Search

BackgroundHuman immunodeficiency virus (HIV) and hypertension are major contributors to morbidity and mortality in South Africa. Effective management of these conditions is critical to population health, yet patient management and retention varies by facility for reasons that are not fully understood. We assessed whether measures of clinic leadership, cohesion, and stress were associated with retention for HIV and hypertension in a cohort of patients in northeast South Africa before and during the Coronavirus disease 2019 pandemic.MethodsWe quantified nursing capacity and service readiness within primary health care facilities in the Bushbuckridge sub-district in Mpumalanga province South Africa. We administered brief scales on facility leadership, cohesion, and stress from January to March 2019, and tested scales for individual and facility-level agreement. We extracted clinical records for patients with HIV and/or hypertension from 2019 to 2021 and quantified treatment retention by quarter. We used generalised estimating equations to assess individual and clinic factors associated with retention in each treatment programme prior to (2019-first quarter 2020) and during (second quarter 2020-2021) the pandemic.ResultsThe nine facilities had a median of 12 nurses on staff and scored 0.83 out of 1.0 on basic service readiness. We collected responses to leadership, cohesion, and stress scales from 54 nurses and counsellors. Scales showed high inter-item agreement and moderate within-facility agreement. From 2019 to 2021, 19 445 individuals were treated for HIV and/or hypertension across seven participating facilities. Two-year retention was 91% for those with both conditions, 82% for those in treatment for HIV alone and 77% for those in treatment for hypertension alone, with 10-15% differences between facilities and high retention during the pandemic period. In addition to those with both conditions, women and adults aged 60-69 were more likely to be retained. Clinic factors were inconsistently associated with patient retention.ConclusionsWhile measures of clinic leadership, cohesion, and stress were generally reliable at individual and facility levels, we found limited evidence supporting a link between these factors and better retention in care. Retention was stable during the Coronavirus disease 2019 pandemic. Men, the youngest and oldest adults, and those without known multimorbidity should be prioritised for retention interventions.

We assess the accuracy of self-reported testing, HIV status, and treatment responses compared to clinical records in Ehlanzeni District, South Africa. We linked a 2018 population-based survey of adults 18-49 years old with clinical data at local primary healthcare facilities from 2014 to 2018. We calculated self-reported testing, HIV status, and treatment, and triangulated findings with clinic record data. We adjusted testing estimates for known gaps in HIV test documentation. Of 2089 survey participants, 1657 used a study facility and were eligible for analysis. Half of men and 84% of women reported an HIV test in the past year. One third of reported tests could be confirmed in clinic data within 1 year and an additional 13% within 2 years; these fractions increased to 57% and 22% respectively limiting to participants with a verified clinic file. After accounting for gaps in clinic documentation, we found that prevalence of recent HIV testing was closer to 15% among men and 51% in women. Estimated prevalence of known HIV was 16.2% based on self-report vs. 27.6% with clinic documentation. Relative to clinical records among confirmed clinic users, self report of HIV testing and of current treatment were highly sensitive but non-specific (sensitivity 95.5% and 98.8%, specificity 24.2% and 16.1% respectively), while self report of HIV status was highly specific but not sensitive (sensitivity 53.0%, specificity 99.3%). While clinical records are imperfect, survey-based measures should be interpreted with caution in this rural South African setting.

Social network analysis depends on how social ties to others are elicited during interviews, a process easily affected by respondent and interviewer behaviors. We investigate how the number of self-reported important social contacts varied within a single data collection round. Our data come from Health and Aging in Africa: a Longitudinal Study of an INDEPTH community (HAALSI), a comprehensive population-based survey of individuals aged 40 years and older conducted over 13 months at the Agincourt health and demographic surveillance site in rural South Africa. As part of HAALSI, interviewers elicited detailed egocentric network data. The average number of contacts reported by the 5,059 respondents both varied significantly across interviewers and fell over time as the data collection progressed, even after adjusting for respondent, interviewer, and respondent-interviewer dyad characteristics. Contact numbers rose substantially after a targeted interviewer intervention. We conclude that checking (and adjusting) for interviewer effects, even within one data collection round, is critical to valid and reliable social network analysis.

