AM and LAS are joint first authors.
STRENGTHS AND LIMITATIONS OF THIS STUDY
This study rigorously evaluates an intervention which was derived from behavioural economics and human-centred design that has already demonstrated preliminary effectiveness in a pilot study, and is now being regionally scaled.
Our data metrics are tied to a clear impact pathway, allowing us to evaluate where or how far along the hypothesised path from the intervention input to the primary effectiveness outcomes we can demonstrate impact.
We will use a rigorous, mixed-methods implementation science study to understand factors relating to programme effectiveness.
While cluster-level outcome data collection preserves anonymity of individual adolescent girls and young women’s (AGYW) sexual healthcare-seeking behaviours, it presents a challenge whereby it may be difficult to differentiate intervention impacts with other community-wide efforts to prevent HIV and unplanned pregnancy among AGYW.
Introduction
Adolescent girls and young women (AGYW; aged 15–24 years) are disproportionately affected by HIV infection and unintended pregnancy in sub-Saharan Africa. There, approximately 44% of the 14 million unintended pregnancies that occur annually worldwide and nearly 20% of incident HIV cases were among AGYW aged 15–24 years, despite them comprising just 10% of the population.1 In Tanzania, nearly one-third of pregnancies occur in AGYW, 23% have unmet need for contraception and 2.1% of AGYW live with HIV.2 3 AGYW face unique challenges in accessing HIV and sexual and reproductive health (SRH) care, including inability to pay for transport or related costs, concerns about privacy and sparsely available AGYW-centred care.4 5 Limited access to and uptake of SRH services among AGYW6 hinders the potential of current and upcoming HIV prevention and SRH strategies such as new formulations for HIV pre-exposure prophylaxis (PrEP)7 or self-injectable contraception.8
While public health facilities fill an important role in SRH care in lower-income countries, interest in other avenues for healthcare provision via private sector channels,9–12 such as small drug shops and pharmacies, is growing.13–17 Community drug shops sell medicine and health products (eg, hygiene, vitamins)13 15 18 and, in lower-income countries, provide over half of all healthcare and are often the only access point in underserved areas.18 Since they effectively expand health systems’ reach and are a gateway to clinical care, many countries are training this cadre to deliver priority initiatives (eg, SRH, diarrhoea, malaria).15 16 19–29 Tanzania’s Accredited Drug Dispensing Outlet (ADDO) Programme is a model for formalising this sector. ADDOs are in nearly every community and can improve access to quality-assured pharmaceuticals, including contraceptives.30 31 For example, adding malaria services to ADDOs’ workflow has improved proper dispensing of antimalarial drugs,32–34 demonstrating the promise of the ADDO platform.
To lower HIV incidence and unintended pregnancy among AGYW, we must first increase access to and uptake of contraceptives and HIV prevention, with HIV testing as a critical gateway.35 Compared with adults, AGYW are less aware of their HIV status despite available testing at clinics.36–38 WHO-recommended HIV self-tests (HIVST) using oral fluid kits are effective for safe and confidential testing.39 40 Studies in Malawi, Zimbabwe and Zambia show high demand for HIVST, especially among adolescents.41–44 Further, better access to contraception at drug shops is an underexplored yet potentially effective strategy for addressing unmet need among AGYW; though overall contraceptive demand is low among AGYW (19%) compared with older women (35%) in Tanzania,45 national data show that AGYW are twice as likely as older women to obtain contraceptives from drug shops. ADDOs provide a largely untapped opportunity to reach AGYW with these underused prevention products, including HIVST and contraceptives; and new strategies are needed to drive demand among AGYW and simultaneously provide girl-friendly safe spaces to learn about and access these products.
Intervention design, theoretical foundation and pilot study
From 2018 to 2019, we used a process integrating human-centred design (HCD) and behavioural economics to develop the Malkia Klabu (Queen Club) intervention. Throughout each stage of the design process (figure 1),46 complementary approaches from HCD and behavioural economics were considered within the context of future scalability of Malkia Klabu. The design process is described in detail elsewhere47; the outcome of this formative work was Malkia Klabu, an enhanced loyalty programme whereby AGYW members can receive free contraceptives (oral pills, emergency contraception, condoms), urine pregnancy tests and HIVST kits while earning punches for shop purchases that can be redeemed for small prizes of increasing value (eg, soap, nail polish).
Enlarge this image.Figure 1. Logic model outlining intervention framework. ADDOs, Accredited Drug Dispensing Outlets; AGYW, adolescent girls and young women; HIVST, HIV self-test; SRH, sexual and reproductive health.
Using a rigorous mixed-methods implementation science evaluation framework, we subsequently tested the feasibility and acceptability of Malkia Klabu in 2019 via a 4-month pilot randomised controlled trial among 40 shops (1:1 intervention/control). We found that shops implementing Malkia Klabu had 4.4 times the number of AGYW customers, distributed over twice as many HIVST kits, 46 times higher contraceptives and 7 times more referrals for clinic-based SRH services relative to business-as-usual comparison shops.48 At the same time, inconsistencies in shopkeepers’ records highlighted the need for better systematic inventory management and sales documentation as the intervention was readied for scale. Nevertheless, these findings demonstrate the feasibility and acceptability of Malkia Klabu among AGYW and shopkeepers. Importantly, we affirmed that the multifaceted solutions incorporated into the intervention served to overcome a number of structural and behavioural barriers for both sets of actors. We describe our planned study protocol involving a cluster-randomised controlled trial (c-RCT) over 24 months (Standard Protocol Items: Recommendations for Interventional Trials checklist in online supplemental appendix 1).
Regionally scaled effectiveness study
This study builds on our pilot study and will directly measure effectiveness of the Malkia Klabu Programme when regionally scaled, while also including metrics to understand adaptability (figure 1). We describe our planned study protocol involving a c-RCT over 24 months.
Methods
Study design
We will conduct a 24-month, unblinded, parallel two-arm type 1 effectiveness implementation49 c-RCT of the Malkia Klabu Programme in approximately 40 health facility catchment areas to achieve the following study aims:
Aim 1: at the population level, evaluate the effectiveness of the Malkia Klabu Programme on HIV diagnoses and antenatal care (ANC) registrations among AGYW in Lake Zone, Tanzania.
