Prioritising place-based strategies for sexually transmitted and blood-borne infection prevention: protocol for the Zingatia Maskan interactive, convergent mixed-methods study in Nairobi, Kenya Authors: Leigh M McClarty, Apollo Kamau Gitau, Nicole Herpai, Joshua Kimani, Antony Karimi Kinyua, Anthony Kiplagat, Murugi Micheni, Festus Muriuki, Daisy Oside, Rhoda Kabuti Wanjiru, Shajy Isac, Lisa Lazarus, Robert R Lorway, Lyle R McKinnon, Sharmistha Mishra, Souradet Y Shaw, Parinita Bhattacharjee, Marissa L Becker, The Zingatia Maskan Study Team Published: 2026 ABSTRACT Abstract AbstractIntroductionAssociations between individual-level ‘risk behaviours’ and sexually transmitted and blood-borne infection (STBBI) acquisition are well documented; however, increasing evidence highlights the importance of place in shaping ecologies of STBBI risk and safety and access to relevant services. HIV prevention efforts have largely focused on population-based approaches resulting in siloed programming that ignores how sex work locations shape risk and safety for adolescent girls and young women (AGYW), female sex workers (FSW) and men who congregate in these spaces. Marked heterogeneity in STBBI prevalence and reported risk across sex work locations in Nairobi County, Kenya, underscores the need for more nuanced location-based prevention strategies. This paper describes the Zingatia Maskan (ZiMa) study, which aims to examine how physical, organisational, social and relational dimensions of sex work locations in Nairobi County shape ecologies of risk and safety for AGYW, FSW and their male sexual partners.Methods and analysisThe ZiMa study will employ an interactive, convergent mixed-methods design, implemented across three interlinked phases. Phase 1 will involve programmatic mapping to identify and describe sex work locations and estimate population sizes. Mapping findings will then inform purposive sampling for peer ethnography and key informant interviews that will subsequently explore social and relational dynamics within locations. Phase 2 will comprise a cross-sectional biobehavioural survey to assess STBBI outcomes, contextual determinants and related programme coverage. In phase 3, findings will be integrated iteratively across phases, with ongoing community engagement, through the ZiMa Community Advisory Board, informing refinement of data collection and interpretation.Ethics and disseminationThis study has received ethical approval from the Amref Ethics and Scientific Review Committee (Kenya) and the Committee for Harmonized Health Impact, Privacy, and Ethics Review (Manitoba, Canada). Study findings will be shared with community members, programme teams and study co-investigators through structured in-person meetings in Nairobi to support collaborative validation and interpretation of results and inform actionable programme refinements, and with national and subnational Technical Working Groups to inform policies. For academic audiences, findings will be presented at international conferences, meetings and workshops and published in open-access, peer-reviewed journals.Keywords: Adolescents, Sexually Transmitted Disease, HIV & AIDS, Preventive Health Services, Africa South of the Sahara STRENGTHS AND LIMITATIONS OF THIS STUDY This study uses an interactive, convergent mixed-methods design that integrates programmatic mapping, peer ethnography, in-depth interviews and a cross-sectional biobehavioural survey. This study uses an interactive, convergent mixed-methods design that integrates programmatic mapping, peer ethnography, in-depth interviews and a cross-sectional biobehavioural survey. Community engagement is embedded throughout all study phases through a dedicated Community Advisory Board and the involvement of community researchers. Community engagement is embedded throughout all study phases through a dedicated Community Advisory Board and the involvement of community researchers. Programmatic mapping across all 17 subcounties of Nairobi will support the development of population-specific sampling frames and probability-based survey sampling. Programmatic mapping across all 17 subcounties of Nairobi will support the development of population-specific sampling frames and probability-based survey sampling. The study combines location-level, network, behavioural and biological data to examine multiple dimensions of sexually transmitted and blood-borne infection (STBBI) risk and safety among three prioritised populations in Nairobi County. The study combines location-level, network, behavioural and biological data to examine multiple dimensions of sexually transmitted and blood-borne infection (STBBI) risk and safety among three prioritised populations in Nairobi County. As this is a cross-sectional study conducted in Nairobi County, findings will be context-specific and causal relationships cannot be inferred. As this is a cross-sectional study conducted in Nairobi County, findings will be context-specific and causal relationships cannot be inferred. Introduction Decades of research have established that sexually transmitted and bloodborne infections (STBBI), including HIV, are shaped not only by individual behaviours1–4 but also by broader social, structural and environmental contexts.5–7 While associations between individual-level ‘risk behaviours’ and STBBI acquisition are well documented, growing evidence highlights the importance of ‘place’ in shaping health outcomes and health inequities.7–10 Specifically, STBBI risk can be shaped by the contexts within which individuals come together and relate to one another and to their environment.8 11 12 Across HIV epidemics globally, including in Kenya, prevention efforts have historically used population-based approaches, prioritising key populations such as female sex workers (FSW). Comprehensive HIV prevention programmes for sex workers have demonstrated significant impact globally through combinations of community-led outreach and facility-based services.13–15 The most successful programmes are community-empowering; integrate behavioural, biomedical and structural components; and are delivered at scale via safe and non-stigmatising platforms.13 16–24 However, adolescent girls and young women (AGYW), including those under 18 years who sell sex, often face restricted access to services due to age-based eligibility criteria within sex worker-specific programmes and broader structural barriers. In particular, AGYW who sell sex or engage in transactional sex but do not self-identify as sex workers are frequently excluded from sex