Contraception knowledge and uptake among in-school adolescents in three South African townships: Baseline findings from the Girls Achieve Power (GAP Year) Trial

Background: South African adolescents experience barriers to sexual and reproductive health (SRH) knowledge and uptake. This study provides insight into contraceptive and other SRH service knowledge, perceptions, and uptake among adolescents in high HIV prevalence settings. Methods: A baseline cross sectional survey was conducted among 3432 grade 8s enrolled into the Girls Achieve Power (GAP Year) trial from 26 public high schools across three South African townships (Soweto, Thembisa and Khayelitsha) (2017 - 2018). An interviewer-led survey collected information on SRH knowledge and perceptions; an audio computer-assisted self-interviewing technique gathered SRH service uptake. Descriptive analysis indicates frequency distribution of socio-demographics and knowledge, uptake and perceptions of SRH services. Chi-square test tested for associations between age and sex and selected variables that measure SRH knowledge and uptake. Results: In total, 2383 learners, completed both survey components. Of these, 63.1% (n=1504) were female and 81.4% (n=1938) aged 12-14. Almost a fifth (18.3%, n=436) had ever had sex and less than 1% had accessed SRH services in the last year. Of the 157 females who had ever had sex, 50.9% had ever used contraception. Of those who had sex in the last three months, 59.0% reported using a contraceptive method. Condom use was inconsistent: almost all females said they had not used or could not remember if a condom was used at last sex. Conclusion: This paper contributes to the evidence strengthening learner SRH education, including the national Integrated School Health Programme. Key themes include the need for age-appropriate, differentiated comprehensive sexuality education (CSE) for the range of ages found in the same grade in South African schools. Education on different contraceptive methods, informed decision-making, and emergency contraception is key. School-based interventions should embrace integrated HIV, STI, and pregnancy prevention messages. Closer links with health services need to be constantly fostered and reinforced.


Introduction
Globally, there has been significant focus on the potential dividends to be gained by investing in and prioritising adolescent health, including sexual and reproductive health (SRH) and HIV. Although great strides have been made since the SRH historical landmark -the International Conference on Population and Development (ICPD) in 1994 -there are still multiple challenges in accessible SRH service provision for adolescents, resulting in poor health outcomes 1,2 . This is particularly so for adolescent girls and young women (AGYW) in sub-Saharan Africa, who continue to experience significant inequalities, challenges and unmet needs for SRH services, together with an associated increased risk of unintended pregnancies, HIV, STIs, complications related to early age pregnancy, and other SRH complications 3,4 . In sub-Saharan Africa, more than 50% of rural AGYW aged 15-24 years, and 42% of those in urban areas have been pregnant before the age of 18, with one in five new HIV infections occurring in this age group, despite representing only 10% of the population 5 . This is primarily attributed to limited information, barriers to access, constraints in exercising their SRH rights, lack of comprehensive programmes and service provision, and in many countries, framed by an unsupportive legislative and policy environment 6,7 .
Although South Africa has progressive and enabling rights-based laws, policies, guidelines and programmes relating to SRH and young people 8-10 , the challenges faced in relation to SRH and HIV mirrors similar trends in sub-Saharan Africa. In terms of HIV, 7% of young people age 15-24 are HIV-positive, with AGYW having a four times higher prevalence rate than their male counterparts (12% versus 3%) 11 . The most recent South African National HIV Prevalence, Incidence, Behaviour and Communication Survey ((SABSSM V) 12 provides insight into some of the most important drivers of HIV in young people in South Africa, including early sexual debut, age disparate sex, and condom use. In terms of sexual debut in young people who have had sex, 13.6% reported having had sex before the age of 15 years. More males (aged 15-24) reported this compared to females (13.6% males vs 7.6% females). This pattern has been consistent over the previous four surveys 12 .
Teenage pregnancy and its consequences are an ongoing problem 13,14 . The 2016 South African Demographic and Health Survey (SADHS) found that 15.6% of South African adolescent females aged 15-19 have begun childbearing. Although adolescent pregnancy has been a major social and economic challenge in South Africa for decades, the prevalence of mostly unintended pregnancies among females aged 15-19 has remained unchanged for the last 20 years 11 . The ramifications of unsupported teenage pregnancy are far reaching and have been well documented, and include early exiting from education 15 , reduced opportunities in employment and further education, socio-economic challenges, associated health risks for both the woman and the infant, increased vulnerably for STIs, HIV and gender-based violence, together with the need for social protection and support [14][15][16][17] .
Contraceptive prevalence in sexually active adolescents aged 15-19 was 60.4% in the 2016 SADHS, with reliance primarily on injectables and male condoms 11 . However consistency in condom use in this age group is poor 12 leaving young women open to unintended pregnancy. Factors contributing to poor contraceptive uptake in adolescents are cross cutting and multi-fold, and include health inequalities, limited contraception knowledge, barriers to accessing services, challenges with regards continuation and correct use, and gender-based violence [18][19][20] .
Knowledge about contraception and various contraceptive options is low among South African youth, this pertains particularly to both long-acting reversible contraception and emergency contraception, where, for example, only 30.9% of 15-19-year-old females knew of intrauterine devices (IUD), and 36% of emergency contraception 21 . A school-based study focusing on secondary school girl learners in grades 10 -12 22 showed that while there was awareness about methods, this information was superficial -58% knew about condoms, 50% injections, 43% oral contraceptive pills, 40% female condom and 10% IUDs, and only 17% knew about emergency contraception, but did not know how to take them. Similarly, there was a lack of information about how to take oral contraceptive pills 22 .
The need to use evidence to understand and design effective, responsive strategies to improve the provision of SRH services for AGYW, and more specifically, preventing teenage pregnancy and the promotion of contraception is now as urgent as ever. However, there is still a gap in research relating to adolescents' use and knowledge of contraception and related services 23 . This study seeks to contribute to the evidence and understanding of the factors influencing adolescent contraception uptake. To this end, we aim to gain insight into the knowledge and uptake of contraception among grade 8 adolescents in a high HIV prevalence setting and utilise the findings to guide interventions to improve access to SRH services, inform comprehensive sexual education (CSE) and strengthen strategies to prevent unintended pregnancies, within the framework of the Integrated School Health Programme (ISHP).

