Sexual reproductive healthcare utilisation and HIV testing in an integrated adolescent youth centre clinic in Cape Town, South Africa

Background HIV prevalence is increasing among South African youth, but HIV counselling and testing (HCT) remains low. Adolescent pregnancy rates are also high. Objectives Innovative strategies are needed to increase HIV and pregnancy screening and prevention among youth. Method The Desmond Tutu HIV Foundation Youth Centre (DTHF-YC) offers integrated, incentivised sexual and reproductive health (SRH), educational and recreational programmes. We compared HCT and contraception rates between the DTHF-YC and a public clinic (PC) in Cape Town to estimate the impact of DTHF-YC on youth contraception and HCT utilisation. Results In 2015, females < 18 years had 3.74 times (confidence interval [CI]: 3.37–4.15) more contraception visits at DTHF-YC versus PC. There were no differences in the contraception and adherence was suboptimal. DTHF-YC youth (aged 15–24 years) were 1.85 times (CI: 1.69–2.01) more likely to undergo HCT versus PC, while male youth were 3.83 times (CI: 3.04–4.81) more likely to test at DTHF-YC. Youth were a third less likely to test HIV-positive at DTHF-YC versus PC. Female sex, older age, clinic attendance for contraception and sexually transmitted infections (STIs), redeeming incentives and high DTHF-YC attendance were all independent factors associated with increased HCT. Conclusion Youth were significantly more likely to access SRH services at DTHF-YC compared with the PC. The differences were greatest in contraception use by female adolescents < 18 years and HCT by male youth. Increased HCT did not increase youth HIV case detection. Data from DTHF-YC suggest that youth-friendly healthcare providers integrated into community youth spaces may increase youth HCT and contraception rates.


Introduction
South Africa has the highest prevalence of HIV infections among adolescents worldwide, accounting for nearly 18% of global HIV infections among 15-to 24-year-old youth in 2016. 1 Although HIV incidence is decreasing, South Africa still had 9.9 new infections per 1000 adults in 2016, with approximately 37% of those new infections in young people aged 15-24 years (22% in young women). 1 The HIV prevalence in a low socioeconomic, high-density township outside Cape Town was estimated to be 25% among residents ≥ 15 years of age in 2008 and 10.6% in 11-to 19-year-olds in 2006. 2,3 With respect to other sexually transmitted infections (STIs), a female adolescent cohort (16)(17)(18)(19)(20)(21)(22) years; mean 18 years) in Cape Town found that > 70% tested positive for at least one STI or bacterial vaginosis; Chlamydia trachomatis prevalence alone was 42%. 4 Survey data from four of nine South African provinces revealed that 19.2% of female adolescents (12-19 years) had been pregnant at least once 5 ; similarly, the 2008 Youth Risk Behaviour Survey showed that 24% of female students (11-20 years) reported at least one pregnancy (majority unplanned). 6 Despite the high prevalence of STIs and pregnancy, HIV testing, condom use and contraception coverage in South African youth remains suboptimal. A survey of 15-to 24-year-olds in KwaZulu-Natal found that only 29% of youth reported previous HIV counselling and testing (HCT), 7 while a population-based survey conducted in four South African provinces documented that less than half of 18-to 24-year-old women used hormonal contraception. 8 The same 2008 youth survey found that 30.7% of high school learners reported consistent condom use, while only 55% of students with STI symptoms had received treatment. 5 Nationally, only 45.8% of the 15-to 24-yearhttp://www.sajhivmed.org.za Open Access old South African youth in 2016 were able to correctly identify ways of preventing sexual transmission of HIV. 1 There remains a critical need to increase youth access to comprehensive SRH services in South Africa. Youth-friendly healthcare settings outside of traditional public clinics (PCs) may promote health-seeking behaviour and facilitate prevention and screening opportunities. To evaluate the impact of DTHF-YC on youth health-seeking behaviour, we compared HCT and contraceptive rates at DTHF-YC to those of youth at one of Cape Town's PC in a township of similar demographics and healthcare access. In order not to stigmatise the communities, the two township names are kept anonymous.

Study setting and population
The Desmond Tutu HIV Foundation Youth Centre and PC both serve isolated low-income, high-density townships situated next to wealthier suburbs of Cape Town. Both townships began as informal settlements but have grown to include mixed formal and informal housing. In both communities, over 90% of people identify as Black African, 80% live on < R3200/month and only about 25% of the residents live in formal dwellings. 10

Statistical analysis
In order to compare SRH utilisation at DTHF-YC versus the PC, HIV testing and contraception visit rates were calculated as the number of services provided (e.g. HIV tests and contraception visits) divided by the total number of youth estimated to be living in the two townships who could potentially access services. Two-way frequency tables and Chi-square tests were used to determine the effect DTHF-YC exposure had on HIV testing and contraception utilisation rates (95% confidence intervals [CIs]).
For DTHF-YC data only, each contraceptive visit was equated to 2-12 months of contraception coverage, depending on the method used. The total number of months of contraception coverage per woman per year was tabulated based on the number of visits and contraceptive type dispensed. Adherence was calculated as the number of months of coverage divided by 12 (perfect use was defined as continuous contraception through a single year).
For DTHF-YC data only, the association between HIV testing and DTHF-YC attendance with or without a clinic visit, incentives, contraception or STI treatment visits and demographic data was estimated using multivariable logistic regression. Age, DTHF-YC attendance and the number of incentive points redeemed for rewards were categorised as high or low based on the median. Quantitative data were analysed using STATA (Version 14, College Station, Texas, USA).

