"SKILLZ," is a mixed methods evaluation of the Grassroots Soccer (GRS) SKILLZ Package based in Lusaka, Zambia. The package is made up of three football-based programs: (1) SKILLZ-Girl - implemented in schools as part of a 10-week program culminating in a tournament event; (2) SKILLZ-Club - implemented as an ongoing extra-curricular activity after the completion of SKILLZ-Girl; (3) SKILLZ-Plus - a clinic based football group targeted at girls that are HIV-positive.

The programs work together to build a continued support system which encourages uptake of Sexual Reproductive Health (SRH) and HIV services, while facilitating ART adherence (for HV-positive participants) and continued engagement with health services over the long-term (whether to contraceptive methods, HIV prevention services, HIV repeat testing, and/or HIV treatment and care).

The study team has developed an enhanced SKILLZ-Girl offering, which will include a comprehensive module on HIVST, contraceptives and PrEP, access to a nurse during the implementation of sessions and the additional offering of HIVST and contraceptive services at the event along with ongoing engagement through the SKILLZ-Club program (Enhanced Arm) , The central hypothesis is that this enhanced curriculum will increase HIV testing and contraceptive uptake compared to the standard SKILLZ curriculum & standard event (SOC Arm). The investigators further hypothesize that the intervention in the enhanced arm will positively and directly affect a number of mediating factors including attendance at soccer events where community-based SRH services are offered, SRH knowledge, empowerment, self-confidence, and perceptions of gender balance, and (reduced) stigma.

For girls found to be HIV-positive, the follow-on SKILLZ intervention (SKILLZ-Plus) has been designed to facilitate linkage to HIV care and treatment, reduce HIV-related stigma, increase disclosure to family and partners, increase feelings of social support, empowerment, self-efficacy, and ultimately adherence to ARVs, viral load suppression (VLS) and retention in HIV care and treatment.

This study will be conducted in up to 32 secondary schools that GRS currently serves in the Lusaka Urban District.

Research aim(s)/General Objective and Specific Objectives Aim 1: (a) Assess the impact of SKILLZ over two years on HIV testing and (b) SRH-related prevention services uptake among adolescent girls aged 16-19 in up to 23 intervention schools (with enhanced HIVST, PrEP, and contraceptive method offerings) compared to up to 23 control schools using a quasi-experimental difference-in-difference approach; and

Aim 1: (b) Describe linkage to care and treatment, and viral load suppression and retention at 6, 12, and 18 months for girls who are identified to be HIV+ during the study, using qualitative, programmatic, and administrative data.

This will be achieved by i) using a quasi-experimental difference-in-difference approach that compares girls aged 16 and over in approximately 23 SKILLZ control schools compared to approximately 23 schools where SKILLZ will be delivered with the enhanced HIVST and a variety of contraceptive method offerings; and ii) through a cohort model using pre-existing SmartCare data on HIV+ individuals in care to match to adolescent girls aged 16 and over from the same district (Lusaka) who are identified to be HIV+ at any time during SKILLZ-Girl or SKILLZ-Club participation.

Aim 2: Examine how the intervention works including lessons learned for future implementation by

1. conducting a process evaluation to identify mediators, predictors, and barriers to uptake of the SKILLZ-Girl, Club, and Plus curricula both quantitatively through mediation and moderation analyses and qualitatively through a sequential explanatory approach using focus groups discussions, interviews and observation with coaches and girls; and by
2. monitoring fidelity

Aim 3: Estimate the short- and long-term cost-effectiveness and return on investment of the SKILLZ Girl, Club, and Plus curricula for improving health outcomes for adolescent girls.

This is a mixed-methods evaluation of the SKILLZ intervention using the following approaches:

A quasi-experimental cohort of school-aged 16-year-old and over females to be followed for a period of 18 month will be evaluated with a difference-in-difference (DID) approach to estimate HIV testing uptake and sexual/reproductive health services use. Primary and secondary outcomes will be measured at several timepoints longitudinally across up to 23 schools offering the SKILLZ-Girl and SKILLZ-Clubs program and up to 23 schools offering the "standard of care" (SOC) and SOC Clubs. Recruitment is anticipated to take 6 months on a rolling basis. The order in which schools will be selected will be random. Participants will be clustered at the school-level; a random sample of girls will be enrolled from each participating school, aged 16-years at the commencement of SKILLZ-Girl and who are in a grade where GRS is offering the SKILLZ-Girl curriculum All enrolled girls will be followed over the 18 months' time period to assess SRH service uptake, including HIV testing. It is estimated that the majority (~98%) of girls testing for HIV will likely be HIV-negative, based on estimates from programmatic GRS data. While HIV-positive girls will continue to participate in the school-based SKILLZ-Girl and Clubs activities and followed for SRH use, they will also be offered contemporaneous enrolment into SKILLZ-Plus upon infection; retention in care and Viral Load Suppression (VLS) at 6, 12, and 18 months will be measured for this subset.