Aim To investigate mid-life employment trajectories in relation to later-life memory function and rate of decline in rural South Africa. Methods Data from the Agincourt Health and Socio-Demographic Surveillance System were linked to the 'Health and Ageing in Africa: A Longitudinal Study of an INDEPTH Community in South Africa' (HAALSI) in rural Agincourt, South Africa (N  =   3133). Employment was assessed every 4 years over 2000–12 as being employed (0, 1, 2 and ≥3 time points), being employed in a higher-skill occupation (0, 1, 2 and ≥3 time points) and dynamic employment trajectories identified using sequence analysis. Latent memory z-scores were assessed over 2014–22. Mixed-effects linear regression models were fitted to examine the associations of interest. Results Sustained mid-life employment from 2000–12 (β = 0.052, 95% CI: -0.028 to 0.132, 1 vs 0 time points; β = 0.163, 95% CI: 0.077 to 0.250, 2 vs 0 time points; β = 0.212, 95% CI: 0.128 to 0.296, ≥3 vs 0 time points) and greater time spent in a higher-skill occupation (β = 0.077, 95% CI: -0.020 to 0.175, 1 vs 0 time points; β = 0.241, 95% CI: 0.070 to 0.412, 2 vs 0 time points; β = 0.361, 95% CI: 0.201 to 0.520, ≥3 vs 0 time points) were associated with higher memory scores in 2014/15, but not subsequent rate of memory decline. Moving from a lower-skill to higher-skill occupation was associated with higher memory function, but a faster rate of decline over 2014–22. Conclusions Sustained mid-life employment, particularly in higher-skill occupations, may contribute to later-life memory function in this post-Apartheid South African setting. [ABSTRACT FROM AUTHOR]

Background Human immunodeficiency virus (HIV) and hypertension are major contributors to morbidity and mortality in South Africa. Effective management of these conditions is critical to population health, yet patient management and retention varies by facility for reasons that are not fully understood. We assessed whether measures of clinic leadership, cohesion, and stress were associated with retention for HIV and hypertension in a cohort of patients in northeast South Africa before and during the Coronavirus disease 2019 pandemic. Methods We quantified nursing capacity and service readiness within primary health care facilities in the Bushbuckridge sub-district in Mpumalanga province South Africa. We administered brief scales on facility leadership, cohesion, and stress from January to March 2019, and tested scales for individual and facility-level agreement. We extracted clinical records for patients with HIV and/or hypertension from 2019 to 2021 and quantified treatment retention by quarter. We used generalised estimating equations to assess individual and clinic factors associated with retention in each treatment programme prior to (2019–first quarter 2020) and during (second quarter 2020–2021) the pandemic. Results The nine facilities had a median of 12 nurses on staff and scored 0.83 out of 1.0 on basic service readiness. We collected responses to leadership, cohesion, and stress scales from 54 nurses and counsellors. Scales showed high inter-item agreement and moderate within-facility agreement. From 2019 to 2021, 19 445 individuals were treated for HIV and/or hypertension across seven participating facilities. Two-year retention was 91% for those with both conditions, 82% for those in treatment for HIV alone and 77% for those in treatment for hypertension alone, with 10–15% differences between facilities and high retention during the pandemic period. In addition to those with both conditions, women and adults aged 60–69 were more likely to be retained. Clinic factors were inconsistently associated with patient retention. Conclusions While measures of clinic leadership, cohesion, and stress were generally reliable at individual and facility levels, we found limited evidence supporting a link between these factors and better retention in care. Retention was stable during the Coronavirus disease 2019 pandemic. Men, the youngest and oldest adults, and those without known multimorbidity should be prioritised for retention interventions. [ABSTRACT FROM AUTHOR]

