The socioeconomic gradient of alcohol use: an analysis of nationally representative survey data from 55 low-income and middle-income countries.

BACKGROUND: Alcohol is a leading risk factor for over 200 conditions and an important contributor to socioeconomic health inequalities. However, little is known about the associations between individuals' socioeconomic circumstances and alcohol consumption, especially heavy episodic drinking (HED; ≥5 drinks on one occasion) in low-income or middle-income countries. We investigated the association between individual and household level socioeconomic status, and alcohol drinking habits in these settings. METHODS: In this pooled analysis of individual-level data, we used available nationally representative surveys-mainly WHO Stepwise Approach to Surveillance surveys-conducted in 55 low-income and middle-income countries between 2005 and 2017 reporting on alcohol use. Surveys from participants aged 15 years or older were included. Logistic regression models controlling for age, country, and survey year stratified by sex and country income groups were used to investigate associations between two indicators of socioeconomic status (individual educational attainment and household wealth) and alcohol use (current drinking and HED amongst current drinkers). FINDINGS: Surveys from 336 287 participants were included in the analysis. Among males, the highest prevalence of both current drinking and HED was found in lower-middle-income countries (L-MICs; current drinking 49·9% [95% CI 48·7-51·2] and HED 63·3% [61·0-65·7]). Among females, the prevalence of current drinking was highest in upper-middle-income countries (U-MIC; 29·5% [26·1-33·2]), and the prevalence of HED was highest in low-income countries (LICs; 36·8% [33·6-40·2]). Clear gradients in the prevalence of current drinking were observed across all country income groups, with a higher prevalence among participants with high socioeconomic status. However, in U-MICs, current drinkers with low socioeconomic status were more likely to engage in HED than participants with high socioeconomic status; the opposite was observed in LICs, and no association between socioeconomic status and HED was found in L-MICs. INTERPRETATION: The findings call for urgent alcohol control policies and interventions in LICs and L-MICs to reduce harmful HED. Moreover, alcohol control policies need to be targeted at socially disadvantaged groups in U-MICs. FUNDING: Deutsche Forschungsgemeinschaft and the National Center for Advancing Translational Sciences of the US National Institutes of Health.

Estimated effect of increased diagnosis, treatment, and control of diabetes and its associated cardiovascular risk factors among low-income and middle-income countries: a microsimulation model.

BACKGROUND: Given the increasing prevalence of diabetes in low-income and middle-income countries (LMICs), we aimed to estimate the health and cost implications of achieving different targets for diagnosis, treatment, and control of diabetes and its associated cardiovascular risk factors among LMICs. METHODS: We constructed a microsimulation model to estimate disability-adjusted life-years (DALYs) lost and health-care costs of diagnosis, treatment, and control of blood pressure, dyslipidaemia, and glycaemia among people with diabetes in LMICs. We used individual participant data-specifically from the subset of people who were defined as having any type of diabetes by WHO standards-from nationally representative, cross-sectional surveys (2006-18) spanning 15 world regions to estimate the baseline 10-year risk of atherosclerotic cardiovascular disease (defined as fatal and non-fatal myocardial infarction and stroke), heart failure (ejection fraction of <40%, with New York Heart Association class III or IV functional limitations), end-stage renal disease (defined as an estimated glomerular filtration rate <15 mL/min per 1·73 m2 or needing dialysis or transplant), retinopathy with severe vision loss (<20/200 visual acuity as measured by the Snellen chart), and neuropathy with pressure sensation loss (assessed by the Semmes-Weinstein 5·07/10 g monofilament exam). We then used data from meta-analyses of randomised controlled trials to estimate the reduction in risk and the WHO OneHealth tool to estimate costs in reaching either 60% or 80% of diagnosis, treatment initiation, and control targets for blood pressure, dyslipidaemia, and glycaemia recommended by WHO guidelines. Costs were updated to 2020 International Dollars, and both costs and DALYs were computed over a 10-year policy planning time horizon at a 3% annual discount rate. FINDINGS: We obtained data from 23 678 people with diabetes from 67 countries. The median estimated 10-year risk was 10·0% (IQR 4·0-18·0) for cardiovascular events, 7·8% (5·1-11·8) for neuropathy with pressure sensation loss, 7·2% (5·6-9·4) for end-stage renal disease, 6·0% (4·2-8·6) for retinopathy with severe vision loss, and 2·6% (1·2-5·3) for congestive heart failure. A target of 80% diagnosis, 80% treatment, and 80% control would be expected to reduce DALYs lost from diabetes complications from a median population-weighted loss to 1097 DALYs per 1000 population over 10 years (IQR 1051-1155), relative to a baseline of 1161 DALYs, primarily from reduced cardiovascular events (down from a median of 143 to 117 DALYs per 1000 population) due to blood pressure and statin treatment, with comparatively little effect from glycaemic control. The target of 80% diagnosis, 80% treatment, and 80% control would be expected to produce an overall incremental cost-effectiveness ratio of US$1362 per DALY averted (IQR 1304-1409), with the majority of decreased costs from reduced cardiovascular event management, counterbalanced by increased costs for blood pressure and statin treatment, producing an overall incremental cost-effectiveness ratio of $1362 per DALY averted (IQR 1304-1409). INTERPRETATION: Reducing complications from diabetes in LMICs is likely to require a focus on scaling up blood pressure and statin medication treatment initiation and blood pressure medication titration rather than focusing on increasing screening to increase diabetes diagnosis, or a glycaemic treatment and control among people with diabetes. FUNDING: None.

