The impact of community-based integrated HIV and sexual and reproductive health services for youth on population-level HIV viral load and sexually transmitted infections in Zimbabwe: protocol for the CHIEDZA cluster-randomised trial

Background Youth have poorer HIV-related outcomes when compared to other age-groups. We describe the protocol for a cluster randomised trial (CRT) to evaluate the effectiveness of community-based, integrated HIV and sexual and reproductive health services for youth on HIV outcomes. Protocol The CHIEDZA trial is being conducted in three provinces in Zimbabwe, each with eight geographically demarcated areas (clusters) (total 24 clusters) randomised 1:1 to standard of care (existing health services) or to the intervention. The intervention comprises community-based delivery of HIV services including testing, antiretroviral therapy, treatment monitoring and adherence support as well as family planning, syndromic management of sexually transmitted infections (STIs), menstrual health management, condoms and HIV prevention and general health counselling. Youth aged 16–24 years living within intervention clusters are eligible to access CHIEDZA services. A CRT of STI screening (chlamydia, gonorrhoea and trichomoniasis) is nested in two provinces (16 of 24 clusters). The intervention is delivered over a 30-month period by a multidisciplinary team trained and configured to provide high-quality, youth friendly services. Outcomes will be ascertained through a population-based survey of 18–24-year-olds. The primary outcome is HIV viral load <1000 copies/ml in those living with HIV and proportion who test positive for STIs (for the nested trial). A detailed process and cost evaluation of the trial will be conducted. Ethics and Dissemination The trial protocol was approved by the Medical Research Council of Zimbabwe, the Biomedical Research and Training Institute Institutional Review Board and the London School of Hygiene & Tropical Medicine Research Ethics Committee. Results will be submitted to open-access peer-reviewed journals, presented at academic meetings and shared with participating communities and with national and international policy-making bodies. Trial Registration https://clinicaltrials.gov/: NCT03719521

HIV and sexual and reproductive health services for youth on HIV outcomes.

Protocol
The CHIEDZA trial is being conducted in three provinces in Zimbabwe, each with eight geographically demarcated areas (clusters) (total 24 clusters) randomised 1:1 to standard of care (existing health services) or to the intervention.The intervention comprises community-based delivery of HIV services including testing, antiretroviral therapy, treatment monitoring and adherence support as well as family planning, syndromic management of sexually transmitted infections (STIs), menstrual health management, condoms and HIV prevention and general health counselling.Youth aged 16-24 years living within intervention clusters are eligible to access CHIEDZA services.A CRT of STI screening (chlamydia, gonorrhoea and trichomoniasis) is nested in two provinces (16 of 24 clusters).The intervention is delivered over a 30-month period by a multidisciplinary team trained and configured to provide high-quality, youth friendly services.
Outcomes will be ascertained through a population-based survey of 18-24-year-olds.The primary outcome is HIV viral load <1000 copies/ml in those living with HIV and proportion who test positive for STIs (for the nested trial).A detailed process and cost evaluation of the trial will be conducted.

Ethics and Dissemination
The trial protocol was approved by the Medical Research Council of Zimbabwe, the Biomedical Research and Training Institute Institutional Review Board and the London School of Hygiene & Tropical Medicine Research Ethics Committee.Results will be submitted to open-access peer-reviewed journals, presented at academic meetings and shared with participating communities and with national and international policy-making bodies.

Introduction
Two-thirds of people living with HIV reside in sub-Saharan Africa (SSA).While there has been a general global decline in new HIV infections, these are suboptimal, and this has been much less marked in youth.In 2019, 30% of people newly infected with HIV in eastern and southern Africa were adolescent girls and young women aged 15-24 years 1 .Compared to other age-groups, youth living with HIV are less likely to be diagnosed and those diagnosed have lower rates of HIV viral suppression once they start antiretroviral therapy (ART) 2 .HIV testing is a prerequisite for accessing care or prevention services, yet population-based survey data from 2015-2017 in Zimbabwe, Malawi and Zambia showed only 52%, 48% and 45% of youth aged 15-24 years reported ever having an HIV test, respectively [3][4][5] .In these countries, it was estimated that only 40-50% of 15-24-year-olds living with HIV were aware of their HIV status compared with 66-73% in all age groups combined (>15 years).Similarly, HIV viral load suppression among those who reported being on ART ranged from 71-85% among those aged 15-24 but 87-91% among those aged >15 years [3][4][5] .Viral non-suppression is associated with morbidity and with increased risk of onward HIV transmission 6 .
Youth face unique personal, social, legal, and structural barriers to access HIV services, including judgemental provider attitudes, intense stigma, and families being "gatekeepers" to accessing healthcare 7,8 .HIV testing is often not a priority, particularly given the barriers to access.
Worldwide, there remains a large unmet need for sexual and reproductive health (SRH) services among youth, including those who are living with HIV 9,10 .The prevalence of curable sexually transmitted infections (STIs) (chlamydia, gonorrhoea, and trichomoniasis) among youth remains very high, and the current approaches for STI management in low-and middleincome settings, i.e., syndromic management has been ineffective in controlling STIs 11,12 .The main limitation of syndromic management is that asymptomatic infections, which are the majority, are not treated 13 .Untreated STIs result in considerable morbidity, including ascending infections, infertility, chronic pelvic pain, and poor birth outcomes.At a public health level, STIs facilitate HIV transmission 14 .
In recent years, rapid molecular tests that do not require sophisticated laboratory infrastructure have become available, raising the possibility offering screening for STIs in high risk groups such as youth to identify asymptomatic infections 15 .This study design is influenced by evidence provided by previous work which can seen here: https://doi.org/10.12688/wellcomeopenres.17531.1 16 .

