Development of an HIV risk index for university students in KwaZulu-Natal, South Africa

Abstract

Background: Human Immunodeficiency Virus (HIV) infection is a global health challenge, with developing countries most affected by the HIV burden. In South Africa (SA), young people aged 15 to 24 years are exceptionally vulnerable to HIV infection. University students are a segment of this vulnerable age group in SA. Significant effort and resources have been invested in HIV prevention and treatment such as availability of free government issue condoms, a large-scale antiretroviral (ARV) programme and intensification of other public health interventions such as male medical circumcision (MMC) and the availability of Pre exposure prophylaxis (PrEP). Nonetheless, HIV still poses a critical public health challenge 40 years post discovery, particularly among young people. Factors that may increase susceptibility include substance usage, tendency to engage in risky sexual behaviours (RSBs) such as transactional sex, inconsistent condom usage and engaging in sex with multiple partners. Poverty may further perpetuate RSBs such as transactional sex for money or to acquire basic needs. University students form part of the country’s economic backbone and are the future leaders of the country, thus context driven HIV prevention interventions for this unique population are imperative. HIV risk is multidimensional and in order to develop effective prevention interventions, risk factors associated with RSBs and HIV seropositivity must be further explored. Aim: The study aimed to analyse the behavioural and socio-economic components of risk associated with HIV and health among student populations in an attempt to design an evidence based risk index. Methodology: This study used a case control study design. Study sites included the four major public sector higher educational institutions (HEIs) in KwaZulu-Natal, South Africa. A structured questionnaire was used to collect data using convenience sampling from the study population with consisted of 375 HIV uninfected students (controls) and 125 HIV infected students (cases). STATA-version 17 was used to conduct data analysis. Data collected from controls was initially used to conduct a preliminary analysis to explore the association between food insecurity which was a socio-economic risk factor and RSBs. Subsequently, the main analysis was undertaken to investigate associations of the following: socio-economic, behavioural and knowledge risk factors with RSBs and HIV seropositivity using bivariate analysis and logistic regression modelling. Results from the case control data analysed was used to develop an HIV risk index tool, for specific use among university students. Key findings: Preliminary data analysis of control data found significant associations between food insecurity and students engaging in transactional sex for money as well as to meet basic needs. Participants who were food insecure were twice as likely to have multiple sexual partners. The case control analysis with socio-economic risk factors revealed a significant association between socio-economic measures, RSBs and HIV seropositivity. Overall, food insecurity increased the likelihood of multiple partners, transactional sex for money, and transactional sex for basic needs. Students who accessed the government funding scheme and shared the bursary/ loan were more likely to be HIV seropositive. Results from the analysis on behavioural risk factors indicated that heavy episodic drinking [aOR: 2.73 (1.38; 5.44), 0.004], drugs before sex [aOR: 7.46 (2.11; 27.88), 0.003], and a higher number of lifetime sex partners (2 to 5 partners) [aOR: 4.22(1.69; 10.54), 0.002] and ≥ 6 partners [aOR: 16.36 (6.18; 43.28), 0.000] increased students’ likelihood of having multiple concurrent sex partners. High risk behaviours such as sex with multiple partners and inconsistent condom use were more prevalent among HIV uninfected students. Data from our study showed that both HIV infected, and uninfected students displayed a better understanding regarding modes of HIV transmission compared to prevention methods. Students with poor knowledge regarding HIV prevention were 2.34 times more likely to be HIV positive and those demonstrating poor knowledge about HIV transmission were 4.79 times more likely to be HIV positive. Several misconceptions regarding HIV transmission methods were prevalent among both HIV infected and uninfected students. The above findings regarding socio-economic, behavioural and knowledge risk factors were used to develop an HIV risk index tool based on logistic regression models. The risk index score comprises 17 factors and used an aggregated score to formulate three risk categories namely, low, medium, and high-risk categories. Questions were formulated for the risk index tool based on variables used in the data analysis. Conclusion: Our findings demonstrate that university students are at an increased risk of engagement in RSBs which may consequently lead to HIV infection. Therefore, it is of critical importance for public health officials to consider the multidimensional aspects of HIV risk in the university student population when designing interventions targeted at this vulnerable group. Risk measurement is important to guide HIV prevention methods such as Pre-exposure prophylaxis (PrEP). The developed risk index tool has the potential to serve as a powerful public health tool for measurement of HIV risk among university students in developing countries. However, further testing and evaluation is required before implementation.