ABSTRACT
ABSTRACT Background The SARS-CoV-2 Beta variant, associated with immune escape and higher transmissibility, drove a more severe second COVID-19 wave in South Africa. Individual patient level characteristics and outcomes with the Beta variant are not well characterized. Methods We performed a retrospective cohort study comparing disease severity and inpatient mortality of COVID-19 pneumonia between the first and second wave periods at a referral hospital in Cape Town, South Africa. Beta variant infection was confirmed by genomic sequencing. Outcomes were analyzed with logistic regression and accelerated failure time models. Results 1,182 patients were included: 571 during the first wave period and 611 from the second wave. Beta variant accounted for 97% of infections in the second wave. There was no difference in crude in-hospital mortality between wave periods (first wave 22.2%, second wave 22.1%; p = 0.9). Time to death was decreased with higher weekly hospital admissions (16%; 95% CI, 8 to 24 for every 50-patient increase), age (18%; 95% CI, 12 to 24 for every 10-year increase) and hypertension (31%; 95% CI, 12 to 46). Corticosteroid use delayed time to death by 2-fold (95% CI, 1.5 to 3.0). Admission during the second wave decreased time to death after adjustment for other predictors, but this did not reach statistical significance (24%; 95% CI, 47 to -2). There was no effect of HIV on survival. Conclusions There was a trend towards earlier mortality during the second COVID-19 wave driven by the Beta variant, suggesting a possible biological basis. Use of oral prednisone was strongly protective. Key points In Cape Town, South Africa, the second wave of COVID-19, dominated by the Beta variant, was associated with decreased time to inpatient death after adjustment for age, comorbidities, steroid use, and admission numbers. Use of oral prednisone was strongly protective.