Aim 2: verify individual-level behaviour change, including recent HIV testing, unmet need for contraception and linkage to care, and describe demand-side pathways (eg, Malkia Klabu reach into AGYW networks) through which Malkia Klabu may reduce HIV and unintended pregnancy among AGYW.
Aim 3: understand supply-side innovations and challenges, and determine how implementation fidelity relates to programme effectiveness. We will use a mixed-methods implementation science study to pinpoint supply-side factors influencing effectiveness (eg, implementation models, programme fidelity, shop characteristics).
Study setting
The study will take place in the Shinyanga and Mwanza provinces in the Lake Zone of Tanzania (figure 2). Shinyanga has a population size of 2.2 million and Mwanza is a home to 3.7 million people according to the 2022 national census report,50 with 835 and 438 health facilities in each province, respectively, the majority of which (56.4% and 7.9%, respectively) were dispensaries. In Tanzania, 22% of women aged 15–19 years have ever been pregnant, with the proportion of pregnancies among AGYW higher in rural versus urban areas (24.9% vs 16.4%) as well as in our study regions, where 16.3% in Mwanza and 20.8% in Shinyanga of women aged 15–19 years have ever been pregnant.51 A recent study of over 400 pregnant AGYW seeking care at ANC clinics in Mwanza found that nearly half tested positive for at least one of six common sexually transmitted infections (STIs).52 The Shinyanga region has the second highest HIV prevalence in Tanzania where 6.2% of AGYW are living with HIV compared with the national average of 3.7%.53
Study arms
To facilitate monitoring and management of general sales data in all ADDO shops as well as the Malkia Klabu Programme processes in intervention shops, the study is collaborating with Maisha Meds, a point-of-sale inventory management system,54 to provide the tablet-based system to participating ADDOs free of charge to facilitate collection and analysis of sales data. All eligible ADDO shops in selected catchment areas willing to participate will receive additional Maisha Meds training during their study training.
In both arms, ADDO staff will be trained to answer questions about HIVST and contraception, and give referrals to specific local study-affiliated public health facilities for confirmatory HIV testing, PrEP, long-acting reversible contraceptives (LARCs; for example, intrauterine devices, implants), ANC, STI services and general SRH services.
Intervention arm
The intervention arm will retain the components of Malkia Klabu developed and tested in the pilot, including the use of a loyalty card (online supplemental appendix 2), gifting a free, opt-out HIVST kit upon sign up, giving free SRH products (oral contraceptives, emergency contraception, condoms, HIVST kits and urinary pregnancy tests) upon request and allowing punches on their card for purchases that can be redeemed for two levels of prizes at future visits.
Consenting and enrolled intervention shops will receive comprehensive, 2-day training to review the objectives of the study, the Malkia Klabu Programme, as well as training in providing girl-friendly services, HIVST kits and how to use the Maisha Meds system. Intervention shops will have referrals for female youth-friendly clinicians who are already stationed at the health facility and will attend a study training focused on the provision of girl-friendly SRH care, as well as peer navigators in the study-affiliated public health facilities. Both peer navigators and the youth-friendly clinicians will log the interactions, noting service requested and services provided, including further referrals, and linking each record to a Malkia Klabu ID without collecting any other identifiers.
Upon completing training, ADDOs will receive all the commodities necessary to implement Malkia Klabu in their shop, including a 1-month supply of HIVST kits reserved for free distribution to AGYW who opt into Malkia Klabu. The kits will be prepackaged with specific referral information and their distribution will be tracked in Maisha Meds along with all other study-related and non-study-related sales of products. Further, all ADDO staff will be regularly reimbursed at fixed prices for the SRH products provided freely to Malkia Klabu members (condoms, oral contraceptives, emergency contraception and pregnancy tests), but which will be independently procured by shops per business as usual; the study will provide HIVST kits.
In addition to the facility-based youth-friendly clinician, ADDO shops randomised to intervention catchment areas will receive contact information for one to two local peer navigators who will facilitate linkage to care to the study-affiliated referral health facility. Peer navigators will be recruited in one of two ways: (1) AGYW already working at referral facilities in a similar role and (2) community volunteers recruited through referrals from community leaders.
Control arm
Shops enrolled in the control arm will receive separate (from intervention shops) 1-day training on HIVST kits and the use of the Maisha Meds system, as well as a 1-month supply (eligible for restocking) of HIVST kits marked for free distribution to AGYW (to be tracked in Maisha Meds) with non-specific referral information to a nearby public referral facility. Contraceptives and pregnancy tests will be procured and sold as per normal business operations.
Randomisation
Approximately 40 catchment areas will be recruited. Starting with a listing of government health facilities in the study regions, the sampling frame will only include catchment areas with three public health facilities or less, at least one of which offers both HIV and ANC services, to ensure accessibility to these services while facilitating data collection for our primary outcome and thus clearer attribution of population-level effects to the intervention. Catchment areas will be randomly assigned 1:1 to intervention or comparison arms within strata of region, population density and antiretroviral therapy (ART) caseload using a constrained randomisation approach to ensure balance in these key covariates.55 We will engage the Regional and District Pharmacy Councils, the regulatory authorities overseeing pharmacy and ADDO registration and monitoring, to recruit ADDO shops in each ward (ranging from ~3 to 10 per ward).
Recruitment of study participants
Study participants will be selected from three populations across study aims:
Aim 1:
Public health facilities: within catchment areas with >1 eligible public health facility, one health facility will be chosen at random, otherwise if just one public health facility exists in that catchment area, it will be recruited.
ADDO shops: within selected catchment areas, we will randomly approach and invite up to eight registered ADDO shops with the aim of enrolling three to four ADDOs per catchment area; shop owners and staff aged 18 years or older will be eligible to participate. Further, shops must be willing to sell or distribute all SRH products given as part of Malkia Klabu to AGYW as well as the general population. Given randomisation is at the catchment area, all ADDOs within a catchment area will belong to the same study arm.