worker-focused programming, even though they are globally recognised as priority populations for STBBI prevention.25–30 Despite this prioritisation, service coverage for AGYW in sub-Saharan Africa remains limited.3 8–10 Existing programming for AGYW has often adopted broad, general population approaches rather than systematically identifying and prioritising subgroups at highest risk, thereby limiting coverage and effectiveness of prevention efforts.31–34 Additionally, STBBI prevention programmes for men, including those who purchase sex, are exceedingly rare in most global contexts.35–40 Men have been called ‘the HIV blind spot’ in sub-Saharan Africa due to the limited attention placed on addressing their health needs.41 Importantly, recent systematic reviews highlight the elevated risk of HIV40 and STBBI38 among men who purchase sex compared with men who do not. Opportunities exist to reach men through innovative programming.36 37 42 Globally, heterogeneity in STBBI burden is observed not only across populations but also across geographic locations. In Nairobi County, Kenya, approximately 39 000 FSW live and work,43 with an estimated average HIV prevalence of 14%.44 However, when examined by sex work location, that is, the physical spaces where women congregate to meet clients and other partners, HIV prevalence varies widely, ranging from 7% to 52% across the county, reflecting substantial geographic and contextual heterogeneity.45 Furthermore, our prior research in Mombasa identified distinct locations where AGYW engaged in transactional sex or sex work congregate alongside FSW.46 Compared with AGYW in the general population, those socialising in these shared locations reported substantially higher vulnerability to STBBI.47–49 These findings underscore the importance of understanding how environmental and ecological factors shape vulnerability. We conceptualise this through the framework of ecologies of risk and safety, which recognises that physical environments, social dynamics, organisational practices and relational networks interact to produce both vulnerability and protection.7 8 50–52 This framework highlights the need to consider interconnections between people and places over time in order to fully capture the complexities that perpetuate health inequities and shape lived experiences of STBBI-related risk and safety. Important gaps remain in how place is defined and operationalised in STBBI prevention research, as well as in understanding how interactions between people and places influence differential exposure to risk and access to services.8 53 54 While identifying where people meet can guide the strategic placement of public health programmes in locations of heightened risk, understanding how individuals are socially connected, and how and why they move through and gather in particular places, provides critical insight into how programming should be designed, delivered and prioritised. Recent work also demonstrates the connectedness between places where sex workers work and their associations with STBBI transmission.55–57 A more nuanced understanding of sex work locations is needed to inform location-based prevention strategies that optimise programme coverage and address unmet needs among AGYW, FSW and their male sexual partners (henceforth, ‘men’) who congregate in these spaces. Mixed-methods approaches are particularly well suited to this inquiry, as they allow for systematic characterisation of locations, quantification of health outcomes and service coverage, and in-depth exploration of social and relational dynamics that cannot be captured through survey methods alone. The Zingatia Maskan (ZiMa) study aims to examine how physical, organisational, social and relational dimensions of sex work locations in Nairobi County shape ecologies of risk and safety among AGYW, FSW and men. The overarching goal is to generate contextualised evidence to inform targeted, location-based STBBI prevention strategies. To achieve this aim, the ZiMa study will be conducted in three interlinked phases. Phase 1 will characterise sex work locations and explore their social, organisational and relational dynamics through programmatic mapping, sex work location characterisation and peer ethnography (objectives 1a and 1b). Phase 2 will employ a cross-sectional biobehavioural survey to examine STBBI prevalence, contextual determinants, affiliation networks and programme coverage (objectives 2 and 3). Phase 3 will integrate findings across study components through joint analyses and community-engaged interpretation. The specific objectives are: Characterise sex work locations where AGYW, FSW and men congregate for sexual partnerships.Identify and describe physical and organisational aspects of sex work locations through programmatic mapping with location characterisation.Explore social and relational aspects of sex work locations through peer-led ethnography. Characterise sex work locations where AGYW, FSW and men congregate for sexual partnerships. Identify and describe physical and organisational aspects of sex work locations through programmatic mapping with location characterisation. Identify and describe physical and organisational aspects of sex work locations through programmatic mapping with location characterisation. Explore social and relational aspects of sex work locations through peer-led ethnography. Explore social and relational aspects of sex work locations through peer-led ethnography. Examine contextual factors shaping experiences of risk and safety for AGYW, FSW and men congregating in sex work locations through a cross-sectional survey with biological data collection.Describe social support, economic security, unmet health needs and experiences of gender-based violence.Compare and contrast characteristics of affiliation network structure between AGYW, FSW and men.Measure prevalence of STBBI, including HIV, among AGYW, FSW and men. Examine contextual factors shaping experiences of risk and safety for AGYW, FSW and men congregating in sex work locations through a cross-sectional survey with biological data collection. Describe social support, economic security, unmet health needs and experiences of gender-based violence. Describe social support, economic security, unmet health needs and experiences of gender-based violence. Compare and contrast characteristics of affiliation network structure between AGYW, FSW and men. Compare and contrast characteristics of affiliation network structure between AGYW, FSW and men. Measure prevalence of STBBI, including HIV, among