HIV self testing uptake and associated factors in Cape Town: a contextual framework

Nyembwe, Doris

HIV self testing uptake and associated factors in Cape Town: a contextual framework

Thesis / Dissertation

2025

Publisher

University of Cape Town

Department

Department of Public Health and Family Medicine

Faculty

Faculty of Health Sciences

Abstract

South Africa bears one of the highest HIV burdens globally, with nearly 8 million people living with the virus. Despite hosting the world's largest antiretroviral therapy (ART) program, HIV-related deaths remain significant, accounting for over 23% of all deaths in 2019. Early detection and timely initiation of ART are essential to prevent transmission, improve quality of life, and reduce HIV-related morbidity and mortality. However, insufficient testing coverage among males and younger individuals remains a concern. HIV self-testing (HIVST) has emerged as a promising strategy to bridge these gaps, offering a private and convenient option for individuals hesitant to access healthcare facilities. The World Health Organization (WHO) endorses HIVST as a complementary approach to enhance access, particularly for populations underserved by traditional testing methods. While research has examined HIVST uptake in various settings, little is known about the specific factors influencing its adoption in Cape Town. Given South Africa's unique socio-economic landscape and the disparities between urban and rural areas, understanding the factors shaping HIVST uptake is crucial for developing tailored interventions. This thesis seeks to address this gap by investigating and analyzing the demographic, socio-economic, and community-level factors associated with HIVST uptake in Cape Town. Methods: This study utilized a cross-sectional design to examine HIV testing uptake and associated factors in Cape Town, South Africa, between January and December 2022. The analysis leveraged routine HIV Testing Services (HTS) programmatic data collected by the Anova Health Institute. The dataset included a total of 266,284 observations: 30,785 for HIV self-testing (HIVST) and 235,499 for conventional HIV testing. Data were drawn from individuals aged 18 years and older across the eight subdistricts of the Cape Town metropolitan area: Eastern, Northern, Southern, Western, Khayelitsha, Klipfontein, Mitchells Plain, and Tygerberg. The data, comprising sociodemographic details and testing information, were deidentified with formal permission from Anova Health Institute and the Department of Health. Individual-level data was recorded through consent forms and HTS registers and subsequently transferred to Red Cap and Power BI for quality checks and analysis. Community-level data, including the number of healthcare facilities, new and registered ART patients, and child acute malnutrition rates, were sourced from City of Cape Town health profiles (2021). Predictors were selected based on a socio-ecological framework, capturing both individual- and community-level factors. Individual-level variables included age, gender, and HIV testing history. Community-level factors encompassed healthcare access (number of healthcare facilities), HIV burden (number of registered and new ART patients), and socioeconomic status (child acute malnutrition rate). Descriptive statistics summarize the frequencies of HIVST, and conventional testing variables stratified by subdistrict, alongside community-level factors. A bivariate logistic regression model was conducted to assess associations between individual predictors and HIV testing options. Subsequently, a multivariate logistic regression model was employed to evaluate the influence of both individual- and community-level predictors on HIV testing choices (conventional vs. HIVST). Odds ratios were calculated with 95% confidence intervals to quantify these associations. This methodology integrates diverse data sources and robust statistical approaches, enabling a comprehensive examination of the factors influencing the uptake of HIV self-testing in Cape Town. Results: The study had a sample size of 265,063 of which 234,853 (88.60%) had utilized conventional HIV testing method and 30,210 (11.40%) opting for self-testing. Majority of individuals undergoing conventional testing are adults aged 25- 49 (63.16%), followed by older adults aged 50+ (17.16%). Similarly, for self-testing, most users are also within the 25-49 age group (63.84%), but there is a higher proportion of young adults aged 20-24 choosing self-testing (23.72%) compared to conventional testing (15.59%). Additionally, adolescents aged 18-19 are more likely to opt for self-testing (7.75%) than conventional testing (4.08%). Regarding gender, females constitute a larger share of those undergoing conventional testing (65.62%) compared to males (34.38%). The trend is similar for self-testing, where females account for 65.32%, and males make up 34.68%. In terms of the last HIV test, self-testing is more prevalent among individuals who were tested within the past 12 months (63.51%), while conventional testing is more common among those whose last test was over a year ago. Subdistrict analysis shows that conventional testing is most frequent in Tygerberg (20.06%) and Khayelitsha (16.87%), followed by Western (13.24%) and Eastern (12.45%). In contrast, self-testing is more widely utilized in Western (19.38%), Southern (15.33%), and Mitchell's Plain (16.97%). The bivariate logistic regression indicated that age was a significant factor influencing self-testing preferences, with the likelihood of using