Aim 2: approximately 560–600 AGYW aged 15–24 years who reside in selected catchment areas and speak Kiswahili will be recruited using respondent-driven sampling (RDS). We will begin our initial search for participants using approximately 40 AGYW ‘seeds’, identified from referrals from the study’s Youth Advisory Boards, local schools, youth organisations (including for out-of-school AGYW) and participating ADDO shops, with one per catchment area. The initial seed will receive brief training from study staff in peer recruitment before being given up to three coupons to distribute to eligible AGYW in their network. This process will be repeated in until we have reached our recruitment target.
Aim 3: we will conduct in-depth interviews (IDIs) with the following groups:
Referral public health facility staff: approximately 8–12 female referral facility staff total aged 18 years or older from participating referral public health facilities in the intervention arm will be recruited for IDIs at study endline.
ADDO shopkeepers: up to 30 ADDO shop owners and staff among participating ADDO shops in the intervention areas will be recruited for IDIs at study endline.
Aim 3 will also include quantitative surveys among shopkeepers already recruited in aim 1, as well as mystery client visits (further described below) which will not entail any participant recruitment.
Outcomes
The primary endpoints for the study are as follows (table 1):
Cumulative number of HIV-positive diagnoses among AGYW aged 15–24 years over the 24-month study period.
Cumulative number of ANC registrations among AGYW aged 15–24 years over the 24-month study period.
Overview of primary and secondary outcomes and data sources to be collected among all shops (intervention and control)
| Aim | Outcome | Data collection approach | Observation period |
| 1 |
| Health facility data abstraction from participating referral health facilities | Cumulative count over 24 months |
| 2 | ART initiation among AGYW | Health facility data abstraction from participating referral health facilities | Cumulative count over 24 months |
| Sales data entered by ADDOs in Maisha Meds | Cumulative count over 24 months | |
| Endline RDS survey among AGYW | Recall period of ever exposure over 24 months | |
| Unmet need for contraception§ | Endline RDS survey among AGYW | Contemporaneous |
*Defined as the number of AGYW who report ever receiving a Malkia Klabu card (ie, enrolling) as the numerator and the total number of AGYW interviewed as the denominator.
†A proportion defined as the number of AGYW who report testing for HIV in the last 6 months as the numerator over the total number of AGYW interviewed.
‡A proportion defined as the number of AGYW who report taking a pregnancy test in the last 6 months as the numerator over the total number of AGYW interviewed.
§Defined as the number of women who do not want to get pregnant but who are not using a method of birth control and are sexually active; the numerator is the number of women not using a method of birth control and who do not desire to get pregnant, the denominator is the total number of sexually active women.
ADDOs, Accredited Drug Dispensing Outlets; AGYW, adolescent girls and young women; ART, antiretroviral therapy; HIVST, HIV self-test; RDS, respondent-driven sampling.
These data will be abstracted from routinely collected health facility data and aggregated to the catchment (synonymous with ward) area level. Our team will abstract from the study-affiliated (ie, referral) health facility and from nearby facilities within the same catchment area. Abstraction will occur annually.
The secondary outcomes, data source and observation period are outlined in table 1.
Data collection
Aim 1
Data for the primary and secondary outcomes will be abstracted from health facility-level records annually and will include aggregate data about: number and outcomes of HIV tests by AGYW, ANC registrations by AGYW, ART initiations by AGYW (aim 2) at baseline, 12 and 24 months, reported by age groups 15–19 and 20–24 years.
Aim 2
HIVST kits and contraception distribution will be tracked in Maisha Meds over the 24-month study period. To measure programme exposure, HIV testing and unmet need for contraception, bilingual (Kiswahili and English) study staff will survey 560–600 AGYW (n=280–300 per region) at study endline. Survey topics will include demographic characteristics, exposure to Malkia Klabu (if any), recent HIV testing, contraceptive use, relationship to the referring participant (eg, friend, relative) and the size of their social network to facilitate the RDS analysis.56 57
We will conduct two rounds of four focus group discussions with AGYW who are registered Malkia Klabu members (n=4 each at 12- and 24-months) with diverse, purposefully selected AGYW (eg, various ages, in and out of school) who will elicit AGYW-level and community-level perceptions of Malkia Klabu over time and contextualise the pathways of potential impact on HIV and unintended pregnancy as suggested in the pilot findings,48 including integration with organic behaviours (eg, visiting ADDOs at the behest of others), the motivational effect of behavioural economics constructs (eg, rewards) and acceptability to community members.
Aim 3
To understand supply-side and implementation heterogeneity factors that may affect study outcomes, further described in online supplemental table 1, we will conduct (1) IDIs with ADDO staff; (2) quantitative interviews with ADDOs at baseline, midline and endline (topics outlined in online supplemental table 1); (3) IDIs with referral facility staff and (4) mystery client visits made by study-trained actors to intervention arm participating shops. Additional details about aim 3 data collection methods and measures are in online supplemental appendix 3.
Sample size
The trial is powered for the primary (aim 1) outcome of cumulative HIV-positive diagnoses among AGYW at 24 months based on facility records of HIV testing by AGYW aggregated to the catchment area. Using publicly available data from 2019 on HIV testing from Shinyanga and Mwanza Regions, along with estimates of HIV prevalence among AGYW (2.1%),58 we estimate a mean of 40 HIV diagnoses (SD=20.9) among AGYW over 24 months in each comparison (control) catchment area. With 40 health facility catchment areas, we will have ≥80% power to detect an increase in mean AGYW HIV diagnoses of ≥18.5 (eg, minimum detectable effect, rate ratio ≥1.46). This 46% increase in AGYW diagnoses is equivalent to the identification of less than one AGYW living with HIV per catchment area per month during the 24-month study period.
Overall, we anticipate that the analysis for the trial will include 1970 HIV test results among AGYW abstracted from referral facility records across all catchment areas. If the prevalence of HIV among AGYW is lower than expected in the comparison group (eg, 1.5%), we are adequately powered at 80% to detect a difference of 13.2 or greater in mean AGYW HIV diagnoses in the intervention arm compared with the control arm over the 24-month period. Likewise, if prevalence of HIV among AGYW is higher than expected in the comparison group (eg, 2.7%), we are adequately powered at 80% to detect increases of 23.7 or greater in mean AGYW HIV diagnoses.