AGYW, FSW and men. Measure prevalence of STBBI, including HIV, among AGYW, FSW and men. Estimate and describe STBBI programme coverage among AGYW, FSW and men across sex work locations using the effective programme coverage framework through a cross-sectional survey. Estimate and describe STBBI programme coverage among AGYW, FSW and men across sex work locations using the effective programme coverage framework through a cross-sectional survey. Methods and analysis Study design An interactive, convergent mixed methods design58 will be used to examine interconnections between places, people and programmes that shape ecologies of risk and safety for AGYW, FSW and men. In this approach, quantitative and qualitative data collection and analyses interact across study phases, with iterative analysis informing ongoing refinement of data collection procedures.58 Study activities commenced in October 2025. Data collection is expected to continue through September 2027, with data analysis, integrated interpretation and knowledge mobilisation activities extending until 2030. Consistent with the study phases outlined above, phase 1 addresses objectives 1a and 1b, phase 2 addresses objectives 2 and 3 and phase 3 focuses on integration of findings. Objective 1a will use programmatic mapping to identify sex work locations, estimate population sizes for each participant group and characterise relevant features of these locations. Objective 1b will employ peer ethnography and key informant interviews to understand how these locations operate and are experienced by people spending time there. Objectives 2 and 3 will be addressed through a cross-sectional survey with a linked biological component to measure STBBI prevalence and behavioural, social and structural factors that shape STBBI outcomes and programme coverage across the study’s population groups. Together, these methods will enhance understandings of how place-based contexts intersect with individual and programmatic factors to shape risk and safety within the local STBBI prevention and care landscape across Nairobi County. Qualitative and quantitative methods will be integrated iteratively throughout the study using a phased approach, with findings from each phase informing subsequent data collection activities, and with ongoing community engagement throughout all phases of the research. In phase 1, results from programmatic mapping and sex work location characterisation will inform sampling strategies and guide location selection for subsequent qualitative and quantitative components. Qualitative findings will support the contextualisation of mapping data by uncovering how the local social and structural milieu shape risk and safety. Insights from this phase will also support the refinement of survey questions and operational definitions for phase 2. In phase 2, the biobehavioural survey will be developed using insights generated through both mapping and qualitative inquiry to ensure that questionnaire content reflects the local context and that the sampling frame aligns with location typologies and population estimates. Finally, phase 3 will involve intentional integration of quantitative and qualitative findings through joint analyses to interpret results within broader social and structural contexts. This community-engaged, convergent design will support a cohesive ecological understanding of how place-based characteristics, population dynamics and programme coverage interact to shape risk and safety. This study design is informed by a Program Science59–61 approach, which emphasises the use of embedded research to generate timely, context-specific evidence that can inform programmatic decision-making and adaptations to optimise population-level impact.59 62 Within this approach, the effective programme coverage framework62 provides an analytic tool to assess and address programme coverage gaps and maximise population-level impact. We will apply the framework to systematically examine where, and to what extent, gaps in STBBI prevention coverage exist for AGYW, FSW and men congregating in sex work locations across Nairobi County. Study setting This study will be conducted in Nairobi County, Kenya’s capital and largest urban centre, with an estimated 2023 population of 4.9 million, making it the most populous county in the country.63 Nairobi is a priority county for HIV prevention, with a 2022 epidemic appraisal suggesting that prevention programmes focused in Nairobi could avert more than 1000 new infections.64 65 The county hosts one of Kenya’s largest and most comprehensive HIV prevention programmes for key populations and AGYW. Programmatic mapping exercises have highlighted a dense landscape of sex work locations in Nairobi, identifying 2032 sex work locations in 2018.43 Recent estimates indicate that over 39 000 FSW work in Nairobi, representing approximately 24% of the country’s total estimated FSW population.66 According to the UNAIDS Sub-national HIV estimates in Priority Populations (SHIPP) tool,28 Nairobi was home to an estimated 515 886 AGYW in 2025, among whom 23.7% reported non-regular sexual partners and 2.8% reported selling sex. The same SHIPP estimates indicate that Nairobi is home to 1 371 913 adult men (aged 15–49 years), among whom 43.7% report having non-regular sexual partners.28 Partners for Health and Development in Africa (PHDA) implements Nairobi County’s largest HIV prevention programme for key populations, serving approximately 34 000 FSW through a network of peer educators and six Sex Worker Outreach Programme (SWOP) clinics. The SWOP clinics operate across 5 subcounties and are supported by extensive networks of peer educators and prevention officers across all 17 subcounties, ensuring coverage and accessibility for key populations across Nairobi. The ZiMa study will involve systematic assessment of sex work locations across all 17 subcounties of Nairobi. This work will be conducted in partnership with Kenya’s National Syndemic Diseases Control Council and Nairobi County, with joint leadership by PHDA and the Institute for Global Public Health at the University of Manitoba. Study population The study will include three groups: AGYW, FSW and men (table 1). Eligibility will be assessed using a brief screening tool, and numbers and reasons for refusal will be recorded. To be eligible for participation, individuals must congregate in sex work locations identified during programmatic mapping (objective 1a) and be capable and willing to provide informed consent. Individuals will be excluded if they are unable