INTRODUCTION

South Africa, like many countries, is striving to achieve the Joint United Nations Program on HIV/AIDS (UNAIDS) 95–95–95 targets: ensuring that 95% of people living with HIV know their status, 95% of those diagnosed receive antiretroviral therapy (ART), and 95% of those on ART achieve viral suppression¹. Achieving these targets requires inclusive and targeted strategies, particularly for key populations who remain disproportionately affected by HIV. This includes ‘Lesbian, Gay, Bisexual, Transgender, Queer, Intersex, Asexual, and other sexual and gender minority groups’ (LGBTQI+). The routine collection of sexual orientation and gender identity (SOGI) data is increasingly recognized as essential for addressing health disparities among LGBTQI+ individuals, monitoring rights-based policy implementation, and informing equitable health responses^2-4. The South African National Strategic Plan (NSP 2023–2028) for HIV, TB, and STIs prioritizes five key populations: female sex workers (FSW) and their clients, transgender (TG) and gender-diverse individuals, gay men and other men who have sex with men (MSM), people who use drugs (PWUD), and those in prisons or closed settings⁵.

According to the Thembisa Model version 4.8, South Africa has achieved 95% diagnosis, 82% ART coverage, and 92% viral suppression⁶. While the overall national HIV prevalence is estimated at 12.7%, it is significantly higher among key populations, with reported prevalence rates of 55% among PWUD, 40% among TG women, 56% among FSW, and 27% among MSM^6,7. The Botshelo Ba Trans study highlighted the disproportionate burden among transgender women, with HIV prevalence estimates ranging from 37% in Cape Town and 50% in Buffalo City to 63% in Johannesburg⁷. In response, the UNAIDS Global AIDS Strategy 2021–2026 calls for 90% of transgender people to have access to HIV services, integrated with mental health support, gender-affirming care, services for intimate partner violence and sexual and gender-based violence, and access to post-exposure prophylaxis, emergency contraception, and psychological first aid⁸.

In South Africa, gender-affirming healthcare services remain a challenge, with limited provision of these services within the public health system. In October 2021, the South African Clinicians Society publicized much-needed guidelines for the provision of gender-affirming healthcare⁹. The guidelines address significant gaps in knowledge and skills of healthcare providers in providing services to transgender and diverse persons, although hormones are outlined as part of the essential drug list; the service is only available at the tertiary-level hospitals⁹.

Despite this progress, routine health information management systems (RHIMS) in South Africa do not disaggregate data by key population groups, including LGBTQI+ individuals. This makes it difficult for program managers and policymakers to allocate resources effectively or tailor interventions to meet their needs. Collecting SOGI data in healthcare settings is critical for identifying and addressing health disparities, particularly among TG individuals. A study by Mulemfo et al.¹⁰ in South African PHC facilities in Gauteng reported frequent misgendering and administrative erasure, with forms limited to binary options and staff assuming gender based on appearance rather than self-identification. These findings are aligned with the study by Ruberg and Ruelos¹¹ which reported that traditional demographic categories often fail to capture the complexity and fluidity of transgender identities, highlighting the need for inclusive and flexible data practices that respect self-identification. These limitations undermine efforts to provide responsive care, support monitoring and evaluation, guide public health planning, and inform policy decisions affecting gender-diverse individuals¹².

The country uses three interlinked electronic registers (TIER.Net) to capture patient-level HIV information at the facility level, and it is integrated with the district health information system (DHIS-2) for reporting data from the sub-district to the national level¹³. However, TIER.Net currently lacks fields for collecting SOGI information and does not include unique identifier codes for key populations, resulting in HIV data that cannot be disaggregated by key population group. To address these gaps, experts have recommended the adoption of a two-step method, which involves asking about both current gender identity and sex assigned at birth, as a best practice for accurately capturing transgender identities¹⁴. Some countries have adopted a two-step question approach to enable the collection and documentation of gender identity information^15-18. This approach involves first asking individuals about their current gender identity, followed by a question about the sex assigned at birth¹⁹. By prioritizing gender identity as the initial question, this method underscores its significance from the perspective of transgender individuals, affirming their lived identity before referencing assigned sex at birth. This approach allows for greater inclusivity, reduces misclassification, and reflects the diverse lived experiences of gender-diverse individuals^2,14,20.