Statistical analysis
We will conduct an intent-to-treat analysis to estimate the intervention’s effect on cumulative HIV-positive diagnoses and ANC registrations among AGYW (aim 1). For each outcome, we will specify a generalised linear regression model (specifying the Poisson family, log link and robust variance estimator): ln[Yi]=β0+β1x1+β2x2+ζj, where Yi is the cumulative number of positive HIV tests or ANC registrations among AGYW in each catchment area i, and x1 represents area assignment to the intervention (x1=1) or comparison (x1=0) group. The model will also control for region, the stratification factor that was used in the randomisation. β1 is the effect estimate of interest: the log rate ratio of HIV diagnoses (or ANC registrations) among AGYW in intervention versus comparison catchment areas (x2 and ζj are the stratification variable and the error term, respectively). We will conduct a permutation test based on residuals from the model as proposed by Gail et al59 which has been shown to perform well in simulation studies to test the null hypothesis of no treatment effect.60 Per Consolidated Standards of Reporting Trials,61 we will report unadjusted and adjusted estimates. Sensitivity analyses assessing effect measure modification or confounding by age will be conducted.
For our aim 2 outcomes, we will use a Poisson model to estimate a rate ratio for the rate of HIVST kits and contraceptives distributed to AGYW over the 24-month study period, controlling for region as the stratification factor and accounting for catchment-level clustering. Using a log binomial model to compute a prevalence ratio, we will also assess effects on recent HIV testing, recent pregnancy testing, programme exposure and unmet need for contraception with the same model specifications for stratification and clustering, as well as the permutation test. We will use the same model specifications for aim 1 to estimate the impact on ART initiations.
For the aim 3 implementation study, we will use an open-coding approach,62 63 based on Proctor’s implementation indicators, to assess ways in which AGYW-level, shop-level, facility-level and community-level factors facilitate or hinder uptake, fidelity, scalability and sustainability.64 65 Data analysis will be iterative, allowing new themes to emerge.66 Data will be independently coded by two team members using Dedoose software.67 Qualitative and quantitative data will be triangulated to describe supply-side innovations and challenges and define relatively high-fidelity and low-fidelity models for the degree to which each shop implemented Malkia Klabu as intended.
Timeline
Recruitment of catchment areas and ADDOs will take place on a rolling basis by region between May and October 2022; recruitment of AGYW will not occur until September 2024. Intervention activities will launch from July to November 2022 and the intervention will run for 24 months, with data collection points at 0, 12 and 24 months for the ADDOs and 24 months for the AGYW (online supplemental table 2). Outcome assessments measured via chart abstraction will be collected 12- and 24-months post-launch and entered into REDCap-enabled tablets.
Data management plans
Every effort will be taken to protect the confidentiality of participants. Our study team will ensure proper and secure storage of all paper forms (eg, consent forms) upon transfer from the interview site or shop to the study offices in a locked cabinet, which only designated staff members will have access to for data entry. Survey data will be collected during the interview on a tablet computer and automatically uploaded into the secure REDCap cloud-based survey platform.
We will retain the study data for up to 10 years after it is complete. At the completion of data cleaning and the primary analysis, we will prepare anonymised datasets. Only the study principal investigators (PIs), co-investigators and designated study members will have access to the data.
Ethics and dissemination
This study has been approved by the National Institutes of Health (NIH), the University of California, San Francisco Institutional Review Board (IRB) and Tanzania’s National Institute of Medical Research (NIMR). The study has convened a Data Safety and Monitoring Board that meets biannually with experts in HIV and adolescent health, and includes representatives from East Africa. Further, the study has convened three regional (given geography, despite two study regions) Youth Advisory Boards with 15–25 AGYW members in each to provide regular feedback on study progress and provide concerns and perspectives related to AGYW SRH.
Any proposed protocol changes will be submitted to US-based and Tanzania-based IRBs and will undergo standard review and approval processes and approval at each institute before proposed changes can be implemented.
The implementing partner is Health for a Prosperous Nation (HPON), a Tanzanian-based non-profit research organisation and the same organisation which carried out the pilot study. All HPON staff will be trained in and have experience with collecting sensitive information and the importance of protecting participants’ confidentiality. Study staff are required to sign a confidentiality agreement ensuring that they will not share participant data. All staff based in Tanzania who will seek informed consent from study participants are fluent in English and Kiswahili, trained in protection of human subjects in research and are experienced in obtaining informed consent. As part of the informed consent process, the research staff will verbally communicate with the participants to ensure that they understand the nature and purpose of the research, their anticipated involvement, the potential risks and benefits to participation, and the voluntary nature of their involvement. Consent forms will be read aloud to potential participants unless individuals insist that they are comfortable reading the form themselves. Potential participants who are willing to provide written informed consent will be invited to enrol in the study. For participants under age 18 years, we requested and received a waiver of parental consent given that Tanzanian law allows minors to independently consent for HIV and SRH treatment. Study participants will have contact information for NIMR and the study PIs on the consent form, a copy of which they will be given, to report adverse events.
Study findings will be disseminated in line with both the NIH data dissemination policy and the National Clinical Trials policy of the USA, including annual updates to the Clinical Trials website, as well as at conferences and in peer-reviewed journals. Local dissemination will include presentations to the local Youth Advisory Boards, the local Regional Medical Offices of Mwanza and Shinyanga, and our stakeholders and partners at the Ministry of Health. Upon conclusion of the study and primary analyses, data will be made available to outside investigators upon reasonable request and with a detailed concept proposal.
Patient and public involvement
Our study involves both AGYW, small drug shops, and regional medical and pharmacy leadership in the development and design of our trial as detailed in the Intervention design, theoretical foundation and pilot study section. Further, our study has convened three regional Youth Advisory Boards to provide feedback and input on study launch, the development of study tools and the eventual interpretation of study results; they will convene regularly until study results are finalised and disseminated.
Provisions for post-trial care
There are no anticipated harms or compensation for trial participation. There are no plans for any post-trial care.