Amendments from Version 1
• The title has been amended slightly, removing reference to ''patterns of contraception use''.
• Text about the GAP Year intervention has been added in the Methods • In the Discussion the GAP Year intervention is now cited as an intervention seeking to address some of the gaps raised from the findings.
• One finding in the Results has been rewritten to provide greater clarity • Recommendations presented in the Discussion and Conclusions have been amended to ensure they focus on those only from our findings.
• Minor grammatical changes have also been made throughout the manuscript Any further responses from the reviewers can be found at the end of the article

Study design and setting
This research is part of the Girls Achieve Power Trial (GAP Year) cluster randomized controlled trial (cRCT) conducted in 26 non-fee-paying high schools, in three peri-urban townships of South Africa: Khayelitsha (Western Cape Province), and Soweto and Thembisa (Gauteng Province) 24-28 . GAP Year was seeking to test the effectiveness of a CSE asset-building intervention aiming to reduce school dropout among adolescent girls between grades 8-10 while shifting gender attitudes and encouraging positive behaviour change among adolescent boys 27 . Half of the schools were randomised to the intervention study arm, whilst the other 13 were in the control arm. The GAP Year intervention adopted a four-pronged approach, across the ecological model: a sports-based after-school intervention, parent intervention which includes dialogues and text messaging, linkage to care and school safety 27 . The aim was to assess school dropout and increase reporting of violence among adolescents as they progress in education. Schools were selected using the following inclusion criteria: mixed sex public high schools in in quintiles 1-3 * which had not been exposed to any asset building interventions in the past six months. The exclusion criteria for the study were: single-sex schools; private schools; schools that have been exposed to similar interventions and public schools that cater for learners with special needs. A baseline cross sectional analysis was conducted for grade 8 learners enrolled in GAP Year to assess the knowledge, patterns and uptake of contraception among adolescent learners. Data collection took place between April 2017 and September 2018 for all 26 schools.