Study rationale
Youth are a priority group for HIV interventions 17 , but existing HIV programmes which are vertical and largely facility-based, have not been successful in engaging youth.We hypothesise that services which address the entire HIV cascade (HIV testing, linkage to care and treatment, ART initiation and support to maintain viral suppression) and are community-based may address some of the barriers to access and improve HIV outcomes.Integrating HIV services with SRH services may better address the needs of youth and are likely to lead to higher acceptability and programmatic efficiency 18 .An important barrier to youth accessing services is healthcare provider attitudes and therefore training and mentorship of healthcare providers to provide youth friendly services is critical to ensure engagement of youth 18,19 .
The Community based interventions to improve HIV outcomes in youth: a cluster randomised trial in Zimbabwe (CHIEDZA trial) investigates whether providing an integrated package of HIV and SRH services in community-based settings by trained providers to youth aged 16 -24 years improves population level HIV viral suppression.The trial embeds another CRT aiming to investigate whether screening for STIs including offering testing, treatment, counselling, and partner treatment (comprehensive management) for chlamydia, gonorrhoea and trichomoniasis reduces population level prevalence of these STIs among youth.

Study objectives.
The aim of this trial is to estimate the impact of a comprehensive community-based package of HIV services, integrated with SRH services and general health counselling for youth aged 16-24 years, on population-level HIV viral load.
The objectives are: 1.To investigate the impact of the intervention on each step of the HIV care cascade (HIV testing, treatment initiation and HIV viral load suppression) among youth at population level 20 .

2.
To investigate the impact of providing STI screening (chlamydia, gonorrhoea and trichomoniasis) and comprehensive management for those testing positive on populationlevel prevalence of these STIs among youth.
3. To measure the uptake of each component of the intervention.

4.
To estimate the cost and cost-effectiveness of the intervention.

5.
To assess the fidelity, feasibility, acceptability, and quality of the intervention as delivered in order to identify likely mechanisms of action for observed outcomes and to inform the factors required for scalability and sustainability.

6.
To estimate population-level prevalence of important health-related risk factors and behaviours among youth.

Study design
CHIEDZA is a two-arm cluster randomised trial (CRT) (Figure 1).The standard of care (SOC) is defined as existing

Amendments from Version 1
This revised version of our manuscript does not include any major revisions, however, provided additional clarification of terms used in the first version of manuscript.This includes additional descriptions of staff roles and the set up within control clusters of the trial.
Any further responses from the reviewers can be found at the end of the article Figure 1.CHIEDZA trial design.
HIV and SRH services, which are largely facility-based, siloed and verticalized.For example, within a primary care clinic HIV treatment would be offered by one provider and other services such as family planning would be offered in another part of the clinic by another provider.The standard of care would also include ad-hoc community-based services from various providers, however, no CHIEDZA intervention will be delivered within the control clusters.
A further CRT is embedded within the parent trial to investigate the impact of providing STI screening and comprehensive STI management integrated within the CHIEDZA intervention on population-level prevalence of STIs among youth in two of three CHIEDZA provinces.