Discussion
Innovative interventions are urgently needed to reduce the disproportionate risk of HIV acquisition and unplanned pregnancy among AGYW in East Africa. Previous community-based studies have focused on community health centres to improve adolescent access and individual behavioural factors, using community-based peer navigators and extensive educational programmes to reach AGYW.57 This study builds upon this body of work by integrating drug shops as providers of SRH care and education while still partnering with community health centres in a differentiated model of care. Our study combines concepts from a comprehensive, youth-focused, HCD process,47 an integrated conceptual framework for behaviour change rooted in behavioural economics,46 and epidemiological methodologies to scale and evaluate the Malkia Klabu Programme. Our pilot study demonstrated strong results on HIVST kit and contraceptive distribution and referrals to care among AGYW.48 If proven efficacious at scale, we will have demonstrated that ADDO shops, which are already ubiquitous in Tanzania and provide basic health services in rural settings, can also serve as integral actors in the prevention of HIV and unplanned pregnancy for AGYW. This would simplify further scaling at a national level or transference of the intervention to beyond Tanzanian where small drug shops equipped with basic, but essential commodities (condoms, emergency contraception, etc) are equally ubiquitous, to transform health services for and reduce disparities in sexual health outcomes among AGYW.
Our study has several features purposely designed to facilitate its implementation and eventual scaling. Shopkeepers will use Maisha Meds, a point-of-sales tablet-based application to manage inventory and sales data. This technology integration will simultaneously promote small business management while facilitating data collection and tracking, and will be an essential component of future, scaled versions of Malkia Klabu, which may be used to channel essential commodities supplied by public sector programmes. Further, our study team has convened three regional Youth Advisory Boards with AGYW who regularly provide feedback on study activities, review enrolment and to provide feedback on implementation challenges throughout the study rollout; this is in direct alignment with the Quality Standards Framework for adolescent SRH interventions.58 Based on feedback from health facility staff and the pilot study Youth Advisory Board, this study will work with the enrolled health facilities to develop a comprehensive differentiated care model that involves existing peer navigators and facility counsellors. Specifically, AGYW products sold at ADDOs will contain referrals to trained peer navigators and girl-friendly providers by name who will be trained to provide next-level services (eg, HIV confirmatory testing, LARC) not offered by ADDOs. Finally, our study is a rigorous evaluation of the scaling of an HCD-derived intervention47 infused with concepts from behavioural economics that has already demonstrated preliminary effectiveness48 and has metrics tied to a clear impact pathway; there are few such evaluations.
This study is not without potential limitations. Given the primary outcome of HIV diagnoses and ANC registrations will be measured at the population level, it will be difficult to disentangle other community-wide efforts to prevent HIV and unplanned pregnancy among AGYW. At the same time, the tracking of individual HIVST kits and contraceptives/urine pregnancy tests among AGYW patrons of ADDOs would have presented a number of logistical, ethical and implementation challenges, especially at scale; measuring the outcomes at a community-level preserves anonymity and privacy of the AGYW and is in line with WHO guidelines for HIVST interventions and measurement.39 40 To account for this potential measurement challenge, we purposively enrolled catchment areas with fewer than three public health facilities and will conduct data abstraction among all health facilities within that catchment area in addition to the primary referral facility for our study. In practice, this approach excluded few catchment areas, as the two provinces from which catchment areas were drawn (Mwanza and Shinyanga) are largely rural. To complement this community-level data, we will collect shop-level interim outputs (ie, distribution of HIVST kits and contraception in intervention areas) in line with our impact pathway, allowing our study to evaluate where or how far along the hypothesised path from the intervention input to the primary effectiveness outcomes we can demonstrate efficacy. The choice to use RDS also brings some limitations, mainly in that it is a non-probability sample as well as unknowns around the network size of AGYW.68 Nevertheless, there are strengths of the RDS, including its strength in recruiting ‘hidden populations’, such as sexually active AGYW, those out of school and/or those who are young mothers, which we believe outweigh historical questions raised about RDS.
By the end of this study, we will have comprehensively evaluated an HCD-derived intervention with integration from behavioural economics to evaluate how our intervention affects HIV diagnoses and ANC registrations among AGYW in two high HIV burden regions of Tanzania via a c-RCT. We will have evaluated supply-side and demand-side innovations and challenges, as well as implementation fidelity, to understand specific mechanisms which may have been barriers or facilitators to our study’s implementation. This is novel as few evaluations of such programmes exist, and the rigorous nature of the evaluation will provide multiple points of data for potential scale-up.
We would like to acknowledge the ADDO owners, shopkeepers and clinic staff for their time and participation in this study. We thank the Tanzanian Pharmacy Council for their support for our study, as well as Elizabeth Shekalaghe specifically, the CEO of the Tanzanian Pharmacy Council. We acknowledge the regional medical officers, regional pharmacists, the Regional Health Management Teams, and the Council Health Management Teams in Mwanza and Shinyanga for their support for this study. Further, we would like to thank the members of the Youth Advisory Boards for their contributions and feedback provided throughout the study timeline. We thank our study staff for their crucial contributions to this work, including Atuganile Kalinjila, Juliet Allan, Lisa Richard, Janeth Msasa, Zaburi Morris, Moza Albert, Janeth Maliga, Joyce Maboko, Victor Brown and Frank Joseph. We thank the Data Safety and Monitoring Board members for important feedback and considerations as we prepare for study launch. We thank Lauren Hunter for her important intellectual contributions to the pilot study which influence this study.
Ethics statements
Patient consent for publication
Not applicable.
Contributors AM, LAS and JL led the structuring, drafting and editing of this manuscript. AM, LAS, SIM, PFN, LJP, KH, CAS-B, WM, M-SKD, AS and JL contributed to the editing and revision of the manuscript. The study ideation and methodological design were led by JL, SIM and PFN.
Funding This work was supported by the US National Institutes of Health (National Institute of Mental Health: R01MH124516, PI: JL).
Map disclaimer The inclusion of any map (including the depiction of any boundaries therein), or of any geographic or locational reference, does not imply the expression of any opinion whatsoever on the part of BMJ concerning the legal status of any country, territory, jurisdiction or area or of its authorities. Any such expression remains solely that of the relevant source and is not endorsed by BMJ. Maps are provided without any warranty of any kind, either express or implied.