Study population
All grade 8 adolescents at selected schools, irrespective of age, sex or race, were invited to participate. The grade 8 learner age range is approximately 12-14 years however due to learners repeating grades and other reasons, the age range is commonly wider with older learners enrolled up to 18 years 29,30 .

Sample size justification
The sample size was computed using cluster-randomized size methodology, suggested by Hayes and Bennett 31 , based on the study's primary outcome measures, (school dropout and increased reporting of GBV among adolescent girls). The effect size of dropout was factored from other similar studies accounting for a large conservative and representative sample size to measure outcomes. Based on other local studies, it was hypothesized a reduction in drop-out rate from 17.8% as reported by Branson, Hofmeyr 30 to less than 14% (estimated effect size of 20%), with an anticipated attrition rate of 5% per year based on a similar local study 32 . These calculations resulted in an upper limit sample size of 2730 adolescent girls and 1850 boys to determine the association between intervention and control school's dropout rate and GBV variables.

Measures
The baseline survey was comprised of two components: firstly, an interviewer-led questionnaire and secondly, an audio computer-assisted self-administered interview (ACASI). The first section lasted between 45 minutes to 1 hour, providing information on the learners' demographic and socio-economic characteristics, knowledge and attitudes pertaining to school safety, social support and social networks, sexuality, gender and norms, sexual reproductive health and rights (SRHR) and care-seeking behaviours. It was conducted by trained fieldworkers and captured directly on an android tablet, formatted with the Research Electronic Data Capture (REDCap) system 33 .
In the second section, administered using the Audio Computer Assisted Self-interviewing (ACASI) method, learners were asked sensitive questions regarding their actual practices and behaviour, including questions on uptake of health care services, contraceptive use, sexual debut and relationships. This section lasted 20-30 minutes. The ACASI method was adopted, was seeking to reduce social desirability bias, and later exported into the REDCap online system. The survey was developed in English and back translated into Xhosa (a commonly spoken language at one site) and then pre-tested among selected learners for comprehension. Participants indicated that they preferred to complete the survey in English and therefore translation for the other sites was not provided. During data collection, learners were assigned unique person identifiers. As such, names and other personal information of the learners were not revealed in the datasets.
The data were analysed at two levels. Descriptive analysis was used to show the frequency distribution of the sociodemographic characteristics of the respondents as well as the knowledge, uptake and perceptions of sexual and reproductive health services. At the bivariate level, a chi-square test of association was employed to test for an association between age and sex categories and selected variables that measure sexual and reproductive health.

Data management and analysis
The tablets that were used to collect data utilised passwordprotected mechanisms to protect the data and the synced data was stored on Wits RHI secured servers. All the data from the REDCap and ACASI systems was exported into Stata 17 34 for analysis. At the univariate level, descriptive frequency tables were used to describe the socio-demographic characteristics of the learners by age and sex. At the bivariate level, variables measuring knowledge, uptake and perceptions around SRH were assessed to examine their association with both sex and age category. Learners' age was presented in groups (12-14 years and 15-18 years), this is because the 15-18-year age group represent a group that are older than the standard age appropriate range for that grade. They comprised of those who have repeated or skipped a year of schooling or started school later. They may represent a more vulnerable group whose education has been affected by social or economic circumstances. It should be noted that the "no" category has been removed for all dichotomous variables [variables with a Yes or No Response] shown in the results tables. * Quintile 1-3 schools are non-fee-paying schools, serving the poorest communities. Quintile 1 is the group of schools in each province catering for the poorest 20% of learners and receiving the highest allocation of Government funding per learner to support these poorer schools.

Ethical approval and considerations
The study was approved by the University of the Witwatersrand Human Research Ethics Committee (#M160940) in September 2016. The study was also approved by the provincial research committees of the Western Cape and Gauteng Departments of Health and of Education. This was followed by the schools' approval and written parental informed consent and learner assent. The participating schools, parents and learners were fully informed about the voluntary nature of participation in the study, and of the confidentiality of data management. All data collection was supervised by the research team.
Interviewer environments were set up to ensure confidentiality. Where feasible, interviewers were the same sex as the learner. Learners could stop the interview process at any time and were free to refuse to respond to any question(s) they felt uncomfortable answering. Social workers were employed to provide psychosocial support to learners during data collection and study intervention and a social harm form was developed to facilitate prompt referrals, where needed. The Good Participatory Practice Framework was adapted and adopted in the GAP Year trial to guide stakeholder engagement throughout the lifecycle 27 .

