Despite improvements in access to antiretroviral treatment over the past decade, sub-optimal HIV care outcomes persist among youth with HIV (YWH) in rural South Africa. Psychosocial stressors could impede improved HIV treatment outcomes within this population. We linked self-reported psychosocial health and demographic data from a cross-sectional survey conducted among YWH aged 12–24 in rural South Africa to individual medical record data, including facility visit history and viral load measurements. Poisson regression with robust standard errors was used to estimate the associations between five psychosocial stressors- heightened depressive symptoms (Center for Epidemiological Studies-Depression scale scores ≥ 16), lower social support (Medical Outcomes Social Support Scale scores ≤ 38), lower resilience (Conner-Davidson Resilience Scale scores ≤ 73), lower self-esteem (Rosenberg Self-Esteem Scale scores ≤ 21), and higher perceived stress (Sheldon Cohen Perceived Stress Scale scores ≥ 10)- and viral non-suppression (viral load ≥ 400 copies/mL) and loss to care (no documented clinic visits within the 90 days prior to survey), separately. A total of 359 YWH were included in this analysis. The median age of study participants was 21 (interquartile range: 16–23), and most were female (70.2%), single (82.4%), and attending school (54.7%). Over a quarter of participants (28.1%) had heightened depressive symptoms. Just 16.2% of all participants (n = 58) were lost to care at the time of survey, while 32.4% (n = 73) of the 225 participants with viral load data were non-suppressed. The prevalence of non-suppression in individuals with lower self-esteem was 1.71 (95% confidence interval: 1.12, 2.61) times the prevalence of non-suppression in those with higher self-esteem after adjustment. No meaningful association was observed between heightened depressive symptoms, lower social support, lower resilience, and higher perceived stress and viral non-suppression or loss to care in adjusted analyses. Retention in care and viral suppression among YWH in rural South Africa are below global targets. Interventions aimed at improving viral suppression among YWH should incorporate modules to improve participant’s self-esteem as low self-esteem is associated with viral non-suppression in this setting. Future studies should longitudinally explore the joint effects of co-occurring psychosocial stressors on HIV care outcomes in YWH and assess meaningful differences in these effects by age, gender, and route of transmission.

Linking a health and demographic surveillance system (HDSS) to data from a health facility that serves the HDSS population generates a research infrastructure for directly observed data on access to and utilization of health facility services. Many HDSS sites, however, are in areas that lack unique national identifiers or suffer from data quality issues, such as incomplete records, spelling errors, and name and residence changes, all of which complicate record linkage approaches when applied retrospectively. We developed Point-of-contact Interactive Record Linkage (PIRL) software that is used to prospectively link health records from a local health facility to an HDSS in rural Tanzania. This prospective approach to record linkage is carried out in the presence of the individual whose records are being linked, which has the advantage that any uncertainty surrounding their identity can be resolved during a brief interaction, whereby extraneous information (e.g., household membership) can be referred to as an additional criterion to adjudicate between multiple potential matches. Our software uses a probabilistic record linkage algorithm based on the Fellegi-Sunter model to search and rank potential matches in the HDSS data source. Key advantages of this software are its ability to perform multiple searches for the same individual and save patient-specific notes that are retrieved during subsequent clinic visits. A search on the HDSS database (n=110,000) takes less than 15 seconds to complete. Excluding time spent obtaining written consent, the median duration of time we spend with each patient is six minutes. In this setting, a purely automated retrospective approach to record linkage would have only correctly Gates Open Research identified about half of the true matches and resulted in high linkage errors; therefore highlighting immediate benefit of conducting interactive record linkage using the PIRL software.

Linking a health and demographic surveillance system (HDSS) to data from a health facility that serves the HDSS population generates a research infrastructure for directly observed data on access to and utilization of health facility services. Many HDSS sites, however, are in areas that lack unique national identifiers or suffer from data quality issues, such as incomplete records, spelling errors, and name and residence changes, all of which complicate record linkage approaches when applied retrospectively. We developed Point-of-contact Interactive Record Linkage (PIRL) software that is used to prospectively link health records from a local health facility to an HDSS in rural Tanzania. This prospective approach to record linkage is carried out in the presence of the individual whose records are being linked, which has the advantage that any uncertainty surrounding their identity can be resolved during a brief interaction, whereby extraneous information (e.g., household membership) can be referred to as an additional criterion to adjudicate between multiple potential matches. Our software uses a probabilistic record linkage algorithm based on the Fellegi-Sunter model to search and rank potential matches in the HDSS data source. Key advantages of this software are its ability to perform multiple searches for the same individual and save patient-specific notes that are retrieved during subsequent clinic visits. A search on the HDSS database (n=110,000) takes less than 15 seconds to complete. Excluding time spent obtaining written consent, the median duration of time we spend with each patient is six minutes. In this setting, a purely automated retrospective approach to record linkage would have only correctly identified about half of the true matches and resulted in high linkage errors; therefore highlighting immediate benefit of conducting interactive record linkage using the PIRL software.