Study setting and population
The trial is being conducted in Zimbabwe, which had an adult (>15 years) HIV prevalence of 12.8% in 2019 1 .The trial is being conducted in three provinces, Harare, Bulawayo, and Mashonaland East.Harare is the capital and largest city in Zimbabwe and the population is predominantly of Shona ethnicity; Bulawayo, is the second largest city in the country, situated 437kms from Harare, and is predominantly Ndebele.Mashonaland East province borders Harare and peri-urban settings in this province were selected.In combination, these provinces represent both main ethnic groups in Zimbabwe.Urban and peri-urban settings were selected as population densities in rural areas are low, making the trial unfeasible.
Definition of study clusters.We defined a cluster as a geographically demarcated area with an estimated population of ~2000-4000 youth aged 16-24 years (based on Zimbabwe 2012 Census estimates) that contained a community centre or hall from which the intervention could be delivered 21 .A cluster had to be serviced by a defined primary care clinic that was not serving another study cluster and was situated within the cluster to ensure integration and collaboration with public-sector services.
Spatial datasets were provided for the study areas by the Zimbabwe Electoral Commission.These included electoral wards (with population numbers), health centres and schools.These data and local field knowledge were used to define a list of possible locations for clusters.These were examined in more detail, overlaying ward populations, clinic and school locations on top of OpenStreetMap data and satellite imagery to show roads, rivers, and buildings 22 .The clusters were defined by first using electoral ward boundaries to identify areas within the required population range.These areas were then modified to take into account natural breaks e.g. a river or a major road to provide sensible cluster boundaries.Where possible, natural boundaries such as green space or industrial areas were used to form the edge of the cluster to minimise contamination.Satellite imagery and the buildings from OpenStreetMap were used to define an area that was estimated to contain between 2000-4000 youth.Once all the clusters were defined the boundaries and their clinics were exported to Google Earth (.kml format), to enable field teams to check the clinic locations and map key points of interest (POIs).Study staff walked around the cluster, marking any POI onto a printout of the OpenStreetMap data while at the same time storing the location and details as a bookmark in a navigation application (MAPSME) 23 .The electronic locations were validated against the paper version locations.
Randomisation.A total of 24 clusters, stratified by province, were randomised in a 1:1 allocation ratio to either the control arm (existing services) or the intervention arm (defined below), so that each province had four intervention and four control clusters (Figure 2).Intervention.The intervention was designed following formative research that included a literature review, qualitative interviews with relevant stakeholders (n=90) and participatory workshops with youth (n=2) recruited from the study communities.Considering youth opinions and the formative research findings, the research team developed an outline intervention.This outline was presented to youth in participatory workshops, and consensus achieved on the intervention's content and configuration including the types of services, location of service delivery, types of service providers and the "branding" of the service (Table 1) (formative work manuscript submitted with protocol manuscript 24 . The intervention is delivered at multi-purpose community centres or halls within the cluster over a 30-month period.An extension of six months beyond the originally planned 24 months was added to mitigate against the effects of the COVID-19-related lockdowns.Services are provided one day a week (same day every week) in each cluster throughout the intervention period (except for public holidays).The deployment of the intervention across the provinces occurred in phases (Figure 3).
One intervention team per province is responsible for delivering the intervention to all four intervention clusters.Each team consists of two nurses, four community health workers (CHWs), one counsellor and two youth workers.Youth workers are young people aged below 24 years at the time of recruitment who supported trial activities by welcoming young people to the CHIEDZA sites, screening for eligibility, and responsible for running activities and information sessions.
Although the intervention teams are affiliated with the trial, their role is primarily to deliver what is recognized by the World Health Organisation (WHO) as a 'best practice' public health interventions 25 .Hence, the intervention teams are regarded as separate from the 'research teams' and the norms and standards governing their activities are those accepted for the implementation of public health interventions rather than those applied to conventional clinical research projects.
The intervention was specifically configured to be "youth friendly" i.e. able to effectively attract youth, meet their needs responsively and retain them in care, as shown in Table 1.In addition, there was a specific focus on intervention providers; the intervention team was selected based on prior experience of working in communities and with youth.A training programme on each of the intervention components was combined with training on provision of youth friendly services, particularly focusing on communication and counselling that is appropriate to age and maturity, and attitudinal training specifically emphasizing respect, confidentiality, non-judgement, and relatability.Ongoing 1-2 weekly debrief meetings that incorporate problem-solving, discussion of complex cases and operational issues ensure that intervention providers are supervised and mentored.
The content of and rationale for the choice of the intervention components is summarised in Table 2.A structured Manual of Operations details intervention activities and guides service delivery (Appendix i).This is updated if any operational procedures are adapted.
HIV testing and care services.HIV testing is conducted according to national guidelines.A mobile-based application (ITHAKA) was used to support HIV self-testing for the first four months of the intervention but discontinued due to logistical challenges.Those who test HIV-positive are offered a choice of being referred to a HIV care clinic of their choice (accompanied by a CHIEDZA team member to the clinic to help facilitate linkage to care) or of accessing care through CHIEDZA.If the latter is selected, the client's HIV records are maintained at the clinic and clinic data updated by CHIEDZA  Table 1.Configuration of the CHIEDZA intervention to provide Youth-Friendly Services.

Environment
• Convenient location • Adequate and attractive space where youth can socialise • Comfortable surroundings • Female and male service providers (to avoid a "female-centred" environment) • Providers known by first names • Encouragement of youth-focused "non-health" activities e.g., music, dancing, darts, pool, soccer, film evenings • Outreach -sensitisation of youth and community gatekeepers

Confidentiality and privacy
• Multipurpose use venue (so clients not marked out) • Visual and auditory privacy • Names and identifying information not recorded a,b

Information
• Paper and electronic IEC materials (posters, videos, manuals) appropriate for various age and literacy levels • Bouquet of SMS messages Flexibility • Services will be provided from 11am to 7pm and at weekends (e.g., CAPS groups) • Drop-in service delivery (no appointments)

Youth as equal partners
• Youth involved in design of space and intervention content and delivery • Youth involved in selection of providers (on interview panel) • Youth Advisory Board for feedback • Youth involved in mobilization and sensitization activities

Appropriate, quality and affordable services
• Wide range of services available (see above) • Choice of products where relevant e.g., menstrual management products, contraceptives, • All services and commodities free of charge • High quality products • Trained and mentored providers • System to "red-flag" youth with concerns Youth living with HIV (regardless of whether they access HIV care through CHIEDZA or not) are offered free HIV viral load

Mobile based interventions.
From 18 January 2021, all CHIEDZA clients have been offered the option of registering to receive a series of short evidence-based SMS messages.The messages were developed 34 and evaluated [35][36][37] elsewhere and adapted for the Zimbabwean context though an iterative process with CHIEDZA clients 38 .The 97 short messages for female recipients (94 for male), delivered over three months and offered in English, Shona and Ndebele contain accurate information about contraception and condom use and promote positive behaviour change 39 .
Community sensitisation and mobilisation.Service delivery is accompanied by peer outreach to promote CHIEDZA and engage youth to achieve high coverage.Outreach teams consist of 16-24-year-old cluster residents who had previously engaged with CHIEDZA.Following engagement training and social mapping of clusters (which highlighted prime community sensitization locations or "hotspots"), the teams focus their activities around these locations.These include shopping centres, bars, sports centres, boreholes, and community snooker tables.Flyer distribution, information dissemination, and in-field live demonstrations of CHIEDZA products (such as reusable pads, menstrual cups, and condoms) are conducted to educate, generate support and strengthen community engagement.
Sensitisation is also conducted door-to-door within the cluster.
Mobilisers also escort eligible youth to CHIEDZA centres where requested.During the walk to the centre, mobilisers answer questions and assuage fears.Ad hoc talent shows, modelling, music, drama, and dance competitions are held at community centres to increase their engagement with the intervention.Strategies accommodate the different seasons, school terms and COVID lockdown levels.
Study outcomes.Trial outcomes will be measured at population level.The primary outcome of the CHIEDZA trial is the proportion of survey participants living with HIV who are virally suppressed (defined as having an HIV viral load <1000 copies/ml).
The primary outcome is the product of the following three secondary outcomes among participants living with HIV: 1. Proportion who are aware of their HIV status 2. Proportion of those aware of their HIV positive status who are taking ART