Process evaluation: A process evaluation will be conducted to better understand the casual pathways and effect mediators and moderators using mixed methods approaches, including supplemental qualitative data collection. To investigate reasons why the program components (SKILLZ-Girl, Club & Plus) might fail or succeed in different schools or communities and assess intervention effect mediators and moderators for HIV testing and contraceptive use uptake we will use a mixed methods approach involving: (1) quantitative analyses of baseline survey data for testing how variables can moderate the observed impact of the intervention, and (2) qualitative data collected from a subset of SKILLZ participants in the evaluation cohort and their coaches to further understand potential mediating pathways.

Fidelity monitoring: The study includes measures to monitor fidelity of the intervention implementation, including receptivity to and understanding of the intervention amongst participants, the perceived appropriateness and relevancy of the intervention for adolescent girls in school, and the extent to which the intervention was delivered per protocol in different communities and schools.

Economic evaluation: Time permitting (given the uncertainty of delays due to COVID), the study will estimate the short- and long-term cost-effectiveness and return on investment of SKILLZ for improving health outcomes for adolescent girls.

Data collection plan and tools

Survey at baseline: A baseline survey will be administered to approximately 50 girls per school randomly selected to participate in the evaluation to collect information on school characteristics, sexual behaviour, age, HIV knowledge, maternal education, baseline HIV status, and sexual/reproductive health uptake and retention. These variables will be used as covariates and moderators for service uptake and retention. The target sample of approximately 1380 girls will respond to a survey self-administered using ODk software upon informed consent. The survey will be available in English, Nyanja and Bemba. Coaches and study staff will on site be help girls navigate the survey. If during baseline, it appears that there has been sufficient time to measure change, or the full SKILLZ-Girl curriculum was not administered, the baseline survey may be re-administered to assist with validation of responses (as opposed to implementing the follow-up survey provided below).

Survey post-SKILLZ-Girl event, 6-, 12-, and 18 months: A total of approximately 30 girls per school (regardless of HIV status) will respond to a ODK survey to collect information on sexual behaviour, HIV knowledge, maternal education, sexual/reproductive health uptake and retention. Post-SKILLZ-Girl survey data will be used to assess the effect of SKILLS Girl and SKILLZ-Club on the outcomes of interest. In schools selected to offer SKILLZ-Club, surveys will be conducted during club attendance approximately 6 months after the SKILLZ-Girl community soccer event. Surveys will be completed individually by the participant on a secure tablet physically or via phone. Study staff will be available to assist participants to navigate the survey if they request assistance. If a participant has opted to not attend the clubs or has since left or changed schools, the coaches and study staff will attempt to trace participants in the community in order to complete the follow-up survey.

In the schools selected not to offer SKILLZ-Club, GRS will arrange for a 2-hour meeting session with all participants still enrolled at that school, where all participants will complete the surveys. If a participant opts to not attend the meeting or has since left or changed schools, the coaches and study staff will attempt to trace participants in the community in order to complete the follow-up survey.

Data management and storage

Quantitative:

GRS will collect programmatic data through a combination of paper and electronic data entry via tablets, with all data stored securely in RedCap. Where paper files are used for programmatic management, data will be manually entered by study staff directly into the RedCap database. De-identified data will made available to GRS co-investigators to enable review of key fidelity and quality indicators in order to monitor implementation in real time.

All paper study files will be stored in secured, locked cabinets located in locked rooms available to study staff only. A secure server will be used to store encrypted study data, including the study database. All personal identifiers will be removed prior to generating the analytical dataset.

Qualitative:

All data will be obtained using digital audio recorders. Recorded data will be transferred on to a lockable computer kept by the co-principal investigator leading the qualitative research. Transcripts will go through a QA/QC process where researchers will cross check randomly selected transcribed verbatim with the recordings.

Data analysis plan The primary outcome is the probability of testing for HIV at any time during the period when SKILLZ is delivered through to 24 months after the final soccer event. The primary analysis for Aim 1 will follow a difference-in-difference (DID) effect between study arms. The outcome of interest will be regressed on an indicator of study arm assignment interacted with a dichotomous variable POST which indicates the outcome after the intervention is implemented; this differentiates baseline and end-line for each study arm, controls for any time dependent variables that might confound the effect, as well as any time-invariant observed or unobserved confounders that might differ between study arm schools. Clustering by school will be accounted for by a generalized estimating equation approach. Intention-to-treat (ITT) will be evaluated by including all girls in the evaluation cohort assigned to each arm rather than all girls who actually participated in SKILLZ.

Inclusion Criteria:

• The student is >16 years of age prior to the first session of SKILLZ-Girl or the student will turn 16 years of age before the last session of SKILLZ-Girl (ie prior to the event)
• Is enrolled in one of the SKILLZ STUDY schools
• Is able to converse in and comprehend either Nyanja, English and/or Bemba
• The Parent/Guardian of the participant has consented for the participant to take part in the study

Exclusion Criteria:

• The student has a mental disability that would interfere with their ability to give informed consent

De-identified IPD will be shared via secure server to co-investigators

Data will be shared with requisite co-investigators on a rolling basis in order to conduct interim analysis

Must be a co-investigator

1 year

This was a school study that was conducted during the COVID-19 period and schools were closed for protracted periods of time. In addition, community mistrust around COVID-19 and COVID vaccinations led to large numbers of pupils not returning to school once schools were re-opened. This limited our ability to follow participants longitudinally. General distrust of the medical system made girls resistant to sharing clinical data around HIV.