HIVST decreasing with age. Individuals aged 20‐24 had 20% lower odds of using self‐testing compared to adolescents aged 18‐19 (OR = 0.80, 95% CI: 0.76–0.85, p < 0.005). Those aged 25‐49 had 47% lower odds compared to the adolescent group (OR = 0.53, 95% CI: 0.51–0.56, p < 0.005), and adults aged 50 and above had 86% lower odds (OR = 0.14, 95% CI: 0.13–0.15, p < 0.005). Additionally, for facility testing those who had tested for HIV within the last 12 months were more inclined towards self‐testing. In contrast, individuals who last tested more than 12 months ago were 77% less likely to choose self‐testing (OR = 0.23, 95% CI: 0.22–0.25, p < 0.005), and those who had never been tested were 40% less likely (OR = 0.60, 95% CI: 0.54–0.67, p < 0.005) to use self‐testing. HIVST was more popular among people living in areas with a high concentration of registered ART patients. Specifically, the odds of choosing self‐testing increased by 32% in high-density areas (≥30,001 registered ART patients) (OR = 1.32, 95% CI: 1.28–1.37, p < 0.005) and by 53% in medium-density areas (20,001–30,000 registered ART patients) (OR = 1.53, 95% CI: 1.49–1.58, p < 0.005), compared to areas with fewer than 20,000 registered ART patients. On the other hand, people living in areas with a higher number of healthcare facilities were more likely to choose conventional HIV testing. The odds of self‐testing decreased by 15% in subdistricts with a medium number of healthcare facilities (15–25 facilities) (OR = 0.85, 95% CI: 0.82–0.87, p < 0.005) and by 27% in areas with a high number of facilities (26 or more) (OR = 0.73, 95% CI: 0.71–0.76, p < 0.005), relative to areas with fewer than 15 facilities. Additionally, communities with a high number of newly enrolled ART patients (≥2,901) showed a 31% lower likelihood of opting for self-testing (OR = 0.69, 95% CI: 0.67–0.72, p < 0.005). Subdistrict variations were evident, with Southern (OR = 2.04, 95% CI: 1.94–2.13, p < 0.005) and Mitchell's Plain (OR = 2.12, 95% CI: 2.03–2.23, p < 0.005) showing more than twice the odds of self‐testing compared to the Eastern subdistrict. Other subdistricts with significantly higher odds of self-testing included Western (OR = 1.63, 95% CI: 1.56–1.71, p < 0.005) and Klipfontein (OR = 1.12, 95% CI: 1.06–1.18, p < 0.005). Conversely, Northern (OR = 0.52, 95% CI: 0.48–0.55, p < 0.005), Tygerberg (OR = 0.55, 95% CI: 0.52–0.57, p < 0.005), and Khayelitsha (OR = 0.97, 95% CI: 0.83–0.91, p < 0.005) had significantly lower odds. Finally, testing preferences assessed through the multivariate logistic regression model highlighted the influence of both individual- and community-level factors. Consistent with the bivariate analysis findings, age remained a strong predictor of HIVST. Individuals aged 20–24 had 23% lower odds of using self-testing compared to those aged 18–19 (OR = 0.77, 95% CI: 0.73–0.82, p < 0.005), while those aged 25–49 had 49% lower odds (OR = 0.51, 95% CI: 0.48–0.53, p < 0.005). The oldest age group (50 years and above) had 86% lower odds of choosing self-testing compared to the youngest group (OR = 0.14, 95% CI: 0.13–0.15, p < 0.005). 5 Individuals residing in communities with a medium (20,001–30,000) and high (≥30,001) number of registered ART patients had 240% (OR = 3.40, 95% CI: 3.18–3.65) and 182% (OR = 2.82, 95% CI: 2.70– 2.96, p < 0.005) higher odds of using self-testing, respectively, compared to those in areas with low ART caseloads (<20,000). Additionally, living in areas with more newly enrolled ART patients was negatively associated with self- testing. Residing in communities with a medium number of new ART patients (1,800–2,900) was associated with 68% lower odds of self-testing (OR = 0.32, 95% CI: 0.29–0.34, p < 0.001), while living in areas with a high number of new ART initiations (≥2,901) was associated with 81% lower odds (OR = 0.19, 95% CI: 0.18–0.21, p < 0.001), compared to areas with a low number of new ART patients (<1,800). Unlike in the bivariate analysis, gender also played a significant role in the multivariate model, with females having 8% lower odds of choosing self-testing compared to males (OR = 0.92, 95% CI: 0.90– 0.94, p < 0.005). Additionally, individuals who received a positive HIV test result had 9% lower odds of having used self-testing compared to those who tested negative (OR = 0.91, 95% CI: 0.84–0.98, p = 0.005). The number of healthcare facilities was also positively associated with self-testing uptake. Living in areas with a medium (15–25) or high (≥26) number of healthcare facilities increased the odds of self-testing by 13% and 34%, respectively (OR = 1.13, 95% CI: 1.09–1.17, p < 0.005; and OR = 1.34, 95% CI: 1.26–1.43, p < 0.005), compared to areas with a low number of facilities (0–14). Conclusion: In conclusion, the study underscores the complex interplay of individual and community-level factors influencing HIV testing preferences in Cape Town. Younger age, recent HIV testing history, male gender, and residence in areas with higher number of registered ART patient and more healthcare facilities were associated with increased likelihood of HIV self-testing (HIVST). Conversely, older age, female gender, living in communities with more newly initiated ART clients, and receiving a positive HIV diagnosis were linked to a lower likelihood of using HIVST. Geographic disparities across subdistricts further highlight the need for targeted, context-specific strategies to enhance HIV testing uptake, particularly among underrepresented groups, and to optimize the reach and impact of self-testing interventions.