Results
As previously reported 24-26,28 , overall, 3432 eligible learners across 26 schools participated in the baseline survey: we included 2383 in the analysis who completed both sections of the survey. In some cases, due to lack of time, learners were unable to complete both components and were excluded from this analysis. Table 1 provides the socio-demographic characteristics of learners, by sex and age group. As previously reported 24-26,28 , of the  Uptake and patterns of contraceptive use in females Table 3 reports on contraceptive history and current use in females who had ever had sex. Males are not presented as most  were unsure about method use their partner may have used/was currently using. Half (51.0%, n=80) who had ever had sex, had ever used a contraceptive method. The injection and condoms were the most common method ever used in the 12-14 age group. In the 15-18 age group almost all (96.4%, n=27) had ever used the injection. Sixty-five females were current contraceptive users although 29 of these had not had sex in the last three months.
Of the 157 females who had ever had sex, 41.2% (n=61) reported having sex in the last three months. Of these 59% (n=36) reported current contraceptive use with two-thirds (69.4%, n=25) using the injection while a quarter (25.0%, n=9) using the condom. Almost half of the injectable users mentioned using condoms as an additional method but not one of these or the condom only group reported use of a condom at last sex. Most knew condoms could prevent HIV and STIs. A small proportion mentioned hormonal methods (implants, injections and oral pills) could prevent HIV and STIs. Similarly, a small number (2.8%) mentioned the vaginal ring.
Of the 80 who had ever used contraceptives, only two said it was their decision to start using a method. Sixty said it was their parents/guardians who made the decision for them, while 18 did not say who had made the decision. None of the females who had ever had sex had ever discussed contraception with their partner.
At the first sexual encounter, just over half (56.1%, n=83) did something to prevent pregnancy, HIV or sexually transmitted infections. Most (82.7%, n=129) knew where to get contraceptives but fewer (26.0%, n= 40) knew that they could get contraceptives without their parents' permission.
Of the 80 females who have ever had sex and ever used a contraceptive method, the majority got the method from a public/ government clinic (33.8%, n=26), followed by a private clinic (23.4%, n=18), pharmacy (10.4%, n=8), corner shop (6.5%, n=5), and friend (5.2%, n=4), while some did not say where they had obtained their method. Of those who reported they had never had sex, a small proportion reported using a contraceptive method.
Five females, of the 157 who had ever had sex, had ever been pregnant (0.3%, n=5): two got pregnant at 14 years, one at 15 years and two at 16 years (p=0.001). Of those who had ever been pregnant, two were pregnant at the time of the survey, one had terminated the pregnancy and two had gone on to have a live birth.