Competing interests None declared.
Patient and public involvement Patients and/or the public were involved in the design, or conduct, or reporting, or dissemination plans of this research. Refer to the Methods section for further details.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
1 UNAIDS. Women and HIV: a spotlight on adolescent girls and young women. UNAIDS; 2019.
2 Ministry of Health, Community Development, Gender, Elderly and Children (MoHCDGEC)[Tanzania Mainland], Ministry of Health (MoH) [Zanzibar], National Bureau of Statistics (NBS),Office of the Chief Government Statistician (OCGS), ICF International. Tanzania demographic and health survey and malaria indicator survey (TDHS-MIS) 2015-16. PR74. Dar es Salaam, Tanzania, and Rockville, Maryland, USA: MoHSW, MoH, OCGS, and ICF International; 2016. Available: http://dhsprogram.com/pubs/pdf/PR74/PR74.pdf
3 Brown K, Williams DB, Kinchen S, et al. Status of HIV epidemic control among adolescent girls and young women aged 15–24 years — seven African countries, 2015–2017. Morb Mortal Wkly Rep 2018; 67: 29–32. doi:10.15585/mmwr.mm6701a6
4 Kuzma EK, Peters RM. Adolescent vulnerability, sexual health, and the NP’s role in health advocacy. J Am Assoc Nurse Pract 2016; 28: 353–61. doi:10.1002/2327-6924.12331
5 Mutea L, Ontiri S, Kadiri F, et al. Access to information and use of adolescent sexual reproductive health services: qualitative exploration of barriers and Facilitators in Kisumu and Kakamega, Kenya. PLoS One 2020; 15: e0241985. doi:10.1371/journal.pone.0241985
6 Sila J, Larsen AM, Kinuthia J, et al. High awareness, yet low uptake, of pre-exposure prophylaxis among adolescent girls and young women within family planning clinics in Kenya. AIDS Patient Care STDS 2020; 34: 336–43. doi:10.1089/apc.2020.0037
7 Clement ME, Kofron R, Landovitz RJ. Long-acting injectable cabotegravir for the prevention of HIV infection. Curr Opin HIV AIDS 2020; 15: 19–26. doi:10.1097/COH.0000000000000597
8 Krovi SA, Johnson LM, Luecke E, et al. Advances in long-acting Injectables, implants, and vaginal rings for contraception and HIV prevention. Adv Drug Deliv Rev 2021; 176: 113849. doi:10.1016/j.addr.2021.113849
9 Waters H, Hatt L, Peters D. Working with the private sector for child health. Health Policy Plan 2003; 18: 127–37. doi:10.1093/heapol/czg017
10 Peters DH, Mirchandani GG, Hansen PM. Strategies for engaging the private sector in sexual and reproductive health: how effective are they Health Policy Plan 2004; 19 Suppl 1: i5–21. doi:10.1093/heapol/czh041
11 Hanson K, Berman P. Private health care provision in developing countries: a preliminary analysis of levels and composition. Health Policy Plan 1998; 13: 195–211. doi:10.1093/heapol/13.3.195
12 Hanson K, Ranson MK, Oliveira‐Cruz V, et al. Expanding access to priority health interventions: a framework for understanding the constraints to scaling-up. J of Intl Development 2003; 15: 1–14. doi:10.1002/jid.963
13 Bloom G, Standing H, Lucas H, et al. Making health markets work better for poor people: the case of informal providers. Health Policy Plan 2011; 26 Suppl 1: i45–52. doi:10.1093/heapol/czr025
14 Goodman C, Brieger W, Unwin A, et al. Medicine sellers and malaria treatment in sub-Saharan Africa: what do they do and how can their practice be improved Am J Trop Med Hyg 2007; 77: 203–18.
15 Liu J, Prach LM, Treleaven E, et al. The role of drug vendors in improving basic health-care services in Nigeria. Bull World Health Organ 2016; 94: 267–75. doi:10.2471/BLT.15.154666
16 Corroon M, Kebede E, Spektor G, et al. Key role of drug shops and pharmacies for family planning in urban Nigeria and Kenya. Glob Health Sci Pract 2016; 4: 594–609. doi:10.9745/GHSP-D-16-00197
17 Yadav P, Glassman A. Harnessing the power of private distribution channels for UHC. Center for Global Development; 2019. Available: https://www.cgdev.org/publication/harnessing-power-private-distribution-channels-uhc
18 Sudhinaraset M, Ingram M, Lofthouse HK, et al. What is the role of informal healthcare providers in developing countries? A systematic review. PLoS ONE 2013; 8: e54978. doi:10.1371/journal.pone.0054978
19 Awor P, Wamani H, Bwire G, et al. Private sector drug shops in integrated community case management of malaria, pneumonia, and diarrhea in children in Uganda. Am J Trop Med Hyg 2012; 87: 92–6. doi:10.4269/ajtmh.2012.11-0791
20 Awor P, Wamani H, Tylleskar T, et al. Increased access to care and appropriateness of treatment at private sector drug shops with integrated management of malaria, pneumonia and diarrhoea: a quasi-experimental study in Uganda. PLoS ONE 2014; 9: e115440. doi:10.1371/journal.pone.0115440
21 Treleaven E, Liu J, Prach LM, et al. Management of paediatric illnesses by patent and proprietary medicine vendors in Nigeria. Malar J 2015; 14: 232. doi:10.1186/s12936-015-0747-7
22 Ajayi IO, Falade CO, Adeniyi JD, et al. The role of patent medicine sellers in home management of childhood malaria: a situational analysis of experience in rural Nigeria. Int Q Community Health Educ 2002; 21: 271–81. doi:10.2190/569A-XLPX-YF5C-H9HU