Proportion of those on ART who are virally suppressed
Other outcomes among survey participants (regardless of HIV status) are: 1. Proportion with knowledge of their HIV status 2. Proportion of females using modern methods of contraception 3. Proportion of females reporting adequate menstrual management 40 The primary outcome of the nested STI trial is the proportion of survey participants who test positive for either NG, CT or TV, ascertained in a randomly selected proportion of participants in the 16 nested STI trial clusters.The secondary outcomes were prevalence of the individual STIs -NG, CT, and TV.
The population-based survey provides an opportunity to understand the prevalence of important health-related risk behaviours and morbidities.These will not be compared by trial arm, and include: 1. Proportion with symptoms of a common mental disorder (measured using the Shona Symptom Questionnaire-14 with a cut-off of 8) 41 2. Proportions who report tobacco, alcohol, and substance use 3. Sexual behaviour (including no. of sexual partners in the past year and condom use) 4. Proportion circumcised (men only) 5. Proportion who have high blood pressure (Systolic >140mmHg and/or diastolic >90mmHg) Ascertainment of study outcomes: endline survey Study outcomes will be assessed using a population-based cross-sectional survey conducted among 18-24-year-olds living in the study clusters at the end of the intervention period (30 months).The age group was selected to ensure maximum exposure to the intervention.

Sampling strategy.
The sampling methodology for the outcome survey combines remote selection methodologies incorporating satellite imagery and traditional random street selection 42 .The OpenStreetMap road network is checked against satellite imagery to ensure all streets were mapped.All streets within a cluster are manually split into segments within the GIS software (ArcGIS version 10.5) 43 either using junctions or features such as school grounds.Segments of streets with no buildings, streets that formed the boundary of the cluster and streets on which the main entrance to buildings is from a different segment are excluded.The average length of segments varies slightly between clusters depending on the density of buildings and streets, ranging from between 100-300m.Each segment is assigned a number and segments are randomly selected.All dwellings in the selected segment will be included in the survey.We estimate between 50-70% of the segments in each cluster will be sampled to reach the required sample size.Maps showing the selected segments and a table with GPS coordinates for the segments will be provided to the survey teams.

Survey procedures.
Research teams will visit survey communities before implementation of the survey to sensitize residents about the trial.Following community sensitisation, all households (defined as a person or group of related or unrelated persons who live together in the same dwelling or unit(s) of a dwelling, who acknowledge one adult male or female as head of the household, who share the same housekeeping arrangements, and who are considered a single unit) in each dwelling in the selected street segments will be enumerated.All individuals aged 18-24 years residing in the enumerated households will be eligible to participate.If a potentially eligible individual is not available at the time of enumeration, up to three repeat visits will be made to enrol the individual.
Following consent, a fingerprint will be collected (see data collection and management section for details) and an intervieweradministered questionnaire will record sociodemographic data, duration of residence and exposure to the intervention.Participants will be asked about knowledge of HIV status, history of HIV testing and care, sexual behaviour, parity, contraceptive use, and menstrual management (females only) and VMMC (males only).In addition, data on smoking and other substance use, mental health, COVID-19 and access to digital technology will be collected.Height, weight, and blood pressure will be measured, and a dried blood spot sample will be collected for anonymised HIV antibody testing and (for those who are HIV antibody positive) HIV viral load testing.
Survey days will be randomised as "STI trial days" for ascertainment of the nested STI trial outcomes.Any participant recruited on a STI day will have a urine sample collected (with additional consent) for STI testing (Trichomonas Vaginalis, Neisseria gonorrhoeae and Chlamydia trachomatis) using the Seegene multiplex real-time PCR (Anyplex II STI Essential Assay, Seegene Inc, Seoul, Korea).Those who test positive for an STI will be followed up for comprehensive STI management.Follow up will be done via telephone (up to three phone calls within two months) and treatment provided for free to trial participants and their sexual partners.
Participants identified with "red flag" conditions (e.g.severe hypertension, experience of violence, severe mental health disorders, alcohol, and drug use disorders) will be counselled and referred to the appropriate health service providers for care.
The survey will occur in phases (by province) following the phased implementation of the intervention (Figure 3).