to provide informed consent, including those impaired by alcohol or drug use. All participants will be screened for eligibility and enrolled while congregating in mapped sex work locations. Table 1Participant group eligibility criteriaParticipant groupsEligibility criteriaAge (years)Sex and gender identitySexual behaviours and partnershipsAdolescent girls and young women (AGYW)15–24Cisgender womenHave been sexually active in the last 6 monthsFemale sex workers (FSW)≥15Cisgender womenHave exchanged sex for money or gifts in the last 3 monthsMale sexual partners (men)≥18Cisgender menHave been sexually active in the last 6 months with someone other than their marital/regular partner Participant group eligibility criteria Patient and public involvement: establishment of the Zingatia Maskan community advisory board Prior to commencing data collection, we will establish a study-specific Community Advisory Board to provide oversight and guidance to all three phases of the study. The Community Advisory Board will guide the identification of community researchers, including peer ethnographers and peer educators to facilitate data collection, advise on safety and security strategies for field teams and contribute to data analyses, interpretation of findings and integrated knowledge mobilisation activities. The Community Advisory Board will comprise representatives from each participant group, with at least four AGYW, four young sex workers (15–17 years), four adult FSW (18+ years) and four men from diverse location typologies and subcounties across Nairobi. Members will be identified in collaboration with PHDA, SWOP and peer educators to ensure broad representation and strong linkage with existing community structures. The Community Advisory Board will meet quarterly throughout the study and will be convened by members of the study team, including investigators from PHDA and the University of Manitoba. Additional ad hoc meetings may be convened, as needed, to support implementation planning, interpretation and dissemination of emerging findings, and development of ancillary studies, or address issues arising during study implementation. Meetings will generally be conducted with full Board membership to promote dialogue across participant groups and facilitate shared interpretation of findings, while separate consultations may be held with representatives of participant groups when deemed necessary to support participant comfort, confidentiality or discussion of population-specific issues. Community Advisory Board members will receive reimbursement for transportation and light refreshments at each meeting, in accordance with locally approved study procedures. Programmatic mapping, population size estimation and sex work location characterisation: study procedure and data analysis (objective 1a) We will conduct programmatic mapping to identify geographically defined sex work locations where individuals meet to buy or sell sex. This process will generate population size estimates for each participant group and provide a systematic characterisation of sex work locations across Nairobi County. Data collection procedures Our mapping approach has been well established in multiple contexts.46 66–68 The approach is community-led and implemented in two stages: (1) level one (L1) identification of sex work locations and (2) level two (L2) validation and characterisation of each location. Level 1 mapping: sex work location identification Level 1 begins with a list of locations, known to PHDA-SWOP, where FSW congregate to meet male sexual partners. Level 2 mapping: validation and sex work location characterisation Level 2 validates and characterises the sex work locations identified in level 1. Data collection pairs (one researcher and one peer educator) will systematically visit each sex work location to confirm its existence and document its name, typology (eg, bar with lodging, street, brothel, etc.) and approximate geographic location (eg, nearest intersection). Researchers will be PHDA-SWOP staff familiar with Nairobi’s sex work locations, and peer educators will be FSW working in or near locations who can facilitate safe access. At each location, researchers will purposively select two to three key informants, such as AGYW, FSW and gatekeepers including owners and staff of the location, to answer approximately 20 brief questions. Questions will generate information about the locations, including the number of AGYW, FSW and men present on typical and peak days; hours of operation; alcohol licensing; age restrictions; staffing structure (eg, security, managers, pimps/madams); local policing practices; presence of violence; and availability of prevention services (eg, condoms, lubricant, outreach). Key informants will also be asked about other nearby sex work locations that may have been missed in level 1. Any newly identified locations will be added to the level 1 master list and visited for validation and characterisation. Level 2 key informant interactions are expected to take approximately 10–15 min. Prior to initiating discussions, researchers will explain the purpose of the mapping activity and confirm the individual’s willingness to participate. Given the brief and informal nature of these discussions, and because no personal identifying information will be collected, formal consent procedures will not be required, and no remuneration will be provided. Where appropriate, mapping team members will purchase non-alcoholic refreshments while visiting sex work locations as a courtesy to the establishments and individuals hosting discussions. Data analysis Mapping data will be triangulated and analysed to describe the distribution of locations by subcounty, categorise typologies of locations, estimate the size of each participant group at locations and characterise overlap of groups within and across identified locations. Data from level 2 will be compiled to describe characteristics and attributes of the physical locations, including availability of prevention products and services. Descriptive summaries will be generated, describing patterns of sexual partnerships at various location typologies. In addition to describing the distribution and characteristics of sex work locations across Nairobi County, objective 1a provides the empirical foundation for objectives 2 and 3. Population size estimates will be used to develop the sampling frames, inform probability-proportionate-to-size sampling procedures and ensure representation across relevant location typologies and sub-counties. Peer-led