Verbal autopsies (VA) are commonly used in Low- and Medium-Income Countries (LMIC) to determine cause of death (CoD) where death occurs outside clinical settings, with the most commonly used international gold standard being physician medical certification. Interviewers elicit information from relatives of the deceased, regarding circumstances and events that might have led to death. This information is stored in textual format as VA narratives. The narratives entail detailed information that can be used to determine CoD. However, this approach still remains a manual task that is costly, inconsistent, time-consuming and subjective (prone to errors), amongst many drawbacks. As such, this negatively affects the VA reporting process, despite it being vital for strengthening health priorities and informing civil registration systems. Therefore, this study seeks to close this gap by applying novel deep learning (DL) interpretable approaches for reviewing VA narratives and generate CoD prediction in a timely, easily interpretable, cost-effective and error-free way. We validate our DL models using optimisation and performance accuracy machine learning (ML) curves as a function of training samples. We report on validation with training set accuracy (LSTM = 76.11%, CNN = 76.35%, and SEDL = 82.1%), validation accuracy (LSTM = 67.05%, CNN = 66.16%, and SEDL = 82%) and test set accuracy (LSTM = 67%, CNN = 66.2%, and SEDL = 82%) for our models. Furthermore, we also present Local Interpretable Model-agnostic Explanations (LIME) for ease of interpretability of the results, thereby building trust in the use of machines in healthcare. We presented robust deep learning methods to determine CoD from VAs, with the stacked ensemble deep learning (SEDL) approaches performing optimally and better than Long Short-Term Memory (LSTM) and Convolutional Neural Network (CNN). Our empirical results suggest that ensemble DL methods may be integrated in the CoD process to help experts get to a diagnosis. Ultimately, this will reduce the turnaround time needed by physicians to go through the narratives in order to be able to give an appropriate diagnosis, cut costs and minimise errors. This study was limited by the number of samples needed for training our models and the high levels of lexical variability in the words used in our textual information. [ABSTRACT FROM AUTHOR]

Telomere length (TL) may be a biomarker of aging processes as well as age-related diseases. However, most studies of TL and aging are conducted in high-income countries. Less is known in low- and middle-income countries (LMICs) such as South Africa, where life expectancy remains lower despite population aging. We conducted a descriptive analysis of TL in a cohort of older adults in rural South Africa. TL was assayed from venous blood draws using quantitative polymerase chain reaction (T/S ratio). We examined the correlation between TL and biomarkers, demographic characteristics, mental/cognitive health measures, and physical performance measures in a subsample of the Wave 1 2014–2015 "Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa" (HAALSI) cohort (n = 510). We used logistic regression to measure the association between TL and mortality through Wave 3 (2021–2022). In bivariate analyses, TL was significantly correlated with age (r = −0.29, p <.0001), self-reported female sex (r = 0.13, p =.002), mortality (r = −0.1297, p =.003), diastolic blood pressure (r = 0.09, p =.037), pulse pressure (r = −0.09, p =.045), and being a grandparent (r = −0.17, p =.0001). TL was significantly associated with age (β = −0.003; 95% confidence interval [CI] = −0.005, −0.003). TL was significantly associated in unadjusted multivariate analyses with mortality, but the relationship between TL and mortality was attenuated after adjusting for age (odds ratio [OR] = 0.19; 95% CI = 0.03, 1.27) and other covariates (OR = 0.17; 95% CI = 0.02, 1.19). Our study is the first analysis of TL in an older adult South African population. Our results corroborate existing relationships between TL and age, sex, cardiometabolic disease, and mortality found in higher-income countries. [ABSTRACT FROM AUTHOR]

What Is 'Success' in Terms of Reducing Under-Five Mortality? When generic criteria to assess achievements are formulated long in advance of endpoints, unforeseen consequences may follow in particular instances. The [...]