Access to Tuberculosis testing among adolescents living with Human Immunodeficiency Virus in the Eastern Cape, South Africa: social factors and theoretical considerations

Abstract

Background: Addressing adolescent tuberculosis (TB) is a critical step towards eliminating TB in high burden countries, especially in HIV endemic communities. South Africa has the highest rates of TB/HIV co-infection and the largest population of adolescents living with HIV (ALHIV) in the world – contributing substantial risk to TB-related morbidity and mortality in this already vulnerable cohort. Previous research on TB has largely overlooked adolescents and ALHIV which has left knowledge – and potential service provision – gaps, but also opportunities for important research. TB among ALHIV is a complex public health challenge, needing to be understood in the context of the unique socio-emotional life stage of adolescence. This dissertation aims to provide insights into the critical first step in the ALHIV TB care cascade: access to TB testing. Through quantitative analysis, I explore the social factors that promote or prevent ALHIV from accessing TB testing in South Africa. Methods: In this longitudinal study, I analysed the Mzantsi Wakho cohort data from n=1046 ALHIV (10-19 years old) from 53 health facilities across the Amathole district of the Eastern Cape. N= 933 (89%) ALHIV – those who participated in the second and third cohort waves – were included in this analysis. Data were collected through self-reported questionnaires, assisted by trained and experienced researchers three times between 2014/2015 to 2017/2018. The selection of social factors that influence access to the outcome of TB testing was informed by an extensive scoping literature review. These factors were initially categorised using WHO's social determinants of health framework, which applies the Ecological Model. Thereafter, factors were filtered through the People Centred Model of TB care – to draw focus to the factors pertaining to the individual (both inter- and intrapersonal) rather than factors imbedded in health systems and services. Analysis was conducted in four steps: First descriptive analyses was used to summarise sociodemographic characteristics, relevant TB clinical data and HIV related factors at each interview (T2 and T3). Secondly, cross tabulation and frequencies of factors were done, comparing ALHIV who tested for TB to those that did not. Thirdly, univariate analysis was performed to identify factors with statistically significant associations with having a TB test or not. Lastly, multivariate regression models of these significant factors were run, both for each time point and over time (across both time points) using a stepwise approach by Hosmer-Lemeshow. The “why” or “how” these specific factors affected the probability of TB testing were then explored through the application of sociological theories and concepts, including the life course approach, social action theory and habitus. Findings: Consistently experiencing the following factors over time were linked to greater odds of TB testing: being 15 years and older (OR 1.43, CI 1.06-1.92, p 0.019), female ALHIV (OR 1.34, CI 1.02-1.75, p 0.033), in a relationship at both time points (OR 1.79, CI 1.23-2.62, p 0.002) and having had a viral load test each year (OR 1.50, CI 1.11-2.02, p 0.008). Having TB symptoms at either wave 2 or 3 was associated with TB testing (OR 1.46, CI 1.08-1.96, p 0.013). At Wave 2, no sim card phone (OR 0.64, CI 0.47-0.85, p 0.002) and having to pay R10 or more to get to the clinic (OR 0.68, CI 0.51-0.92, p 0.011) were associated with lower odds of TB testing, while viral load testing in the past year (OR 1.74, CI 1.26-2.40, p 0.001), living in a rural setting (0R 1.54, CI 1.10-2.16, p 0.012), being 15 years and older (OR 1.60, CI 1.19- 2.15, p 0.002) and reporting any TB symptoms (OR 1.72, CI 1.29-2.30, p< 0.001) were associated with higher odds of TB testing. At Wave 3, when most of the participants were in late adolescence being 15 years and older (OR 1.61, CI 1.19-2.19, p 0.002), living in informal housing (OR 1.58, CI 1.07-2.37, p 0.023), being in a relationship (OR 1.58, CI 1.15-2.18, p.005), experienced community violence (OR 1.43, CI 1.05-1.96, p 0.023), food security (OR 1.53, CI 1.11-2.11, p 0.010) and experienced any TB symptoms (OR 1.65, CI 1.25-2.20, p 0.001) had higher odds of reporting TB testing. Discussion and Conclusion: In this Eastern Cape cohort of ALHIV, factors linked to where ALHIV live (living rurally, cost to get the clinic more than R10, living in informal housing and having experienced community violence) as a reflection of the deep structural issues that shape health symptoms and healthcare access, who they are (age, sex) and their close emotional and nutritional support (being in a relationship, food security) have shown to strongly influence TB testing. Some of these factors are directly linked to increasing risk of TB exposure or vulnerability to TB: rural residence, informal housing and unsafe communities. To delve into why these factors shaped TB testing in ALHIV, sociological theories and concepts were applied to these findings. This dissertation took a holistic approach to bridge a critical knowledge gap in ALHIV's entry into TB care, extending our biomedical understanding with applied sociological frameworks. The work of this dissertation could enhance the current HIV services package offered to ALHIV by creating an awareness and identifying adolescents that may not be reached by current TB testing services. With this insight, TB services in South Africa, and perhaps broader afield, can introduce targeted interventions and social protection measures tailored to address adolescent TB testing, particularly in terms of integrating TB testing into HIV services.