ethnography: study procedure and data analysis (objective 1b) Peer ethnography69 and in-depth interviews will be used to explore the social contexts, norms and practices (eg, gendered power dynamics, safety and ‘risk’ behaviours) shaping interactions within sex work locations mapped in objective 1a. This qualitative component will enhance understanding of how people who congregate in sex work locations navigate and experience the spaces and how these interactions shape local ecologies of risk and safety. Peer ethnography involves trained community members who systematically observe, document and reflect on behaviours and social dynamics within their own networks.69 As trusted insiders, peer ethnographers can more readily access sex work locations and draw on the tacit knowledges70 necessary to interpret social norms, practices and relational dynamics within these environments.71 72 Data collection procedures Nine peer ethnographers will be organised into three teams, each including one representative from each participant group. Peer ethnographers will be nominated through consultations with the study Community Advisory Board and SWOP staff and will participate in a 1-week training covering community-based research ethics, ethnographic methods (participant observation and field note writing73 74) and in-depth interviewing. Data collection guides for observation, field notes and interviews will be refined during training and finalised at its conclusion. Following training, ethnographers will conduct 11 weeks of field work across eight purposively selected sex work locations in selected subcounties, with two teams conducting fieldwork in three locations and one team working in two locations. Location selection will be informed by mapping findings and consultations with the Community Advisory Board and mapping teams to ensure diversity in location typologies, demographic characteristics of people who congregate there and physical and organisational characteristics of the locations. During field work, ethnographers will prepare field notes with ‘thick descriptions’75 of each location, documenting social interactions, spatial dynamics and other contextual features. Teams will meet weekly with the study’s social science leads to debrief, discuss emerging observations and refine their contextual understanding. Towards the end of the 3 weeks of fieldwork in each location, teams will conduct three to four in-depth interviews with key informants (total n=24–36), including individuals working in or around sex work locations (eg, waitstaff, security personnel, owners or managers, boda boda drivers and other local knowledge holders). Interviews will explore social and relational dynamics within and surrounding sex work locations and the broader environments in which they operate. Written, thumbprint or verbal informed consent will be obtained prior to participation. With participant consent, interviews will be audio-recorded, transcribed and translated for analysis. Interviews are expected to last 60–75 min. Participants will receive a standard honourarium of Kenyan Shillings (KES) 1000 (approximately $C11) to compensate for their time and transport costs. Data analysis Following data collection, peer ethnographer teams will be trained in a simplified qualitative analysis process that adopts a meaning-centred, interpretivist approach developed in our prior work.72 76 With technical guidance from the study’s social science leads, peer ethnographers will manually construct coding trees composed of interlinked parent-child nodes.77 78 Parent nodes will be defined collaboratively and iteratively reconciled among analysts to produce a coherent and unified coding framework. Memo-writing and analytic notetaking will be integrated throughout the process to help identify emergent themes, which will be used to refine and finalise code definitions. The unified coding system will then be applied to the full dataset of interview transcripts and ethnographic field notes using NVivo 14. To protect confidentiality, participant names, location identifiers and any other potentially identifying information will be removed during analysis. Biobehavioural survey: study procedure and data analysis (objectives 2 and 3) Objectives 2 and 3 will be addressed through a cross-sectional biobehavioural survey. This component integrates structured data across key demographic, social, economic, behavioural and health-related domains, alongside STBBI testing, to provide a multidimensional understanding of how contextual factors shape heterogeneity in STBBI acquisition and programme coverage among AGYW, FSW and men within and across mapped locations. Sampling and recruitment A multistage, random sampling strategy will be used to obtain representative samples of AGYW, FSW and men. Stage 1 involves random selection of locations; stage 2 involves random selection of participants within those locations. Mapping data will generate three sampling frames (one per participant group) comprising all validated locations identified during the mapping exercise. In stage 1, a fixed number of locations per participant group will be randomly selected from the validated list of locations, stratified by subcounty and location typology. Location selection will use probability-proportionate-to-size, based on population size estimates from mapping and location characterisation. In stage 2, each selected location will be assigned a random week, day and window of time for participant enrolment. Trained community members (‘social mobilisers’) will identify individuals present during that window, randomly approach every Xth person to screen for eligibility, and then enrol eligible participants following informed consent. If an individual declines participation, the next eligible person will be approached, and reasons for refusal will be systematically documented. The survey sampling frame will be developed using the programmatic mapping data generated in objective 1a. Sample sizes for each participant group will be calculated to estimate HIV prevalence using the following formula: where p denotes estimated HIV prevalence, DEFF represents the design effect associated with the sampling strategy and d is the desired precision. Calculations will assume 95% confidence (z-score=1.96), a design effect of 1.5 to account for the stratified random sampling design and desired precision levels of ±4% for FSW, ±2% for AGYW and ±1.5% for men. Prevalence estimates used in these calculations will draw on data from our previous work (14% among FSW in 202379 and 