Understanding the differential impacts of COVID-19 among hospitalised patients in South Africa for equitable response

Background. There are limited in-depth analyses of COVID-19 differential impacts, especially in resource-limited settings such as South Africa (SA).Objectives. To explore context-specific sociodemographic heterogeneities in order to understand the differential impacts of COVID-19. Methods. Descriptive epidemiological COVID-19 hospitalisation and mortality data were drawn from daily hospital surveillance data, National Institute for Communicable Diseases (NICD) update reports (6 March 2020 - 24 January 2021) and the Eastern Cape Daily Epidemiological Report (as of 24 March 2021). We examined hospitalisations and mortality by sociodemographics (age using 10-year age bands, sex and race) using absolute numbers, proportions and ratios. The data are presented using tables received from the NICD, and charts were created to show trends and patterns. Mortality rates (per 100 000 population) were calculated using population estimates as a denominator for standardisation. Associations were determined through relative risks (RRs), 95% confidence intervals (CIs) and p-values <0.001. Results. Black African females had a significantly higher rate of hospitalisation (8.7% (95% CI 8.5 - 8.9)) compared with coloureds, Indians and whites (6.7% (95% CI 6.0 - 7.4), 6.3% (95% CI 5.5 - 7.2) and 4% (95% CI 3.5 - 4.5), respectively). Similarly, black African females had the highest hospitalisation rates at a younger age category of 30 - 39 years (16.1%) compared with other race groups. Whites were hospitalised at older ages than other races, with a median age of 63 years. Black Africans were hospitalised at younger ages than other race groups, with a median age of 52 years. Whites were significantly more likely to die at older ages compared with black Africans (RR 1.07; 95% CI 1.06 - 1.08) or coloureds (RR 1.44; 95% CI 1.33 - 1.54); a similar pattern was found between Indians and whites (RR 1.59; 95% CI 1.47 - 1.73). Women died at older ages than men, although they were admitted to hospital at younger ages. Among black Africans and coloureds, females (50.9 deaths per 100 000 and 37 per 100 000, respectively) had a higher COVID-19 death rate than males (41.2 per 100 000 and 41.5 per 100 000, respectively). However, among Indians and whites, males had higher rates of deaths than females. The ratio of deaths to hospitalisations by race and gender increased with increasing age. In each age group, this ratio was highest among black Africans and lowest among whites.Conclusions. The study revealed the heterogeneous nature of COVID-19 impacts in SA. Existing socioeconomic inequalities appear to shape COVID-19 impacts, with a disproportionate effect on black Africans and marginalised and low socioeconomic groups. These differential impacts call for considered attention to mitigating the health disparities among black Africans.

New insights into HIV epidemic in South Africa: key findings from the national HIV prevalence: incidence and behaviour burvey; 2012

This article presents key findings from the 2012 HIV prevalence; incidence and behaviour survey conducted in South Africa and explores trends in the HIV epidemic. A representative household based survey collected behavioural and biomedical data among people of all ages. Chi-squared test for association and formal trend tests (2002; 2005; 2008 and 2012) were used to test for associations and trends in the HIV epidemic across the four surveys. In 2012 a total of 38 431 respondents were interviewed from 11 079 households; 28 997 (67.5%) of 42 950 eligible individuals provided blood specimens. HIV prevalence was 12.2% [95% CI: 11.4-13.1] in 2012 with prevalence higher among females 14.4% than males 9.9%. Adults aged 25-49 years were most affected; 25.2% [95% CI: 23.2-27.3]. HIV prevalence increased from 10.6% [95%CI: 9.8-11.6] in 2008 to 12.2% [95% CI: 11.4-13.1] in 2012 (p 0.001). Antiretroviral treatment (ART) exposure doubled from 16.6% in 2008 to 31.2% in 2012 (p 0.001). HIV incidence in 2012 among persons 2 years and older was 1.07% [95% CI: 0.87-1.27]; with the highest incidence among Black African females aged 20-34 years at 4.5%. Sexual debut before 15 years was reported by 10.7% of respondents aged 15-24 years; and was significantly higher among male youth than female (16.7% vs. 5.0% respectively; p 0.001). Reporting of multiple sexual partners in the previous 12 months increased from 11.5% in 2002 to 18.3% in 2012 (p 0.001). Condom use at last sex dropped from 45.1% in 2008 to 36.2% in 2012 (p 0.001). Levels of accurate HIV knowledge about transmission and prevention were low and had decreased between 2008 and 2012 from 31.5% to 26.8%. South Africa is on the right track with scaling up ART. However; there have been worrying increases in most HIV-related risk behaviours. These findings suggest that there is a need to scale up prevention methods that integrate biomedical; behavioural; social and structural prevention interventions to reverse the tide in the fight against HIV

Age at sexual debut: a determinant of multiple partnership among south african youth

Age at sexual debut is an important determinant of HIV infection. The paper investigates the effects of age at sexual debut on sexual behaviour among South African youth. Among 2 875 respondents who ever had sexual intercourse, 39% had early sexual debut (sexual debut at age 16 years and below). Males (44.6%) were significantly more likely than females (35.1%) to report early sexual debut (odds ratio (OR)=1.45, p-value <0.001). Multiple sexual partners are significantly more common among those that had early sexual debut (10.4% vs. 4.8%) than those who had late sexual debut, (OR=2.29, p-value<0.001). Those aged 15 to 19 years were 1.4 times more likely to report multiple partners compared to those aged 20 to 24 years. Delaying sexual debut is a strategy many national programmes are promoting. The results of this study provide additional arguments to support such initiatives and show the need to strengthen intervention targeting youth (Afr. J. Reprod. Health 2010; 14[2]:47-54)