Sample size and statistical analysis
The prevalence survey will recruit 700 youth in each cluster (16 800 total) including 300 within each Harare and Bulawayo cluster (4 800 total) for STI testing.The proportion of youth living with HIV who are virally suppressed (primary outcome), is expected to be 43% in the control arm (60% diagnosed × 84% linked to care × 85% virally suppressed), based on ZIMPHIA estimates 5 .The community prevalence of HIV in the 18-24-year-old population is estimated to be 3%.
With a coefficient of variation of 0.25, the study will have 80% power to detect a difference of 19% (i.e.62% prevalence of viral suppression in the intervention arm) and 90% power to detect a difference of 23% (66% prevalence of viral suppression) 44 .A 66% prevalence of the primary outcome could be achieved by, for example, reaching 80% diagnosis, 91% linkage to care and 91% viral suppression.With a coefficient of variation of 0.3, the study has 80% power to detect a difference of 23% and 90% power to detect a difference of 27%.
For the nested STI trial, with 16 clusters, a sample of 300 youth in each cluster for STI testing will have 90% power to detect an STI prevalence in the intervention arm of 10% compared to 17% in the control arm, with a coefficient of variation of 0.25.
CONSORT guidelines for analysis of CRTs will be followed with CONSERVE guidelines followed to report the trial modifications made as a result of the COVID-19 pandemic 45 .A flowchart will be created showing the number of clusters eligible, the number allocated to each arm and the number of participants per cluster in the cross-sectional survey.Cluster-level analyses will be used to adjust for between-cluster variability, as recommended for trials with fewer than 15 clusters/arm 44 .Descriptive analysis will be used to compare cluster-level characteristics of the two arms, with adjustment for variables that are unbalanced between arms (avoiding variables likely to be affected by the intervention) and for stratum.
All outcomes are binary.For each outcome, the risk for each cluster will be calculated, and shown by strata and arm.The mean and SD of the log risk will be used to estimate the geometric mean and associated 95% CI for each arm of the study.Linear regression of the log mean risk on strata and arm will be used to estimate the risk ratio and 95% CI.The approximate variance for the mean risks will be obtained based on the residual mean square from a two-way ANOVA on arm and strata.A 95% CI for this will be calculated from the variance using a t-statistic.

Data collection and management
For the intervention, each client is registered using a fingerprint using SIMPRINTS software which converts a fingerprint into a Global Unique ID (GUID) (Simprints Technology, Cambridge, UK).SIMPRINTS software is integrated with an Android-based data capture application (Survey CTO Collect) (SurveyCTO, Massachusetts, USA) which is used to record the services accessed by clients on each visit.At the first visit age, sex, date of birth and initials are recorded along with the fingerprint but no identifying information such as name or address is collected.At subsequent visits, the client is identified by fingerprint, which enables providers to check whether the client has accessed the service previously.This allows every client visit and service accessed to be tracked across the intervention period.If biometric identification is not possible or clients decline to provide fingerprints a backup registration process consisting of a unique ID number is allocated to the client.For clients who test HIV-positive, identifying data including name and address and contact details are collected to facilitate linkage to care and register them into the national HIV programme.Details of HIV care including ART prescribed, monitoring test results, and any incident clinical events are recorded on national HIV programme patient records.Data are imported into an ACCESS database for cleaning and quality controlled using automated real-time quality checks.
Endline survey data will be collected on electronic tablets using the SURVEYCTO platform.Biometric records obtained from participants in the cross-sectional survey will be matched against biometric records of clients accessing the intervention to assess degree of contamination (the proportion of control cluster survey participants who accessed the intervention) and coverage of the intervention (proportion of intervention cluster survey participants who accessed the intervention).

Process evaluation
A detailed mixed methods process evaluation is embedded in the CHIEDZA and nested STI CRT.The process evaluation is based on the MRC Process Evaluation Framework exploring three core functions of the intervention components: i) implementation, ii) mechanisms of impact and iii) context 46 .An intervention logic model describing the inputs, activities, outputs, outcomes, and intervention impact has been developed (Figure 4).The situation, causal assumptions and anticipated conditions in the logic model were informed by the formative research 24 , existing literature and contextual understanding of the study setting.The logic model reflects how the intervention is intended to work.From the outset, the intervention was designed as an adaptative model that would be adjusted in response to implementation and ethical needs, and contextual events during the trial period.The process evaluation therefore had three objectives.First, to assess fidelity, acceptability, feasibility and quality of delivered intervention components, to provide guidance for sustainable and scalable implementation of the intervention.
Second, to identify likely mechanisms that produce the trial's observed outcomes.Third, to offer real-time feedback during the implementation to inform adaptation of the intervention.
The mixed methods approach in the process evaluation includes both sequential explanatory and sequential exploratory approaches 47,48 .Qualitative data are collected through 1) nonparticipant observation of CHIEDZA sites, 2) interviews with both CHIEDZA providers and clients, 3) interviews with youth living in intervention communities and 4) observations of study team meetings.Individual level quantitative data include 1) client demographic information 2) number of attendances over time, 2) uptake of different services at each visit and 3) intervention coverage.As part of the process evaluation, all organisations (health facilities, community-based organisations and non-governmental organisations) providing similar services to those provided by the intervention (e.g.family planning, HIV testing etc) across all clusters are mapped to understand the contribution of these to the outcome.Data are collected longitudinally at multiple time points.Data collection and analysis processes are iterative, and data are triangulated during analysis.

Economic evaluation
An economic evaluation will be undertaken from the health provider perspective to estimate the incremental cost-effectiveness of the intervention.Primary costing will be undertaken from to estimate the economic costs of all observed resources used to provide the interventions and SOC.Economic costing will be based on global costing guidelines 49 , and will provide evidence of the total and unit cost of CHIEDZA and STI services and routine HIV and STI services through the SOC.Overall economic costs will include start-up, capital and recurrent costs, including personnel and consumables, and will be estimated using financial costs obtained from expenditure records and economic costs obtained from micro costing.Total costs will be apportioned to usage of service components to estimate unit costs.Incremental cost-effectiveness will be estimated using effect estimates comparing (a) HIV testing and treatment under CHIEDZA and the SOC and (b) STI screening in CHIEDZA and syndromic management under the SOC.In the intervention arm, data on genital symptoms and STI positivity will inform diagnostic accuracy of syndromic management, which will then be applied to calculate the incremental cost-effectiveness of STI screening relative to syndromic management within CHIEDZA.