5.5% among AGYW in 201880) and recent estimates for the general population of cisgender men (2.8% in 202481). To maximise inclusion of younger participants who sell sex, recruitment and screening procedures will be implemented by trained community-based researchers with experience working with FSW and AGYW in Nairobi County. Eligibility screening questions will focus on behaviours and partnership characteristics, rather than identity labels, recognising that some oversampling through locations selected from the FSW sampling frame is expected to increase opportunities to enrol younger participants engaged in sex work who may otherwise be under-represented. To achieve an adequate sample of young sex workers, under 18 years old, an additional 1–2 FSW will be randomly selected at each sampled sex work location, with a target of enrolling an additional 150–200 participants. To account for anticipated non-response, sample sizes will be inflated by 20%, consistent with previous field experience. The final rounded sample sizes will be 760 FSW, 899 AGYW and 838 men, totaling N=2497 participants (table 2). Table 2Target sample sizes for each participant groupP valueSample sizeOversampling for young FSW, <18 yearsFinal sample size, adjusting for 20% non-responseFemale sex workers (FSW)0.14434200760Adolescent girls and young women (AGYW)0.055749–899Male sexual partners (men)0.028698–838Total 1881 2497 Target sample sizes for each participant group Data collection procedures Eligible participants will complete both components of the biobehavioural survey, including finger-prick blood, venous whole blood and urine samples. Written, thumbprint or verbal informed consent will be obtained from all participants before enrolment and data collection. To maintain confidentiality, each participant will receive a unique, non-identifying alphanumeric study identification (ID) used to label survey tools and biological specimens. Survey participation, including questionnaire and biological specimen collection, is expected to take up to 120 min. Participants will receive a standard honorarium of KES 1000 (approximately $C11) to compensate for their time and transportation costs. After informed consent, trained researchers will administer face-to-face surveys at or near sex work locations. Surveys will capture information on informal and formal social support,82 mental health,83 substance use,84 economic security,85 gender and social norms, gender-based violence, stigma, sexual and reproductive health needs, sociodemographic characteristics, sexual partners, use of programme services and use of other sexual and social networking locations. For affiliation network analyses, participants will report where they met sex partners in the past 6 months. Reported locations will be categorised into broad typologies (eg, virtual, indoor, outdoor) drawing on a previously established framework86 adapted to the Kenyan context through our prior work43 44 46 and community and Community Advisory Board consultations. Standardised, locally validated survey items will be adapted from national and international instruments (eg, Integrated Bio-Behavioural Survey;87 Kenya Demographic and Health Survey 2022;88 Alcohol, Smoking and Substance Involvement Screening Test;89 Patient Health Questionnaire-9;84 Multidimensional Scale of Perceived Social Support83) and our prior studies.79 80 90 All instruments will be field-tested for clarity, cultural and contextual relevance, and acceptability of biological sample procedures. Data will be entered directly into tablets using SurveyCTO91 and transmitted daily via a secure virtual private network to a central server housed at PHDA for real-time monitoring and quality control. Following the survey, participants will provide blood and urine samples. HIV testing and pre-test and post-test counselling will follow Kenya’s national algorithm.92 Venous blood will be used for confirmatory HIV testing, per national guidelines,92 and viral load quantification. Syphilis testing will follow the national tiered algorithm using Rapid Plasma Reagin screening and Treponema pallidum Hemagglutination Assay confirmation. Urine specimens will be tested for Chlamydia trachomatis, Neisseria gonorrhoeae, Trichomonas vaginalis, Haemophilus ducreyi, herpes simplex virus types 1/2, Mycoplasma spp. and Ureaplasma spp. using the HybriSpot 12 platform (Master Diagnóstica, Spain).93 All specimen collection, transport and laboratory procedures will follow the PHDA laboratory standard operating procedures, including internal quality assurance and participation in external quality assessment protocols. Samples will be transported daily to the PHDA laboratory and regularly processed in batches. Participants may provide contact information to receive results. Regardless of test results, FSW participants will be linked to SWOP clinics for comprehensive treatment and combination prevention services; AGYW and men will be referred to Nairobi County public health facilities. Participants unwilling to provide contact information may collect results at any SWOP clinic using their study ID. Through the consent process, participants may opt into long-term storage of specimens for up to 5 years. For those who consent, remnant samples will be stored at −80 °C at PHDA for resolving discrepant results or use in future research. Stored specimens will be linked to behavioural data but will not be linked to personal identifiers. Objective 2a analysis plan Descriptive univariate and bivariate analyses will be conducted for each participant group to assess social and economic stability, sexual partnerships, sexual and reproductive health needs and experiences of gender-based violence by location and other key variables (eg, age, socioeconomic status, marital status, sex/gender, location typology, subcounty, age at first sex). These analyses will enable examination of inequalities within and between groups and assessment of clustering by location, typology and subcounty. All analyses will apply standardised weights to account for the stratified sampling design. Objective 2b analysis plan Affiliation network analyses will begin by visualising networks based on named locations. STBBI prevalence will then be calculated for each named or sampled location, or by location typology, depending on observed prevalence patterns. Standard network measures, including centrality (degree, eigenvalue/eigenvector-based metrics) and degree (connectedness, compactness), will be generated and compared across participant groups. Where feasible, network