Discussion
We set out to explore adolescent experience with SRH service knowledge and uptake, in particular contraception in three peri-urban settings in South Africa. The learner age profile confirmed what has been reported in other South African studies -that a considerable proportion are above the standard expected age for the grade 35 . These learners are known to experience a number of challenges 30,35,36 , and may be missing out on age-appropriate school-based comprehensive sexuality education (CSE) which would be received if they were in a higher grade.
Although over half the learners reported having accessed healthcare services in the last year, very few reported accessing SRH services specifically, and less than 5% attended a dedicated youth service. Knowledge about HIV and SRH services and rights are important elements for promoting the sexual health and development of adolescents. We found that learners who had used healthcare services commonly experienced a lack of confidentiality and the ability to ask questions, and were concerned about discrimination, this was especially so for those who mentioned healthcare workers as their primary source of SRHR information. Addressing barriers to health care is an integral part of SRH promotion 37 , highlighting the need to improve young people's experience of using health services which are cognisant of their rights.
Dismantling barriers, improving access, and developing effective responses to adolescent SRH and contraceptive programming is vital and there is a growing body of literature attempting to assess and define effectiveness of programmes 37,38 . Studies show that high impact interventions include improved access to youth friendly services 39 , school-based interventions 40 , and in particular well designed, and age-appropriate CSE 38,41 , with an emphasis on focusing on early adolescents, where patterns of behaviour are being formed 42 . In addition, the provision of quality, youth-friendly contraceptive services, sensitive to and responsive to the needs of young people is a fundamental requirement for promoting the SRH of young people 43 , as well as services that promote informed choice and decision-making, including long acting reversible contraception in the method mix 44 . A gap in programmes focussing on younger adolescents and the need to ensure that interventions include this cohort has been highlighted -for example, the Global Early Adolescent Study, which highlights the importance thereof in terms of shaping behaviour patterns, constructing gender norms, and paving the transition into older adolescence for both boys and girls 45,46 . In addition, the need for interventions targeting adolescent males as well has been underscored 47 . The GAP Year intervention seeks to respond to this gap by providing sex specific SRH information through a participatory, sport-based, afterschool intervention 27 .
School teachers and mothers were noted as the most common source of SRH information. This highlights two important priorities -to equip parents with the knowledge and skills to communicate effectively with their children about sex, as well as the need to provide teachers with the knowledge and skills to provide information, supported by comprehensive curricula and policies. Interventions focussing on parents have been shown to be constructive but neglected and should be included in programmatic interventions 48 In terms of sexual experience, over half of the learners were dating or in a relationship and just under a fifth, (18.3%) had ever had sex. Two-thirds had their first sexual experience between the ages 11-14 years, and a small number below this age. Of those who had had sex, many did nothing or did not remember if a method was used to prevent pregnancy, HIV or STIs during that first sexual encounter. The importance of programmes focussing on early adolescents is reinforced, including age appropriate SRH information be provided to ensure adolescents, including very young adolescents, are equipped with the knowledge around safe sexual practices, rights and gender awareness 45,46 . The GAP Year afterschool intervention seeks to address this gap in knowledge 27 . Interestingly, learners reported that parents were commonly involved in their decision to use contraception -this is a potential area for further research.
There was similar contraceptive ever and current use in sexually active females to previously reported data in this age group 11 . It was interesting to note that a number of females who had not had sex in the last three months reported that they were using a contraceptive method as were some who reported never having had sex. Although this was not probed in more detail, this may reflect an intention to have sex, or prevention in the event that they have sex. This, too, is an interesting trend to further research.
Contraceptive method use was similar to national surveys in South Africa with injectables and condoms as the main methods of contraception ever and currently used in sexually active learners 11 . This is similar to other research among AGYW in South Africa 20 , and highlights the need to promote other available contraceptive options, particularly the use of long acting, reversible contraceptive methods which are both highly effective, do not rely on regular visits to the clinic, nor clients remembering to take them daily 55 . The subdermal contraceptive implant, together with the IUD are an important part of the method choice offered to women and are supported by South Africa's national contraceptive guidelines 54 . Both are suitable for adolescents 56 , and, together with short acting methods, form the basket of methods available to women in the public sector and need to be promoted 44,54 . In addition, only two girls mentioned having ever used emergency contraception, and noting that only half who had had sex had ever used contraception, the need to promote emergency contraception is also an important, but underused option 57 . The Standard Operating Procedures for the Provision of Sexual and Reproductive Health, Rights and Social Services in Secondary Schools 9 clearly states the need to inform learners of emergency contraception and where it is available. These guidelines should be specifically targeted to educators involved in school SRH programme delivery.
There are several factors associated with contraceptive uptake, and these all need to be accompanied by strengthened adolescent and youth friendly services and improved outreach concerning available SRH and HIV prevention and services -for example, it was shown that some learners thought the hormonal contraception could prevent HIV and STIs, the majority of respondents reported not using protection at sexual debut, and a very low percentage used condoms at last sex. This calls for ongoing messaging concerning dual protection and the promotion of condom use and other HIV and STI prevention options, such as oral PrEP.
Knowledge, access to, and use of contraception plays a significant role in averting and decreasing millions of unintended pregnancies, births, abortions, and maternal deaths each year. Improving knowledge about contraception, including options, benefits, correct use, where and how to access reliable methods, what to do if a person has unprotected sex, as well as an understanding of their rights can help AGYW make informed decisions that can positively impact their SRH, education and psycho-social well-being 39,58 .
This all serves to highlight the need for interventions that work across the ecological model -empowering individuals; promoting supportive relationships with sexual partners 59 , parents, and peers and teachers and at the community level, such as the GAP Year intervention 27 .