23 Sieverding M, Beyeler N. Integrating informal providers into a people-centered health systems approach: qualitative evidence from local health systems in rural Nigeria. BMC Health Serv Res 2016; 16: 526. doi:10.1186/s12913-016-1780-0
24 Beyeler N, Liu J, Sieverding M. A systematic review of the role of proprietary and patent medicine vendors in healthcare provision in Nigeria. PLoS ONE 2015; 10: e0117165. doi:10.1371/journal.pone.0117165
25 Stanback J, Otterness C, Bekiita M, et al. Injected with controversy: sales and administration of Injectable contraceptives in drug shops in Uganda. Int Perspect Sex Reprod Health 2011; 37: 24–9. doi:10.1363/3702411
26 Stanback J, Miller R. Radical common sense: community provision of Injectable contraception in Africa. In: Kulczycki A, ed. Crit Issues Reprod Health. Netherlands: Springer, 2014: 265–84. Available: http://link.springer.com/chapter/10.1007/978-94-007-6722-5_13
27 Akol A, Chin-Quee D, Wamala-Mucheri P, et al. Getting closer to people: family planning provision by drug shops in Uganda. Glob Health Sci Pract 2014; 2: 472–81. doi:10.9745/GHSP-D-14-00085
28 Hoke TH, Wheeler SB, Lynd K, et al. “Community-based provision of Injectable contraceptives in Madagascar: “task shifting” to expand access to Injectable contraceptives”. Health Policy Plan 2012; 27: 52–9. doi:10.1093/heapol/czr003
29 Stanback J, Lebetkin E, Orr T, et al. Family planning provision in drug shops technical consultation participants. sale and provision of Injectable contraceptives in drug shops in developing countries: conclusions from a technical consultation. Contraception 2015; 91: 93–6. doi:10.1016/j.contraception.2014.11.007
30 Pharmacy Council. Kitabu Cha Mafunzo Kwa Watoa Dawa Katika Maduka Ya Dawa Muhimu [Book of Training for Drug Suppliers in ADDOs]. 1st edn. Dar es Salaam, Tanzania: Ministry of Health and Social Welfare, 2012.
31 Rutta E, Senauer K, Johnson K, et al. Creating a new class of pharmaceutical services provider for Underserved areas: the Tanzania accredited drug dispensing outlet experience. Prog Community Health Partnersh Res Educ Action 2009; 3: 145–53. doi:10.1353/cpr.0.0063
32 Rutta E, Liana J, Embrey M, et al. Accrediting retail drug shops to strengthen Tanzania’s public health system: an ADDO case study. J Pharm Policy Pract 2015; 8: 23. doi:10.1186/s40545-015-0044-4
33 Rutta E, Kibassa B, McKinnon B, et al. Increasing access to subsidized Artemisinin-based combination therapy through accredited drug dispensing outlets in Tanzania. Health Res Policy Syst BioMed Cent 2011; 9: 22. doi:10.1186/1478-4505-9-22
34 Alba S, Dillip A, Hetzel MW, et al. Improvements in access to malaria treatment in Tanzania following community, retail sector and health facility interventions - a user perspective. Malar J 2010; 9: 163. doi:10.1186/1475-2875-9-163
35 World Health Organization. Delivering HIV test results and messages for re-testing and counselling in adults. World Health Organization; 2010. Available: http://www.who.int/hiv/pub/vct/hiv_re_testing/en/
36 Doyle AM, Mavedzenge SN, Plummer ML, et al. The sexual behaviour of adolescents in sub-Saharan Africa: patterns and trends from national surveys. Tropical Med Int Health 2012; 17: 796–807. doi:10.1111/j.1365-3156.2012.03005.x
37 Mavedzenge SN, Luecke E, Ross DA. Effective approaches for programming to reduce adolescent vulnerability to HIV infection, HIV risk, and HIV-related morbidity and mortality: a systematic review of systematic reviews. J Acquir Immune Defic Syndr 2014; 66: S154–69. doi:10.1097/QAI.0000000000000178
38 Napierala Mavedzenge SM, Doyle AM, Ross DA. HIV prevention in young people in sub-Saharan Africa: a systematic review. J Adolesc Health 2011; 49: 568–86. doi:10.1016/j.jadohealth.2011.02.007
39 World Health Organization. WHO recommends HIV self-testing [Internet]. Geneva, Switzerland. WHO/HIV/2016.21. World Health Organization. Available: http://www.who.int/hiv/pub/vct/who-recommends-hiv-self-testing/en/
40 World Health Organization. Guidelines on HIV self-testing and partner notification: supplement to Consolidated guidelines on HIV testing services [Internet]. Geneva, Switzerland. 2016. Available: http://www.who.int/hiv/pub/vct/hiv-self-testing-guidelines/en/
41 Choko AT, Desmond N, Webb EL, et al. The uptake and accuracy of oral kits for HIV self-testing in high HIV prevalence setting: a cross-sectional feasibility study in Blantyre. PLoS Med 2011; 8: e1001102. doi:10.1371/journal.pmed.1001102
42 MacPherson P, Webb EL, Choko AT, et al. Stigmatising attitudes among people offered home-based HIV testing and counselling in Blantyre, Malawi: construction and analysis of a stigma scale. PLoS One 2011; 6: e26814. doi:10.1371/journal.pone.0026814
43 Napierala Mavedzenge S, Baggaley R, Corbett EL. A review of self-testing for HIV: research and policy priorities in a new era of HIV prevention. Clin Infect Dis Off Publ Infect Dis Soc Am 2013; 57: 126–38. doi:10.1093/cid/cit156
44 Napierala Mavdzenge S, Luecke E, Lopez A, et al. HIV Testing among Key Populations, Adolescent Girls and Men in Eastern and Southern Africa: A Review of Research, Policy and Programming. Research Triangle Park, NC: RTI International, 2016.
