Effectiveness of BNT162b2 and Ad26.COV2.S vaccines against COVID-19-related hospitalisation amongst adult members of a private health insurance plan in South Africa during the Delta and Omicron periods: A test-negative case-control study.

BACKGROUND

COVID-19 vaccine effectiveness estimates from Africa are limited. These data can guide decisions on selecting priority groups in vaccine programs. This study estimated VE for BNT162b2 and Ad26.COV2.S against COVID-19-related hospitalisation, stratified by age group, time since vaccination, and HIV-infection status for three SARS-CoV-2 surges in South Africa.

METHODS

We applied a test-negative case-control design to hospitalisations for acute respiratory infections amongst members of a medical insurance plan during the delta (9 May 2021-18 September 2021), omicron BA.1 (28 November 2021-5 February 2022), and BA.4/5 (17 April 2022-28 May 2022) variant periods. All data, including vaccination history, were extracted from insurance plan claims. Logistic regression models adjusted for age, comorbidities, time since vaccination, income level and documentation of previous SARS-CoV-2 infection, were used to calculate VE.

RESULTS

BNT162b2 was protective against COVID-19-related hospitalisation for all variant periods (VE 89.3 % (95 % CI, 85.9-91.9) for delta, reduced to 31.4 % (95 % CI, 19.1-41.9) and 22.7 % (95 % CI, 2.2-38.9) for omicron BA.1, and BA.4/5 respectively). VE estimates for Ad26.COV2·S, although lower than BNT162b2, were protective for all periods (48.8 % (95 % CI, 39.6-56.5), 19.8 % (95 % CI, 5.8-31.6), and 45.0 % (95 % CI, 29.8-57.0) for delta, omicron BA.1, and BA.4/5 respectively). Protection against severe infection was shown in those ≥60 years and younger age groups, as well as in people living with HIV (PLWH) and HIV-uninfected individuals.

CONCLUSION

Vaccination offered significant protection against COVID-19-related hospitalisation in PLWH and the elderly, and is therefore an effective means of reducing severe outcomes in these high-risk populations in South Africa. VE against BA.4/5 waned with time since vaccination suggesting boosters may be necessary.