Strengths and limitations
There are strengths and limitations that should be considered when reviewing these findings. Conducted in 26 schools in three highly populated diverse townships of South Africa, this study can be generalised to other South Africa settings. However, the study was cross sectional therefore only representing one point in time. The final study sample size was less than originally planned however the results maintain power to draw these conclusions. There were many participants who didn't complete both components of the survey which reduced the sample size when analysing specific variables. There was a higher proportion of learners aged 12-14 years in relation to those aged 15-18 years increasing power which results in these significant age differences: therefore, the results should be interpreted with caution given that the proportion of learners in these age groups is unbalanced. These limitations are unlikely to alter our primary findings.

Conclusion
This paper aimed to build on the body of evidence to guide school-based interventions to improve the SRH of school-going learners. The paper highlights the need to encourage health seeking behaviour and promote the idea that health services not only for problems but important for prevention and health promotion -especially in relation to sexual health, including HIV, STI and pregnancy prevention. This means the importance of knowing where services are located, as well as building partnerships with public health clinics to ensure that services are accessible and responsive to the needs of young people. The importance of school teachers and parents as the primary source of SRH information for learners requires an expanded, enhanced, programme to equip both teachers and parents to provide accurate, relevant and accessible information. Several points related to life orientation, life skills and CSE were underscored including the need for age-appropriate, differentiated approaches to cater for the range of ages, and particularly for over-aged learners, in South African schools; the need for CSE to start with early grades; as well as the need to deal with issues rooted in young people's realities and sexual lives -such as communicating with sexual partners about safer sex, using protection at first sexual encounters (and thereafter) and the ongoing need to negotiate and use condoms. In addition, the need for education in relation to contraceptive options and sexual and reproductive health rights was highlighted.

Data availability
These data are comprised of aggregated survey responses, data codebooks and the survey tool. These data are available from: Underlying data This project contains the following underlying data, found at: ", https://doi.org/10.

Consent
Written informed consent for publication of the learner's details obtained from the learners and their parents/guardian.

Megan A. Cohen
Emory University, Atlanta, Georgia, USA This is a cross-sectional baseline study on SRH knowledge and practices of grade 8 schoolchildren in South Africa. The information was gathered as part of a larger interventional study. Knowledge and practices are important topics and relevant to readers. However, several issues remain to be strengthened.
There is now too much information on the GAP Year intervention trial included such that it is confusing.
○ I do not see any information provided on knowledge of contraceptive methods. Was there any collected? I would definitely recommend including that if so.

○
The discussion still appears to state conclusions and recommendations that are more editorializing and are not directly derived from the results. I really do not see where this was amended per the recommendation of a prior reviewer, except to add information on where the GAP Year trial aimed to address issues.

○
For example, in the paragraph starting with "school teachers and mothers were noted as the most common source of SRH information," the authors then state that an important priority is "to equip parents with the knowledge and skills to communicate effectively with their children about sex," but no information in this study actually addressed whether the students felt that their parents lacked knowledge or that they felt they could not communicate with them; that is assumed by the authors.

○
As another example, the authors mention in the discussion the need to promote availability of other contraceptive options such as the contraceptive implant and the IUD, but there is no actually information presented regarding knowledge on contraceptive methods. Therefore, it is unclear if the participants were aware of the implant as an option but chose another method instead, or if they were unaware of the methods. ○ I have some concerns regarding discrepancies in the data reported in the tables, see below.

Methods:
Too much information is reported on the GAP Year trial which confuses what this study actually entails. I suggest editing information to a general, concise overview of the trial. Particularly I would remove information on randomization as that is irrelevant to this study. 1.
The sample size justification reflects that of the actual trial. I would clearly state that. Were the outcomes of this study powered for in any way?

