and relevant to adolescents.

Strengths and limitations
Whilst this qualitative study is not an evaluation of the broader GAP Year Trial (trial results to be published elsewhere), the FGDs provide insight to the SRH knowledge gained as well gaps in information of some learners who participated in the GAP Year intervention. Male and female learners who had participated in the programme were asked different questions during the FGD and so comparison between groups, with regards to sex-specific knowledge and experiences was not always possible. However, there is sufficient data to get a sense of the different issues encountered by each group, and where differences have been noted, these have been highlighted.
There are also potentially a few gaps in the information gathered. Data regarding non-participation was not collected during the recruitment phase for the FGDs. Participants were not asked any questions about LGBTQI (lesbian, gay, bisexual, transgender, queer and intersex) relationships and their access to SRH services, so any challenges or experiences for these adolescent groups have not been explored. Furthermore, although participants were asked about transactional sex relationships, there were no probes further exploring age disparate relationships and the impact of this on SRHR.
There is a high portion of missing data in the demographic data presented in this manuscript, resulting in a skewed representation of male participant demographic responses. The FGDs were evenly split according to male and female groups, and therefore it is likely that much of the missing baseline data would have been female participant responses. Although baseline demographic data were extracted for the FGD participants, they are not directly linked to FGD participant responses. Furthermore, they were collected two years prior to the FGDs and therefore it is possible that some of the socio-demographic information could have changed over time.
For example, the proportion of the FGD participants who were sexually active at baseline, may have changed by the time of the FGD.
Since FGD participant responses are not directly linked to the demographic data, and because of potential changes over time, it is not possible to determine which FGD participants had actually engaged in sexual activity. For this reason, some FGD participants may not have actually accessed SRH, HIV and family planning services, and therefore may have responded to questions based on hearsay, or social desirability bias.

Conclusions
There is a paucity of literature exploring SRH and contraceptive knowledge and perceptions of younger adolescents in South Africa. This research has been important in providing information on the SRHR knowledge and perceptions of a group of adolescent learners. Furthermore, these learners demonstrated that they had acquired knowledge during the GAP Year intervention, highlighting that an afterschool model of education provision within a school setting is a feasible model for provision of SRHR education to learners.
Findings from this study highlight what can be included in future interventions. Youth focussed interventions should focus on information provision to facilitate understanding and informed decision making. For example, the mechanism of action of contraceptives should be described to address and dispel myths around method use. Full information about different methods available, including LARCs, are important for method choice.
In addition, condom promotion should provide practical information and instructions on condom use. The provision of information enables informed choice, decision-making, and method use, and could facilitate uptake of contraceptives. Furthermore, more work needs to be done to limit barriers to accessing healthcare services and to facilitate more youth friendly facilities, including reconsidering healthcare training strategies. Finally, discussions about HIV/SRH prevention need to integrate together with messages about choice and informed decision-making relating to both HIV prevention and contraceptive options.

Data availability
Underlying data The transcripts have been deidentified; however, due to the nature of the topic, they are not openly available. The transcripts can be accessed by emailing the corresponding author (mpleaner@mweb.co.za). A valid request, which will be considered by the authors, is required to access the transcripts.

Jane Harries
Division of Social and Behavioural Sciences, School of Public Health, University of Cape Town, Cape Town, South Africa This is a well written article on an important topic. I suggest a copy edit as I picked up a few typos.
My comments are meant to strengthen the article and many of the issues have already been addressed by the previous reviewer.
Abstract: "NVivo was used for data analysis" -please change to 'NVivo was used for data management'. Software packages can assist in data management and sorting but can not analyze qualitative data.

Myths:
The term "myths" throughout the manuscript in relation to contraceptive use is somewhat problematic and is not clearly defined. Local beliefs need to be acknowledged and examined and too often local beliefs are not interrogated and loosely defined as myths. These so called myths are not really discussed apart from related to implants being removed -"belief that it is stolen from the arm and used for drugs" this has been reported as real in SA and is not surprising with high levels of violence -see reference Harries et al. (2019 1 ).
Typos: "I leant that when sleeping with a girl you might get a disease without using a condom … " replace 'leant' with 'learnt'. Pg.13: "the viability of improving access to contraception at schools needs to be further researched and explored." Should be "The viability ... ".

Results:
Results section could be more synthesized as there are many quotes per sub heading that are saying similar things. The narrative gets lost with similar quotes on one topic area. This is up to the authors to decide.
relationships. It would add depth to include discussion of how sexuality is situated in young people's perceptions and experiences relating to SRH and sexual behaviour. How was sexuality covered in the GAP Year curriculum and materials?
There could be a more in-depth discussion of the sociocultural framing of transactional sex and relationships in these communities, that normalise it, and in which people consider it a viable source of income. Also in terms of sociocultural informed gendered expectations of exchange.