Population-based cross-sectional seroprevalence study of SARS-CoV-2 infection in urban residential areas in Ghana.

Introduction: Estimates of SARS-CoV-2-specific IgG antibodies using robust designs and serological assays have been a key component in assessing and reporting the overall occurrence of COVID-19 infection burden in the general population. The objective of this study was to estimate the seroprevalence and characteristics of COVID-19 infection transmission in Ghana. Methods: We conducted a cross-sectional survey using a two-stage stratified random sampling of community household members from February 2021 to February 2022 in Ghana, to estimate SARS-CoV-2 seroprevalence. Bayesian models with poststratification weighting of the demographic structure of the sample population were used for seroprevalence estimates, accounting for the uncertainty of diagnostic assay in a hierarchical model. Exposure-wide analysis was performed to evaluate the characteristics of the widespread SARS-CoV-2 infection. Results: On average, two members per household were recruited from 299 households in Accra, 348 in Kumasi and 268 in Tamale. Test-performance adjusted seroprevalence estimate was 42.8% (95% CI 37.5% to 48.5%) in Kumasi, 52.6% (95% CI 46.5% to 59.4%) in Accra and 81.7% (95% CI 74.6% to 89.0%) in Tamale. The poststratification for age, sex and household structure raised the overall seroprevalence estimates marginally to 43.9% (95% CI 38.1% to 50.1%) in Kumasi, 53.3% (95% CI 46.9% to 60.1%) in Accra and 84.7% (95% CI 76.4% to 95.6%) in Tamale. These estimates indicate that COVID-19 surveillance and reporting largely underestimates the true extent of infections and herd immunity in Ghana, with an estimated infection-to-case ratio of 19:1 in Kumasi; 85:1 in Accra and 49:1 in Tamale, as of March 2021, October 2021, February 2022, respectively. The seroprevalence estimate was not affected by the sex and age of the study participants. In an exposure-wide analysis, after adjusting for temporal biases in sample collection, age and sex, the average number of contacts per day was the only significant exposure variable associated with increased SARS-CoV-2 seropositivity, with an OR between 1.5 and 1.8. Conclusions: Our data highlight a sustained national transmission of COVID-19 disease through individual contacts and suggests a situation of asymptomatic endemic circulation of the SARS-CoV-2 virus and potential herd immunity.