2.
Results: Table 1-how does the 12-14 year old group have 14 people with sexual debut at 15-17? 1. Table 2 -footnote a for "SRH services required" states multiple choice question not asked of males, but there are data provided for the male participants. Please explain. It would also be helpful to provide the n of the sample asked if it differs from the n=2337 listed above, but again it is not clear if this is actually a different subsample since data are presented for male participants.

2.
Table 2 -What is the n of participants who were asked the question "Requested contraceptive services at last visit?" 3.
There is an "*" on "Main Source of SRHR information" that does not correspond to any footnote I can see. However there are no male data presented -perhaps that was supposed to correspond to the note for footnote a "multiple choice question, not asked of males?" 4.
And please clarify "multiple choice" -it appears that means "select all that apply" such that the values may surpass 100%. I believe the correct terminology should be "multiple response" or "multiple answer" instead of "multiple choice."

5.
Table 3 -please specify in the table that the question "contraceptive method ever used" is calculated based only on those who have used a contraceptive method as the denominator. For example I would consider putting the "n" in the table per age group in the same line as the heading, or list the total "n" before the footnote.

6.
Table 3, footnote a -again would recommend changing to "multiple response" or "multiple answer".

7.
Table 3 "Current contraceptive method used" -it looks like this is also a multiple response question? Please clarify and specify if so.

8.
Table 3 -Why is there no p-value listed for "know that condoms can prevent HIV and STIs"? 9.
Table 3 -The numbers are not adding up correctly for "if no, who made the decision for you?" States n=80 but 60 answers and 100% in each group chose parents. I think it is because there are 18 missing responses, but this is confusing. In this case would change the n to be n=60.

10.
Reviewer Report 25 May 2023 Section 2: Adjust "SRH and HIV". SRH includes HIV / AIDS / STI and contraceptive access and use. ○ Section 3: "the prevalence of mostly unintended pregnancies" Suggest reformulating, unclear what "mostly" means here, nor what data supports it. ○ Section 5: First sentence is circumvoluted, I suggest reformulating. ○ Table 1: Significant difference for currently dating / ever had sex by girls might be indicative of reporting bias and needs to be discussed in Limitations. Literature consistently shows that adolescent girls underreport sexual activity.

Table 2
Significant difference in "ever participated in an SRH program" for girls indicates that girls are much more frequently targeted by this kind of program and should be discussed in relevant section (esp. as authors advocate for gender-based interventions). It might also explain why girls are significantly more likely than boys to access youth clinics (because SRH programs may have made them aware of their availability). Table 2 presents knowledge of contraceptive services but not of contraceptive methods. Manuscript title, background and methods section should be adjusted to reflect this. Table 3: Is "Thigh sex" a traditional method? In that case it should be lumped with "Withdrawal" and other non-modern contraceptive methods.

Discussion:
I re-iterate the comment of a previous reviewer who found the discussion and conclusion too generic and not sufficiently aligned with the specific findings of the study.
The first section does not reflect the specific findings presented in the previous section. For example, looking at the different age groups (since this is an important point highlighted by the authors in the recommendations) Table 1 and 2 suggest that older adolescents are more likely to have been exposed to SRH programs (which makes sense considering they have been alive / in the target demographic for a longer period of time), yet they are less comfortable asking SRH questions from service providers. Regardless of these differences in SRH exposure though, Table 3 indicates that there are no significant differences in use patterns between age groups.
The authors do not discuss supply-side issues and barriers that may challenge their proposed recommendations. What factors could limit the feasibility of implementing those recommendations considering the resource-limited health provision environment of South African townships?
Section 3 and 4 of the Discussion presents the GAP Year as a solution to the issues highlighted in the manuscript which might read as confirmation bias, particularly as the study uses only pre-intervention data. We suggest removing or reformulating to present how GAP Year specifically addresses these issues.
The last three paragraphs of this section present conclusions that are not based on the study findings. For example, data about participants thinking that hormonal contraception can prevent STI / HIV is not mentioned in any of the previous sections.
It might be more appropriate to talk about "respondents" rather than "learners" presuming that not all students participated in the survey? (Participation bias).
The last section could also be construed as confirmation bias in advocating for ecological models such as the one used by GAP Year.