45 National Bureau of Statistics (NBS) [Tanzania], ICF Macro. Tanzania demographic and health survey 2010, Dar es Salaam, Tanzania. FR243. NBS and ICF Macro; 2011. Available: http://dhsprogram.com/pubs/pdf/FR243/FR243%5B24June2011%5D.pdf
46 Liu JX, Vallin J, Chiu C, et al. Designing for two: how enhancing human-centered design with behavioral Nudges unlocked breakthroughs to promote young women’s psychological safety and access to reproductive care in Tanzania. Soc Sci Med 2023; 320: 115683. doi:10.1016/j.socscimed.2023.115683
47 Hunter LA, McCoy SI, Rao A, et al. Designing drug shops for young women in Tanzania: applying human-centred design to facilitate access to HIV self-testing and contraception. Health Policy Plan 2021; 36: 1562–73. doi:10.1093/heapol/czab084
48 Hunter LA, Rao A, Napierala S, et al. Reaching adolescent girls and young women with HIV self-testing and contraception at girl-friendly drug shops: a randomized trial in Tanzania. J Adolesc Health 2023; 72: 64–72. doi:10.1016/j.jadohealth.2022.08.013
49 Curran GM, Bauer M, Mittman B, et al. Effectiveness-implementation hybrid designs. Med Care 2012; 50: 217–26. doi:10.1097/MLR.0b013e3182408812
50 United Republic of Tanzania. Ministry of Finance and Planning National Bureau of Statistics. Administrative Units Population Distribution Report, Dar es Salaam, Tanzania, Available: https://www.nbs.go.tz/nbs/takwimu/Census2022/Administrative_units_Population_Distribution_Report_Tanzania_volume1a.pdf
51 Ministry of Health (MoH) [Tanzania Mainland], Ministry of Health (MoH) [Zanzibar], National Bureau of Statistics (NBS), Office of the Chief Government Statistician (OCGS), and ICF. Tanzania demographic and health survey and malaria indicator survey 2022 key indicators report. Dodoma, Tanzania, and Rockville, Maryland, USA. MoH, NBS, OCGS, and ICF.
52 Hokororo A, Kihunrwa A, Hoekstra P, et al. High prevalence of sexually transmitted infections in pregnant adolescent girls in Tanzania: a multi-community cross-sectional study. Sex Transm Infect 2015; 91: 473–8. doi:10.1136/sextrans-2014-051952
53 SO AAA. Nutritional status and associated factors among adolescents living with HIV in Shinyanga municipal Council, Shinyanga. 2021.
54 Maisha Meds. Maisha Meds; 2023. Available: https://maishameds.org/
55 Moulton LH. Covariate-based constrained randomization of group-randomized trials. Clin Trials 2004; 1: 297–305. doi:10.1191/1740774504cn024oa
56 Yauck M, Moodie EEM, Apelian H, et al. Sampling from networks: respondent-driven sampling. Epidemiol Methods 2021; 10: 20200033. doi:10.1515/em-2020-0033
57 Crawford FW, Wu J, Heimer R. Hidden population size estimation from respondent-driven sampling: a network approach. J Am Stat Assoc 2018; 113: 755–66. doi:10.1080/01621459.2017.1285775
58 Tanzania Commission for, Zanzibar AIDS Commission AC. Tanzania HIV impact survey (THIS) 2016-2017: final report. Dar es Salaam, Tanzania. 2018.
59 Gail MH, Mark SD, Carroll RJ, et al. On design considerations and randomization-based inference for community intervention trials. Stat Med 1996; 15: 1069–92. doi:10.1002/(SICI)1097-0258(19960615)15:11<1069::AID-SIM220>3.0.CO;2-Q
60 Li F, Turner EL, Heagerty PJ, et al. An evaluation of constrained randomization for the design and analysis of group-randomized trials with binary outcomes. Stat Med 2017; 36: 3791–806. doi:10.1002/sim.7410
61 Campbell MK, Elbourne DR, Altman DG, et al. CONSORT statement: extension to cluster randomised trials. BMJ 2004; 328: 702–8. doi:10.1136/bmj.328.7441.702
62 Miles MB, Huberman AM. Qualitative Data Analysis: An Expanded Sourcebook. SAGE, 1994.
63 Tolley EE, Ulin PR, Mack N, et al. Qualitative Methods in Public Health: A Field Guide for Applied Research. John Wiley & Sons, 2016.
64 Proctor EK, Landsverk J, Aarons G, et al. Implementation research in mental health services: an emerging science with conceptual, methodological, and training challenges. Adm Policy Ment Health Ment Health Serv Res 2009; 36: 24–34. doi:10.1007/s10488-008-0197-4
65 Proctor E, Silmere H, Raghavan R, et al. Outcomes for implementation research: conceptual distinctions, measurement challenges, and research agenda. Adm Policy Ment Health Ment Health Serv Res 2011; 38: 65–76. doi:10.1007/s10488-010-0319-7
66 Maxwell JA. Qualitative Research Design: An Interactive Approach. Sage publications, 2012.
67 Dedoose version 8.0.35, web application for managing, analyzing, and presenting qualitative and mixed method research data [Internet]. Los Angeles, CA. SocioCultural Research Consultants, LLC; 2018. Available: www.dedoose.com
68 Navarrete MS, Adrian C, Bachelet VC. Respondent-driven sampling: advantages and disadvantages from a sampling method. Medwave 2022; 21: e8513. doi:10.5867/medwave.2022.01.002528
© 2024 Author(s) (or their employer(s)) 2024. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ. http://creativecommons.org/licenses/by-nc/4.0/ This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/ . Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
; McCoy, Sandra I 3
; Njau, Prosper F 4 ; Packel, Laura J 2 ; Kassim Hassan 1 ; Solorzano-Barrera, Camila 5
; Maokola, Werner 4 ; Mi-Suk Kang Dufour 6 ; Amon Sabasaba 1 ; Liu, Jenny 5 1 Health for a Prosperous Nation, Dar es Salaam, United Republic of Tanzania
2 Division of Epidemiology, School of Public Health, University of California, Berkeley, Berkeley, CA, USA; Institute for Health Aging, Department of Social and Behavioral Sciences, School of Nursing, University of California, San Francisco, San Francisco, California, USA
3 Division of Epidemiology, School of Public Health, University of California, Berkeley, Berkeley, CA, USA
4 Ministry of Health and Social Welfare, Dar es Salaam, United Republic of Tanzania
5 Institute for Health Aging, Department of Social and Behavioral Sciences, School of Nursing, University of California, San Francisco, San Francisco, California, USA
6 Division of Prevention Science, UCSF, San Francisco, California, USA