Ethics and dissemination
The As the intervention components are established public health interventions (e.g.HIV testing, HIV care and provision of SRH services), consent is implied when clients take up intervention activities and specific written consent to participate is not obtained.National guidelines stipulate that those aged 16 years and older can give independent consent to accessing HIV and SRH services.Written, informed consent is obtained for any qualitative interviews conducted as part of the process evaluation.Zimbabwean guidelines stipulate that those aged under 18 years should have parental consent to participate in research but due to the risk of desirability bias affecting responses and the minimal risk associated with participation in interviews, a waiver for parental consent for participants aged 16-18 years was obtained from ethics review bodies.
For the endline survey, eligible individuals are shown a video of the study procedures enacted and narrated by the study team, emphasising that participation is voluntary and that their information will be kept confidential.The video is shown on a tablet with narration in English, Shona or Ndebele and consent is provided electronically on the tablet with a signed paper copy maintained by participants.The video is available online for participants.
Findings will be presented locally to policy makers and other relevant stakeholders at provincial and national level meetings, and at local academic meetings.Results will also be shared with participating communities and collaborators.International dissemination will be through academic peer-reviewed publications in open-access journals and presentations at international meetings.A study report will be sent to regulatory authorities and a policy brief will be developed to facilitate translation of findings into health policy where appropriate.

Trial status
At the time of manuscript submission, the intervention is ongoing in two provinces and data collection for the endline survey has begun in one province.

Discussion
The CHIEDZA trial is the first to investigate the impact of an approach that addresses every step of the HIV cascade on HIV care outcomes among youth.The integration of components is hypothesised to have a synergistic effect because the outcome of each step of the HIV care cascade is conditional on that of the previous step.The trial speaks strongly to WHO recommendations to develop and evaluate client-centred approaches that simplify and adapt HIV services across the cascade and reduce pressure on heavily over-burdened clinics i.e. differentiated service delivery 50 .
The outcome is of public health importance because viral load suppression reduces HIV-associated morbidity and mortality in individuals.It also reduces risk of HIV transmission with those being virally suppressed considered to be at negligible risk of transmitting HIV to their partners (Undetectable=U ntransmissible or "U=U") 6,51,52 .The trial will assess outcomes at population level, and therefore the outcome is dependent both on the inherent effectiveness of the intervention as well as on its coverage.
Notably, the intervention incorporates comprehensive SRH services, HIV prevention and general health information and counselling.As well as being essential components of HIV service delivery, integration of these services will likely lead to higher acceptability and greater potential for scalability 14,29 .Such a model could be adapted to serve as a platform for delivery of services for other chronic conditions e.g. common mental disorders, for which lay counsellor interventions have recently been shown to be highly effective in Zimbabwe and are being scaled-up nationally 53 .
The nested STI CRT investigates the impact of providing STI screening, free treatment of those testing positive and partner notification on population-level prevalence of STIs.The STI trial will provide data not only on the feasibility and acceptability of STI screening among youth, but also data on the additional yield over and above syndromic management.In addition, the complete STI care cascade including partner notification will be investigated.
The CHIEDZA trial has several strengths: uptake of different intervention components at individual level will be accurately tracked across the intervention, which will provide critical insights into service use patterns.The trial has a biological outcome measure, and the use of biometrics will enable the quantification of intervention coverage and contamination.Taking advantage of the large population-based outcome survey, the project will provide much needed data at population level on health-related behaviours and risk factors for the leading causes of morbidity in youth.The study incorporates a detailed process evaluation and study of cost-effectiveness to inform scalability should it prove effective.We acknowledge several limitations: the study excludes rural settings; the nested STI trial was conducted in two of the three provinces and STI screening was implemented for the last 12 months of the intervention (due to funding constraints), which may not provide sufficient coverage.The COVID-19 pandemic has influenced the trial intervention.Intervention implementation was interrupted for a six-week period in March 2020.Subsequently, adaptations have been made to service provision during lockdowns and curfews.These have included limiting social activities, restricting number of clients and reduced operating hours 54 .The intervention period was extended by six months to mitigate against some restrictions related to the COVID19 pandemic.The impacts of these adaptations are being investigated in the ongoing process evaluation In summary, this pragmatic CRT addresses a critical public health issue, and the embedded detailed process evaluation and cost-effectiveness studies will inform scalability and translation into policy and practice should the intervention be effective.Although publication is late in the study timeline, there are the benefits of publishing a study protocol close to data analyses including transparency of research processes and allowing for collaboration.
A few specific comments.

Overall:
Please check and correct for the appropriate use of tense.

Introduction:
Regarding the statement on general global decline in HIV.UNAIDS Global AIDS Update 2022: "Faltering progress meant that approximately 1.5 million new HIV infections occurred last year-more than 1 million more than the global targets.In too many countries and for too many communities, we now see rising numbers of new HIV infections when we needed to see rapid declines." Authors indicate that asymptomatic STIs are not treated with syndromic management.However targeted diagnostic management would probably miss asymptomatic STIs too as there will not be an indication for testing.

Study design:
Are the researchers comparing facility-based care to community-based care?
Is the integration of services an intervention or delivery strategy?
It is not clear what existing standard of care services for youth with HIV in the facility entails-is there a "one stop shop" etc.?
Is BP (and other vital signs) routinely measured under health information as high BP is an outcome?Normal BP for an adolescent is less than 120/80 or 140/90mmHg?