Strengths and limitations:
The sample of 26 schools was not selected in any representative way (or the randomization technique was not adequately presented in the Methods), therefore the results can NOT be generalized to other South African settings. (Large samples are not sufficient to generate external validity).

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Yes Are the conclusions drawn adequately supported by the results? No https://doi.org/10.21956/gatesopenres.14915.r32119 © 2022 MacPhail C. This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Catherine MacPhail
School of Health and Society, University of Wollongong, Wollongong, NSW, Australia This paper reports on the baseline findings from an intervention programme to increase contraceptive knowledge and use among young South African adolescents in three township communities.
Although the GAP Year intervention is included in the title of the paper, there is no real mention of the intervention in the paper. Would it be possible to include some details of the proposed intervention (likely completed already given the dates of this baseline data collection)? It seems that including some ideas of how the GAP Year intervention might address some of the issues outlined in the discussion would enhance the paper by reporting on how the findings might be addressed in the intervention.

Abstract:
Remove the apostrophe from Grade 8's ○

Introduction:
Suggest adding a comma to "there are still multiple challenges in accessible SRH service provision for adolescents, resulting in poor health outcomes." ○ "limited information on contraception knowledge" seems to be convoluted language and might benefit from some additional editing ○ Change "Knowledge about contraception and contraceptive options is low among South African youth,. . ."

Methods:
The recruitment of Grade 8 participants was undertaken across two school years. The authors might wish to comment on whether this was an intentional plan (and to what end) or to comment on any implications on this method in the limitations section.
○ It is interesting that sensitive questions about sexuality, gender, and norms were included in the interviewer-administered questionnaire component and not in the ACASI component. I assume that this was because these were general questions and not specifically about the participant's own identity?
○ Please comment on the use of a Xhosa translation for the W Cape schools but no equivalent translation for the Gauteng schools. What was the reasoning behind this decision? This should potentially be further discussed in the limitations section if you believe that this might have impacted on the accuracy of information or completion. It would be useful here to see completion by province -given that quite a large number of students did not complete both parts of the survey, might language problems have contributed in the ○ Gauteng cohort?

Results:
Table 2 results note that there is a significant difference in knowledge of right to accessing health care by gender (p<0.001), yet the reporting in the text says "Over two thirds of learners (66.8%, n=1557) perceived that they have a right to access health care without being discriminated or stigmatised by health workers: this was similar among males and females." This doesn't seem consistent.

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Yes

Are sufficient details of methods and analysis provided to allow replication by others? Yes
If applicable, is the statistical analysis and its interpretation appropriate? I cannot comment. A qualified statistician is required.
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
Competing Interests: I was employed by the Wits RHI from 2003-2012 and held an honorary position with this organisation after moving to Australia in 2012. I have not however worked with Melanie Pleaner or Mags Beksinska since before 2012. Previously working with the authors has not impacted my ability to provide an impartial review of this paper.
Reviewer Expertise: adolescent sexual and reproductive health, particularly HIV prevention I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
Author Response 01 Jul 2022

Alison Kutywayo
Thank you for your comments on our manuscript.
We have noted your comments and have addressed them as follows: Although the GAP Year intervention is included in the title of the paper, there is no real mention of the intervention in the paper. Would it be possible to include some details of the proposed intervention (likely completed already given the dates of this baseline data collection)? It seems that including some ideas of how the GAP Year intervention might address some of the issues outlined in the discussion would enhance the paper by reporting on how the findings might be addressed in the intervention. Response: Added a sentence in the Methods section to elaborate on the GAP Year intervention and then also mentioned the intervention in the Discussion, in line with this comment.
We have changed the text to make this more clear.
general observations and dissemination of knowledge. I suggest the authors to narrow down their recommendations and focus on what they found alone. Response: In addressing this comment, we have removed the general observations in the Discussion and Conclusion to ensure it is inline with our findings.