Ethical considerations
Ethical approval was received from the University of Cape

Significant predictors of Human immunodeficiency virus testing
The most significant predictor of HIV testing at DTHF-YC was obtaining STI treatment (Table 4). Symptomatic youth who received STI treatment were 2.69 times more likely to

Discussion
The DTHF-YC created an integrated health, educational and recreational programme in order to increase youth access to comprehensive SRH services. The PC had made a number of adolescent-friendly adaptations to increase youth utilisation of health services given the staffing, work hours and budget constraints of a public clinic. We compared the two clinics to test our hypothesis that exposure to an incentivised, integrated youth centre and clinic would increase youth utilisation of SRH services, with a primary focus on increasing access to HIV testing and contraception.
We demonstrated that youth adolescent healthcare utilisation was markedly higher at DTHF-YC in comparison with PC. Nearly four times more female adolescents under 18 years had contraception visits at DTHF-YC versus PC. Although DTHF-YC had more contraception visits, patients at both clinics opted for similar types of contraception (injectables). Similarly, implant and intra-uterine device use was low in both clinics. Intensive community outreach may be needed to increase youth interest in the more effective implant and intra-uterine contraception options.
Despite increased use of contraception at DTHF-YC, adherence was poor (average female yearly use at DTHF-YC was approximately 50%). Reasons given during informal discussions included forgotten appointments, too busy to return to clinic (despite 'adolescent-friendly' hours), travel outside the province or interruption of contraception between relationships. This contraception adherence pattern has implications for how pre-exposure prophylaxis (PrEP) might be used by an adolescent female population if it were readily available in the South African public sector alongside contraception as a part of a HIV prevention package. Given that PrEP requires up to seven days of daily dosing to reach adequate levels, it may be ineffective for adolescent females to cycle on and off PrEP as they do for contraception as most female patients only restarted contraception after a new relationship commenced. 14,15 Public health education strategies should engage adolescent girls about the benefits of using both contraception and PrEP continuously until they have a more prolonged period of either abstinence or monogamy to maximise prevention strategies.
The DTHF-YC model successfully increased the rate of youth HIV testing, particularly in male youth. Nearly twice the numbers of HIV tests were performed at DTHF-YC than PC, with nearly four times the number of tests in males. The Desmond Tutu HIV Foundation Youth Centre introduced the following youth-friendly services: (1) extended hours five days per week, (2) dedicated youth-friendly nurses with decreased wait times, (3) geographic separation from adult services, (4) close proximity to the high school, (5) a safe and fun space for youth to spend time, (6) free computer access, In addition, we suspect that DTHF-YC attracts an in-school youth population who may be less at risk than their out-ofschool peers. HIV testing at PC for healthy women is often linked to mandatory antenatal care, supporting the theory that healthy non-pregnant youth who choose to test might be at lower risk than the general population. All South African youth need access to friendly healthcare services. The Desmond Tutu HIV Foundation Youth Centre successfully increased HIV testing and contraception utilisation for sexually active youth who sought healthcare, irrespective of HIV risk. Nonetheless, additional community outreach and non-clinic-based strategies should be employed to reach the most vulnerable out-of-school or unemployed youth who may not attend youth centres or clinics, no matter how convenient.

Limitations of the study
This study has several limitations. Because PC data were only available at the unit of services provided (HIV test, family planning visit), a comparison at the individual level was not possible. As a result, the width of our confidence intervals may be underestimated owing to our inability to account for clustering. Nonetheless, since tested individuals at DTHF-YC had a median of one HIV test per year, we suspect that clustering would have had minimal impact on the HIV testing data. There was more than one contraception visit per person per year. However, given the high degree of statistical significance of our findings, it seems unlikely that clustering would have qualitatively changed our contraception results. Finally, as HIV testing is not rare, the reported odds ratios might actually present a less accurate approximation of the risk ratio of the tested independent variables.

Conclusion
In conclusion, our data suggest that convenient, confidential, youth-friendly SRH services associated with youth social spaces and activities can increase healthcare utilisation, specifically contraception and HCT. Innovative strategies such as community or school-based outreach programmes that include contraception, HCT and SRH services, including PrEP, are needed to blend the success of the DTHF-YC with existing public facilities and healthcare workers.