Study setting and population:
Definition of an adult (>15 years) -is this correct?
Are Harare, Bulawayo and Mashonaland provinces or cities?
The cluster has to be serviced by a defined primary care clinic.Are these clinics the "control clinics"?
Please define what youth workers are.
HIV testing and care services -who collects the ART from the HIV care clinic if client opts for care via CHIEDZA?"Youth as equal partners" -Love this.
Table 2-Linkage of care and ART initiation-Is there TB prophylaxis provided?
Is the standard of care "annual" HIV testing/screening in Zimbabwe?
Besides VL monitoring, are there ALT or CD4 count monitoring?
Family planning-IUCD and Implants-which sites provided these services?LARC is by far superior.
Partner notification process-done for all partners?
Healthy sexual behaviour -male clients are not on the same table row as VMMC.
Where is the online link for mental health screening?

Sample size.
Is it supposed to state: 80% power-detect a difference of 23% and 90% power to detect a difference of 19% Intervention is focused on youth, endline surveys are performed in adults.
Thank you.

Is the rationale for, and objectives of, the study clearly described? Yes
Is the study design appropriate for the research question?Partly

Are the datasets clearly presented in a useable and accessible format? Yes
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: HIV and NCD prevention, management 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, however I have significant reservations, as outlined above.

Introduction:
Regarding the statement on general global decline in HIV.UNAIDS Global AIDS Update 2022: "Faltering progress meant that approximately 1.5 million new HIV infections occurred last year-more than 1 million more than the global targets.In too many countries and for too many communities, we now see rising numbers of new HIV infections when we needed to see rapid declines." Thank you for noting this point which we agree appropriately reflects the trend in terms of HIV incidence globally.We have now clarified in the text that although there has been some decline in HIV incidence this remains suboptimal.
Authors indicate that asymptomatic STIs are not treated with syndromic management.However targeted diagnostic management would probably miss asymptomatic STIs too as there will not be an indication for testing.This is an important clarification which we have now corrected throughout the manuscript.Our trial offered STI screening to individuals who were asymptomatic.As such we did not provide targeted diagnostic management and were able to test and treat both individuals with and without symptoms for the three STIs.

Study design:
Are the researchers comparing facility-based care to community-based care?
We are comparing community-based care to existing services in communities-these are predominantly facility-based.There may however be vertical, single-service programmes for example a family planning delivery programme and many community-based organisations provide information and education on sexual and reproductive health.However, these would be expected to be equally distributed across the two arms.
Is the integration of services an intervention or delivery strategy?
The intervention is both the integrated services and community-based delivery.
It is not clear what existing standard of care services for youth with HIV in the facility entails-is there a "one stop shop" etc.? Is BP (and other vital signs) routinely measured under health information as high BP is an outcome?Normal BP for an adolescent is less than 120/80 or 140/90mmHg?
No it isn't.The BP is not a trial outcome.Taking advantage of such a large populationbased survey, we used it to get population level data on health behaviours and conditions of relevance to youth as such data is scarce in young people.We clarify in the manuscript that these were not trial outcomes.

Study setting and population:
Definition of an adult (>15 years) -is this correct?
In the context of HIV prevalence-yes that is correct.
Are Harare, Bulawayo and Mashonaland provinces or cities?
They are provinces.Bulawayo and Harare are also the two largest cities and are also administered as provinces -each with an equivalence of a provincial health directorate.
The cluster has to be serviced by a defined primary care clinic.Are these clinics the "control clinics"?
No, these are not control clinics.These are clinics from where the commodities such as ART was accessed and where clients who registered for HIV care in CHIEDZA would be registered within the HIV programme, where statistics about family planning, and other services would be recorded.This was to ensure that the intervention was not creating parallel reporting structures.
Please define what youth workers are.
Youth workers were young people aged up to 24 years at the time of recruitment.Their specific role was to engage youth who came to the service, and a organise social activities and help foster a youth-friendly, safe environment.They underwent training, were renumerated but did not provide any consultation or specific services other than running information sessions such as condom demonstrations and how to use a menstrual cup.We have now added a definition of youth worker in the manuscript.
HIV testing and care services -who collects the ART from the HIV care clinic if client opts for care via CHIEDZA?
The CHIEDZA nurses did."Youth as equal partners" -Love this.
Table 2-Linkage of care and ART initiation-Is there TB prophylaxis provided?
TB screening was performed.TB Prophylaxis, while in the guidelines for several years, was not routinely implemented until 2021 when 3HP was rolled out.Prior to 2021, 3 months of isoniazid was recommended in the guidelines but had low rates of offer and uptake and was therefore not routinely implemented.
Is the standard of care "annual" HIV testing/screening in Zimbabwe?
Yes, the current standard of care is annual testing for those who have previously tested HIV negative.
Besides VL monitoring, are there ALT or CD4 count monitoring?
CD4 count was performed at initial assessment.ALT and CD4 count monitoring is not standard of care.It is only performed if clinically indicated.
Family planning-IUCD and Implants-which sites provided these services?LARC is by far superior.
All intervention sites did…however in an effort to ensure that we integrated with what was available already we partnered with an NGO providing family planning to deliver LARCs.Thus, we were reliant on their schedules to deliver LARCs and we recorded when they were present a spart of the process evaluation-and we report on this.
We are unclear whether a response to the comment "LARC is by far superior" is required?
Partner notification process-done for all partners?
Yes, this was offered to all clients who had an STI.
Healthy sexual behaviour -male clients are not on the same table row as VMMC.