The need to strengthen the government’s RHIMS has become urgent, particularly following the termination of PEPFAR-funded key populations programs in South Africa, due to the global freeze on US-funded aid announced in January 2025²¹. As external support diminishes, national ownership of key populations data collection anchored in government-managed tools and robust monitoring and evaluation systems capable of disaggregating data by key populations is critical for sustaining service delivery, ensuring accountability, and guiding effective resource allocation.

Countries such as Kenya and Ghana offer practical lessons for integrating key populations’ unique identifier codes into RHIMS. The two countries’ success was driven by strong political will, investment in digital infrastructure, training of healthcare workers, and active engagement with the key population community to build trust and reduce stigma^22-24. South Africa could adopt similar strategies by embedding unique identifier codes into existing platforms like TIER.Net or the national health insurance system while ensuring inclusive design through collaboration with key population-led organizations. Hence, understanding the current structure and variables within TIER.Net is essential for assessing how the system can be adapted to collect those missing variables, reduce misclassification, and support person-centered care. Within this context and framework, this study aimed to analyze the structure and content of routine HIV program datasets within the South African RHIMS for the inclusion of Key Populations Unique Identifier Codes.

METHODS

RESULTS

DISCUSSION

As countries work toward achieving the 95–95–95 HIV targets and ending the HIV/AIDS epidemic as a public health threat by 2030, strengthening RHIMS is essential for robust epidemiological analysis, identifying service delivery gaps, and supporting differentiated HIV care tailored to the needs of key populations. The observed patterns in HIV positivity, retention, and viral suppression are broadly consistent with findings reported in the Limpopo. According to the Limpopo Department of Health Annual Performance Plan 2024/25, the HIV test positivity rate was 0.5% among children aged 12–59 months, 1.1% among children aged 5–14 years, 1.2% among youth aged 15–24 years (excluding ANC), 2.7% among adults aged 25–49 years (excluding ANC), and 4.4% among adults aged ≥50 years (excluding ANC). Retention on ART at 12 months stood at 65.2% among adults and 75.9% among children²⁵. The province recorded an 84% viral load testing rate at 12 months, with 89% of those tested achieving viral suppression²⁵. Similar trends of relatively low HIV positivity, high ART initiation, and suboptimal retention at 12 months have been documented, underscoring persistent challenges in sustaining long-term engagement in care. The comparatively lower retention observed in this study may reflect contextual differences in service delivery or limitations inherent in routine health information systems, particularly in tracking patient continuity across facilities. Viral suppression outcomes are largely comparable to provincial estimates, indicating effective treatment among individuals retained in care. However, the persistence of sex-based disparities in treatment outcomes is consistent with existing evidence and highlights the need for targeted, differentiated interventions. Collectively, these findings emphasize the importance of strengthening retention strategies and enhancing routine data systems to improve longitudinal monitoring of HIV outcomes.

The continued reliance on binary sex and gender fields in RHIMS renders transgender individuals largely invisible within health data systems, contributing to inappropriate clinical decision-making and missed opportunities for essential preventive screenings for those with non-binary or transitioned anatomy⁴. A study by Mulemfo et al.¹⁰ found that healthcare workers in South African primary healthcare facilities often assumed patients’ gender based on appearance, contributing to misgendering, the exclusion of LGBTQI+ identities in health records, and limited access to appropriate services.