Cognitive differences associated with HIV serostatus and antiretroviral therapy use in a population-based sample of older adults in South Africa.

Previous clinical studies have reported adverse cognitive outcomes for people living with HIV (PLWH), but there are no population-based studies comparing cognitive function between older PLWH and comparators without HIV in sub-Saharan Africa. We analyzed baseline data of 40 + years-old participants in "Health and Aging in Africa: A Longitudinal Study of an INDEPTH Community in South Africa" (HAALSI) cohort. We measured cognition using a battery of conventional instruments assessing orientation, immediate- and delayed-recall, and numeracy (N = 4560), and the Oxford Cognitive Screen [OCS]-Plus, a novel instrument for low-literacy populations, assessing memory, language, visual-spatial ability, and executive functioning (N = 1997). Linear regression models comparing cognitive scores between participants with and without HIV were adjusted for sex, education, age, country of birth, father's occupation, ever-consumed alcohol, and asset index. PLWH scored on average 0.06 (95% CI 0.01-0.12) standard deviation (SD) units higher on the conventional cognitive function measure and 0.02 (95% CI - 0.07 to 0.04) SD units lower on the OCS-Plus measure than HIV-negative participants. We found higher cognitive function scores for PLWH compared to people without HIV when using a conventional measure of cognitive function but not when using a novel instrument for low-literacy settings.

Validity of hemoglobin A1c for diagnosing diabetes among people with and without HIV in Uganda.

Sub-Saharan Africa (SSA) is facing a growing co-epidemic of chronic HIV infection and diabetes. Hemoglobin A1c (A1c) may underestimate glycemia among people living with HIV (PLWH). We estimated the validity of A1c to diagnose diabetes among PLWH and HIV-uninfected persons in rural Uganda. Data were derived from a cohort of PLWH and age- and gender-matched HIV-uninfected comparators. We compared A1c to fasting blood glucose (FBG) using Pearson correlations, regression models, and estimated the sensitivity and specificity of A1c for detecting diabetes with FBG ≥126 mg/dL as reference standard. Approximately half (48%) of the 212 participants were female, mean age of 51.7 years (SD = 7.0) at enrollment. All PLWH (n = 118) were on antiretroviral therapy for a median of 7.5 years with mean CD4 cell count of 442 cells/µL. Mean FBG (89.7 mg/dL) and A1c (5.6%) were not different between PLWH and HIV-uninfected ( P > 0.50) groups, but the HIV-uninfected group had a higher prevalence of A1c >5.7% (33% vs. 20%, P = 0.024). We found a relatively strong correlation between A1c and FBG (r = 0.67). An A1c ≥6.5% had a poor sensitivity (46%, 95% CI 26-67%) but high specificity (98%, 95% CI 96-99%) for detecting diabetes. More work is needed to define an optimal A1c for screening diabetes in SSA.