data will be linked to location characterisation data to explore associations among location features, participant characteristics and service coverage. Given the sensitivity of location information in criminalised and stigmatised contexts, strict ethical safeguards will be applied. All named locations will be anonymised and assigned coded identifiers (eg, ‘Bar with lodging 1,’ ‘Street 2’), and only anonymised labels will appear in outputs. Map-based data visualisation will be developed in consultation with the Community Advisory Board and community members, for example, aggregated choropleth mapping may be used instead of dot maps, and if dot mapping is deemed acceptable, locations within Nairobi subcounties will be randomly displaced within subcounties to protect confidentiality. Affiliation network analyses will be conducted using UCINET V.6 and R. Objective 2c analysis plan STBBI prevalence will be calculated for all participant groups. Biological samples will be anonymously linked to survey data using study IDs to explore associations among location and participant characteristics, programme coverage and STBBI prevalence. Stratified analyses will be conducted by key location-level and participant-level variables and programme coverage. For each infection, logistic regression models will estimate crude associations with participant group, programme coverage and location characteristics, generating ORs and 95% CIs. Multivariable logistic regression models will then assess the independent effects of these variables on each infection, adjusting for key sociodemographic factors (eg, age, education). Adjusted ORs (AORs) and 95% CIs will be reported. Interaction terms (eg, participant group*programme coverage; participant group*location) will be tested using maximum likelihood methods (p<0.05). All quantitative analyses will be conducted using Stata V.19 (College Station, Texas). Objective 3 analysis plan The survey will measure coverage of STBBI prevention and treatment interventions offered through the national key population programme in Nairobi County, including condom and lubricant provision, STBBI testing and treatment (including syndromic management), violence prevention and response, pre-exposure and post-exposure HIV prophylaxis, antiretroviral therapy and mental health supports. Survey items will be structured to generate variables required for the programme coverage cascade element of the effective programme coverage framework,62 which measures four coverage indicators: (1) required coverage, (2) availability coverage, (3) contact coverage and (4) utilisation coverage. Cascade analyses will characterise programme coverage, identify groups and locations with persistent coverage gaps and support hypothesis generation on potential drivers of these gaps. Coverage cascades will be constructed for each intervention and participant group. Indicators will be calculated as proportions of the total sample or relevant subgroups and calculated independently of preceding cascade steps (ie, participants need not appear in earlier steps to be included in subsequent ones). To examine inequalities, cascade indicators will be disaggregated by primary location and participant characteristics and visualised using equiplots.94 All cascade analyses will be conducted in Stata V.19. Integration of mixed methods findings and study procedures Integration will occur throughout the ZiMa study and not solely during the final analytic phase. Findings from programmatic mapping and sex work location characterisation (objective 1a) will inform sampling for peer ethnography (objective 1b) and the biobehavioural survey (objectives 2 and 3), while mapping and qualitative findings will be used to refine survey domains, operational definitions, questionnaire content and data collection procedures. These iterative linkages will ensure that subsequent study activities are responsive to emerging evidence and in alignment with local contexts and lived experiences. Phase 3 will build on this iterative integration by formally integrating findings from all study objectives using a convergent mixed-methods approach.58 Quantitative findings related to location characteristics, affiliation networks, STBBI prevalence and programme coverage will be compared and interpreted alongside qualitative findings describing social, organisation and relational aspects of sex work locations. Integration will occur through iterative team discussions, development of joint displays58 95 and triangulation of findings to identify areas of convergence, complementarity and divergence across data sources. Preliminary integrated findings will be reviewed with the ZiMa Community Advisory Board, community researchers and programme partners through structured interpretation workshops. These discussions will support contextualisation of findings, validation of emerging interpretations, identification of alternative explanations and development of actionable recommendations for place-based STBBI programming in Nairobi County. Insights arising from these workshops will be documented and incorporated into final analyses and knowledge mobilisation activities. Discussion The ZiMa study uses an interactive, convergent mixed-methods design to understand how the physical, organisational, social and relational dimensions of place shape ecologies of risk and safety within sex work locations for AGYW, FSW and men in Nairobi County. This approach leverages the complementary strengths of qualitative and quantitative data to triangulate findings and deepen inference: qualitative data elucidate lived experience, power dynamics and norms, while quantitative data measure prevalence, patterns and programme coverage across settings. Mixed-methods designs are well suited to criminalised and stigmatised contexts where research must be grounded in community knowledge and contextual realities.96 By combining community-informed qualitative inquiry with systematic quantitative measurement, this approach supports the generation of evidence grounded in lived realities and strengthens its relevance for programme and policies. Nairobi County has the highest estimated number of FSW in Kenya66 and a high general population HIV burden,64 making it a strategically important context for STBBI prevention efforts. Although a well-established prevention programme serves FSW, AGYW and men who congregate in sex work locations remain insufficiently integrated into programming. Recent reductions in external funding, including the US government stop-work order in January 2025, further