This study assessed the opportunities for incorporating key populations’ unique identifier code and gender identity into the national RHIMS to strengthen the monitoring of the HIV care cascade in South Africa. Our findings confirmed that the current health information systems cannot capture key population identifiers, making it difficult to track continuity of care and assess true engagement across the HIV care cascade. This undermines programmatic planning, limits responsiveness to equity gaps, and prevents accurate monitoring of differentiated HIV services. Our study found that RHIMS can currently disaggregate HIV cascades only by sex and age groups. Meanwhile, in South Africa, estimates suggest there are 87214 transgender women, 28065 transgender men, and 66076 gender non-conforming individuals²⁷, who remain largely invisible in routine data systems. While it is well established that key populations carry a higher HIV burden than the general population, routine data-derived HIV cascades are frequently based on proxy indicators, making it impossible to track individuals across the cascade without the use of unique identifier codes¹⁵. This deficiency is particularly concerning given the intersectional vulnerabilities of key populations to HIV, which are exacerbated by stigma, discrimination, and other health-seeking barriers²⁸. According to the Ritshidze²⁹ report, 75% of the key populations who participated in the community-led monitoring survey indicated they accessed public health facilities, but the data collection tools at these facilities could not indicate their key populations status. In trying to assist countries in closing this gap, the WHO released consolidated strategic information guidelines focusing on improving individual-level routine data to strengthen person-centered services for HIV and other related infections⁴. Yet the current South African health information systems still have critical limitations.

This significant disparity necessitates further disaggregation of HIV outcomes by key populations and LGBTQI+ subgroups to identify hidden disparities and ensure equitable care. Global guidance and emerging best practices support the inclusion of SOGI data in RHIMS to improve outcome tracking and health equity³⁰. Studies have emphasized the significance of disaggregating HIV viral suppression outcomes by sex assigned at birth, current gender identity, and sexual orientation, as this approach is crucial for identifying hidden disparities, enhancing clinical decision-making, and promoting equitable, person-centered HIV care among sexual and gender minority populations¹⁷. A study conducted by Shi et al.¹⁶ reported that cisgender sexual minority women had a significantly higher risk of HIV virological failure compared to cisgender heterosexual women, further reinforcing the need for disaggregated data to address inequities in HIV treatment outcomes¹⁶. The Botshelo Ba Trans study reported high HIV prevalences among transgender women in South Africa, with 63.1% in Buffalo City, 49.4% in Cape Town, and 53.5% in Johannesburg, underscoring the urgent need for targeted, inclusive, and gender-affirming health interventions⁷.

The WHO, UNAIDS, and other global health bodies strongly support the use of the two-step gender identity question, which involves asking both current gender identity and sex assigned at birth, as a best practice for accurately capturing data on transgender and gender-diverse populations and advancing rights-based, inclusive public health responses^{8,12,14,31,32}. In South Africa, national policies recognize these needs, but challenges persist in implementation. The South African National LGBTI HIV Plan (2017–2022)³³ and the National Strategic Plan (NSP) 2023–2028⁵ acknowledge the challenges faced by key populations, emphasizing person-centered services; however, the current RHIMS cannot provide data to monitor the impact of those policies. South Africa’s progress toward inclusive, person-centered HIV care requires urgent RHIMS reform to close visibility gaps for TGs, through the incorporation of both key populations’ unique identifier code and gender identities. Incorporating SOGI data and key populations’ unique identifier code into RHIMS is both feasible and ethically necessary. Achieving this will require sustained political commitment and multisectoral collaboration, including the involvement of civil society, policymakers, donors, and technical partners to ensure that inclusive data systems are both effectively designed and sustainably implemented.

CONCLUSIONS

As countries work towards achieving the global HIV targets and ending the HIV/AIDS epidemic by 2030, strengthening routine health information systems remains essential for complete monitoring of progress across the HIV care cascade. This study provided insight into overall program performance, showing evidence of progression from diagnosis through treatment initiation, retention, and viral suppression based on routinely collected service data. However, despite the availability of these key program indicators, the current RHIMS framework does not allow their disaggregation by key population groups because SOGI variables and KP unique identifier codes are not captured. Given that key populations account for a disproportionate share of HIV infections, incorporating unique identifiers for key populations into the system would enable more targeted interventions, better resource allocation, and improved program management. The country is already one step ahead with the implementation of HPRS and the EMR. It is therefore crucial to consider the inclusion of those missing variables in the new system while it is at the piloting stage.