This has been corrected.
Where is the online link for mental health screening?
We are not clear to what the reviewer is referring to by this comment Sample size.
Is it supposed to state: 80% power-detect a difference of 23% and 90% power to detect a difference of 19%.
We have 90% power with a 23% difference and 80% power with a 19% difference.No changes have been made to the manuscript.
Intervention is focused on youth, endline surveys are performed in adults.
Endline survey performed in individuals aged 18-24yrs.16-24-year-olds were eligible for services.The age difference is to ensure that all outcome survey participants had had at least 2 years "exposure" to the intervention.
Thank you.

Introduction:
First paragraph "In these countries, it was estimated that only 40-50% of 15-24-year-olds living with HIV were aware of their HIV status compared with 66-73% of those aged >15 years" -it appears comparing the same age groups -do you mean compared to all age groups?Likewise for following sentence

•
Well-established linkages to other services a Biometrics used for recording service uptake and pseudonyms used for managing STI test results b Except where required for clinical care e.g., registration with national HIV care programme IEC -Information, Education and Counselling SMS -Short Message Service CAPS -CHIEDZA Adolescent Peer Support staff when drugs are collected for supply through CHIEDZA.This ensures that clients remain part of the national HIV programme.HIV treatment is provided according to national guidelines and there are clearly defined referral pathways to a health facility should there be any clinical indications (e.g., severe toxicity, incident symptoms, suspected treatment failure).

Figure 4 .
Figure 4. CHIEDZA Logic Model describing the Implementation of the intervention as intended, intentions to adapt, and the anticipated outcomes and Impact.

No, there
aren't one-stop shop services -in HIV clinics, HIV treatment and adherence support is offered.Other services are accessed in primary care clinics, by donorfunded vertical programmes or through the government and Ministry of Health.Even the DREAMS programme which also focuses on youth does not offer an integrated services, but beneficiaries are referred to different partners for specific services.The concept of offering integrated HIV and SRH services is the subject of this study.We have clarified this in the methods under the study design.

○
Study rationale:What do you mean by existing programmes being "vertical"?○Study design:It would be useful to describe the current standard of care pathway in more detail -what care do those in control clusters receive?○Economic evaluation:If successful, how will the study assess whether the intervention reduces long term costs of sequalae of untreated HIV / STI ○ Is the rationale for, and objectives of, the study clearly described?YesIs the study design appropriate for the research question?YesAre sufficient details of the methods provided to allow replication by others?YesAre the datasets clearly presented in a useable and accessible format?

Table 2 . Outline of the CHIEDZA intervention and rationale for choice of components. Intervention component Service provided Eligibility Rationale HIV services covering the whole care cascade
following week, but clients who have a positive test result are also contacted by telephone.A minimum of five attempts over three months are made to contact clients with positive CT/NG test results before considering them lost-to-follow-up.For the purposes of follow-up, clients are identified by pseudonyms and no identifying information is collected.All those with positive tests who are contactable and/or return to the CHIEDZA service are provided comprehensive management, including treatment, risk reduction counselling and partner notification slips.Partners are treated free of cost regardless of their age and place of residence.
aHIV viral load at 6 months, 12 months and annually thereafter for clients initiating ART and accessing HIV care at CHIEDZA.For clients accessingARTelsewhere but attending CHIEDZA for other services, viral load testing offered opportunistically b Choice of one option (pads or cup) that can be swapped after 3 months c Disposable pads used with a cup initially to provide reassurance that there is no leakage d IUCD and implants delivered by partner providers who come at deliver the services at CHIEDZA e Repeat STI testing offered after 3 months f Flavoured, textured and high quality, with packaging designed to be appealing to youth g Includes an information manual with information on age-appropriate general health and other issues (also available online) h Referral to health and other social and educational services relevant to the issue OMT: Oral mucosal HIV test; TB: Tuberculosis; CrAg: Cryptococcal Antigen; ART: Antiretroviral Therapy; SRH: Sexual and Reproductive Health; IUCD: Intrauterine Contraceptive Device; STI: Sexually Transmitted Infection; VMMC: Voluntary Male Medical Circumcision; IEC: Information, Education and Counselling circumcision (VMMC) services have been identified for each of the intervention clusters and expedited referral is arranged for clients requesting VMMC.In cases of intimate partner violence (IPV) or sexual assault, youth are referred for to defined services.Nested STI CRT Screening for Chlamydia trachomatis (CT), Neisseria gonorrhoea (NG) and Trichomonas vaginalis (TV) is offered to clients in the last 12 months of the intervention in the eight intervention clusters (Harare and Bulawayo).Eligibility includes not having a previous test in the last 3 months.Testing for TV infection is carried out in women only and is performed on self-collected vaginal swabs using a lateral flow assay (Sekisui Diagnostics, Massachusetts, USA) with results available within 15 minutes.Testing for CT and NG infection is performed on urine samples using the GeneXpert platform (Cepheid, California, USA).All clients who take up CT/NG testing can collect results from the CHIEDZA service the Information, education, and counselling.Information, education, and counselling (IEC) materials about SRH, HIV and general health issues are available in the form of video clips and a health manual available at the centres and online through the CHIEDZA website.General health counselling is provided, with onward referral to other health service providers for relevant care where appropriate e.g.mental health.

Table 1 -
info sent by SMS message.What proportion of the youth have mobile devices?

Table 1 -
info sent by SMS message.What proportion of the youth have mobile devices?