underscore the need for efficient service delivery models.97 In this context, place-based programming approaches that reach multiple population groups congregating within shared environments offer a strategy to optimise coverage and address unmet prevention and treatment needs. Limitations The study shares the inherent limitations of cross-sectional designs, including the inability to infer causal relationships from data collected at a single time point. Self-reported data from interviewer-administered surveys and in-depth interviews may be affected by social desirability and under-reporting biases for criminalised or stigmatised behaviours. Asking participants to recall past experiences and behaviours may also introduce recall bias. Because sampling is limited to physical sex work locations, individuals who work exclusively online or in less visible spaces may be under-represented, and the dynamic nature of sex work locations may lead to temporally limited mapping findings. Although the qualitative phase relies on small, purposely selected samples and is not intended to be generalisable, it will yield rich, contextual insights into lived experiences. These insights are critical for informing tailored, context-specific interventions for priority populations in urban sex work settings. HIV and STBBI programmes in Kenya have historically been donor-funded and organised as population-specific interventions, despite evidence of overlapping sexual networks and shared environments among AGYW, FSW and men.46 Focusing on locations where these groups meet to form sexual partnerships offers opportunities to better understand how place shapes risk, safety, service access and programme coverage. The ZiMa study will generate critical, context-specific evidence on the physical, organisational, social and relational dimensions of sex work locations in Nairobi County and their associations with STBBI outcomes and related service coverage. This work aims to inform the development and evaluation of place-based approaches to STBBI prevention in Kenya and other settings where similar dynamics exist. Ethics and dissemination Ethics approvals and considerations Ethical approval has been obtained from the Amref Ethics and Scientific Review Committee (AMREF-ESRC P1936/2025) in Kenya and the Committee for Harmonized Health Impact, Privacy, and Ethics Review (HS26996) in Manitoba, Canada. A research permit was also granted by Kenya’s National Commission of Science, Technology and Innovation (NACOSTI/P/25/4179107). For the biobehavioural survey and qualitative interviews, written, thumbprint or verbal consent will be obtained prior to participation. Consistent with Kenyan guidelines for adolescent HIV and sexual and reproductive health research98 and approvals provided by the relevant ethics committees, participants aged 15–17 years who meet the criteria for mature minor status will be permitted to provide consent on their own behalf. Mature minors are defined as adolescents aged 15 years or older who live independently of parents or guardians and manage their own affairs. Parental permission will not be required for these participants. For level 2 programmatic mapping, researchers will conduct brief, informal discussions with key informants regarding characteristics of sex work locations. Researchers will explain the purpose of the activity and confirm willingness to participate prior to initiating discussions. No personal identifying information will be collected; formal consent procedures are not required. Regardless of participation, all individuals approached for enrolment will be offered linkage to appropriate health facilities for prevention and treatment services, and participation decision will not affect access to services through PHDA-SWOP clinics or public facilities. Participants in the study’s qualitative and survey components will receive standard honorarium of KES 1000 (approximately $C11) to compensate time and travel costs. All required approvals from the Government of Kenya will be obtained prior to the transfer of data between collaborating institutions (University of Manitoba and PHDA). Dissemination of results Findings from each study phase will be shared with ZiMa Community Advisory Board members; community researchers; PHDA teams; SWOP’s peer educators, programme team and clinical team; and co-investigators through structured in-person meetings. These sessions will support collaborative validation and interpretation of results and inform actionable programme refinements, while also strengthening research capacity among community members. Results will be disseminated to Nairobi County, the National Syndemic Diseases Control Council and the National AIDS and STI Control Program Technical Working Groups to inform county-level and national-level strategies and policies. For research communities, findings will be presented at international conferences, meetings and workshops and submitted as manuscripts to peer reviewed journals. Acknowledgements We are deeply grateful to our colleagues and community members whose time, lived experience and expertise ground this study: Lydia Achieng, Brenda Adhiambo, Sophie Andati, Peter Arimi, Rosemary Awino, Wendy Awino, Mitzy Gafos, Mitchel Halima, Chidumebi Idemili, Jacob, Zaituna Khamisi, Bernadette Kina Kombo, Teckler Lisa, Huiting Ma, Stephen Macharia, Byrone Mbilu, Celestine Mugambi, Faith Mumbua, Helgar Musyoki, Carol Ngunu, Mwanaisha Njeri, Emily Nyariki, Linet Saya and Samuel Tony. Their collaboration strengthens every stage of the ZiMa study. Footnotes Funding: This work is supported by the Canadian Institutes of Health Research [funding reference numbers: PJT-196046 and 197851]. Prepublication history for this paper is available online. To view these files, please visit the journal online (https://doi.org/10.1136/bmjopen-2026-121267). Patient consent for publication: Not applicable. Provenance and peer review: Not commissioned; externally peer reviewed. Patient and public involvement: Patients and/or the public were involved in the design, conduct, reporting or dissemination plans of this research. Refer to the Methods section for further details. Collaborators: Lydia Achieng, Brenda Adhiambo, Sophie Andati, Peter Arimi, Rosemary Awino, Wendy Awino, Mitzy Gafos, Mitchel Halima, Chidumebi Idemili, Jacob, Zaituna Khamisi, Bernadette Kina Kombo, Teckler Lisa, Huiting Ma, Stephen Macharia, Byrone Mbilu, Celestine Mugambi, Faith Mumbua, Helgar Musyoki, Carol Ngunu, Mwanaisha Njeri, Emily Nyariki, Linet Saya, and Samuel Tony. References