The relationship between the combined effects of life-course trauma and HIV on cognition in rural South African adults: A secondary data analysis
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of the Witwatersrand, Johannesburg
Abstract
Background:
Lifecourse traumatic events (TE) refers to both childhood and adult trauma. Childhood TE refers to a spectrum and domains of adverse experiences occurring before the age of 18. Global and South African-based research suggests that life-course TE are associated with both physical and mental disorders including HIV and poor cognition (also referred to as neurocognitive impairment (NCI)). While data suggests that life-course TE, HIV and NCI are highly prevalent in South Africa and risk of NCI has been researched, little is known of the combined effect of life-course TE and HIV on NCI in adulthood. To fill the gap in the literature, I analysed data from the Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa (HAALSI), from rural Bushbuckridge (Mpumalanga province).
Study Aim
To examine the combined (interaction) effect of TE and HIV infection on the cognition of older adults in South Africa.
Methods:
I analysed (secondary) data from the HAALSI study. The main exposures analysed were: life-course TE and HIV-positive status. A pre-analysis phase consisted of several exploratory steps to define the exposure and outcome: First TE data which had been measured by the English Longitudinal Study of Aging (ELSA) life history questionnaire were examined. The main exposure variable was coded 1 if one experiences a TE and 0 otherwise from seven potential TE. NCI, the main outcome variable was defined based on measuring cognitive domains: orientation and memory. A score≥1.5 standard deviations (SD) below the mean of the baseline
cognitive function distribution on the cognitive assessment (managing to know the date, day, month and president) will mean no NCI while a score below the means some degree of NCI). The main outcome variable was cognition measured by the Oxford Cognitive Screener (OCS-plus). Based on the validation of Tablet-Based OCS-plus in HAALSI, OCS-Plus included nine brief tests measuring cognitive ability in nine different domains, including language, memory (intentional and incidental), and executive functioning (task switching-alternating jobs between tasks), attention (auditory), and praxis (doing things). A stand-alone application called OCS- Plus was developed using Matlab and Psychophysics Toolbox for Windows Surface Pro tablets. Further preparatory steps for the analysis included: the exposure variables (HIV status and composite trauma-defined as the presence of at least one trauma item-) were created as follows: 1. HIV status was categorized into HIV positive and negative and coded 1 and 0 respectively. 2. Composite trauma was coded 1 if ever experienced any of the seven trauma items and 0 if none were experienced. Four groups of participants were created by exposure status. These were: 1. HIV positive and also experienced TE. 2. HIV negative but experienced TE. 3. HIV positive but TE negative. 4. HIV negative and trauma negative. Descriptive statistics were calculated for both exposures and outcomes as well as relevant sociodemographic variables. Both unadjusted and adjusted logistic regression techniques were employed to examine the combined effect of life course TE and HIV on NCI in older adults. The adjusted logistic regression models were done: 1) handling HIV and composite trauma separately and 2) considering the interaction term of HIV and composite trauma. Variables such as education, employment status, age, nationality, gender, hypertension, stroke, HIV status, marital status and composite trauma were considered confounders and adjusted accordingly.
Results:
Of the 5,059 study participants recruited and residing in the Agincourt study area in Bushbuckridge, 65% of the study participants had experienced at least 1 TE, and the most common trauma experienced were “ever experiencing severe financial hardship which was experienced by 58.74% of the participants, 39.73% whom “ever experienced a natural disaster” and 23.30% “ever experienced a death of a relative or friend” and the prevalence of NCI was 7% ( n=352).. The median age of participants was 64 (IQR: 55-74) years; 53.07% of the study participants were females; 44.42% had no formal education, and 72.63% of the study participants were not working. In the multiple logistic regression model with the interaction term, the odds of having NCI decreased by 64% (AOR=0.36; 95%CI: 0.25-0.52) and 59%
(AOR=0.41; 95%CI: 0.24-0.75) among those who had some primary school (1-7 years) and some secondary school (8+ years), respectively compared to those with no education. The odds of having NCI decreased by 98% (AOR=1.98; 95%CI: 1.05-3.72) among those who were not working compared to those employed. A one-year increase in age was associated with a 5% increase in the odds of having a cognitive impairment (OR=1.05; 95%CI: 1.04-1.07). Those who were married had 37% (OR=0.63, 95%CI: 0.47-0.84) reduced odds of having NCI compared to those who were married. Those without composite trauma had 90% (AOR=0.10;
95%CI: 0.07-0.15) reduced odds of having NCI compared to those who have composite trauma. Those who were HIV positive and had experienced composite trauma had an increased odds of 1.78 (95% CI: 1.04-3.04) of having NCI compared to those who are HIV negative and had not experienced composite trauma. In the interaction model, we found no association between HIV status, stroke, or hypertension and NCI. However, the results of the interactionmodel suggested a significant association between HIV and the composite trauma score on NCI.
Conclusions:
The results suggest that lifecourse TE and HIV infection influence NCI. The full HAALSI cohort could be employed to examine the effect of TE data collected at baseline and incident NCI at later waves
Description
A Dissertation submitted in fulfillment of the requirements for the degree of Master of Science: Infectious Disease Epidemiology The faculty of Health sciences School of Public Health: Division of Epidemiology and Biostatistics to the Faculty of Science, University of the Witwatersrand, Johannesburg 2024
Keywords
Lifecourse traumatic events (TE), HIV, Rural South African adults:, UCTD
Citation
Kupa, Nkgodi Obed. (2024). The relationship between the combined effects of life-course trauma and HIV on cognition in rural South African adults: A secondary data analysis [Master’s dissertation, University of the Witwatersrand, Johannesburg]. WireDSpace.