Background: Geospatial factors can affect SARS-CoV-2 transmission and have been used to prioritise COVID-19 surveillance and response activities. However, no information exists on the geospatial factors associated with SARS-CoV-2 prevalence in Zambia. We assessed the relationship of geospatial factors and SARS-CoV-2 prevalence during the first wave in Zambia.
Methods: We did a cross-sectional study of SARS-CoV-2 prevalence in 5 districts in July 2020, during the upslope of the first COVID-19 wave in Zambia. In each district, 16 enumeration areas (EAs) were randomly selected and 20 households from each EA were sampled. SARS-CoV-2 prevalence was calculated as the number of persons with a positive-SARS-CoV-2 PCR test divided by the number tested. We analysed EA geospatial data for population density; socioeconomic status (SES) (lower scores indicate lower vulnerability); literacy; and access to water, sanitation, and hygiene. Generalized estimating equations were used to measure adjusted odds ratios (aORs) and 95% confidence intervals (CI) for SARS-CoV-2 prevalence with geospatial factors, adjusting for clustering by district, in R.
Results: Overall SARS-CoV-2 prevalence was 76 infections per 1,000 population during the study period, with a median EA prevalence of 45 infections per 1,000 population (interquartile range, 0–111). Less dense, lower SES vulnerability, and urban EAs had greater odds of high SARS-CoV-2 prevalence (population density aOR: 0.95 [0.95–0.99]; SES vulnerability aOR: 0.45 [0.30–0.69]; urban aOR: 1.51 [95% CI=1.03–2.22]).
Conclusions: SARS-CoV-2 prevalence was higher in sparsely-populated, wealthier, urban EAs early in the COVID-19 epidemic in Zambia. The timing of this study early in the first wave could have impacted our findings, with wealthier enclaves affected earlier because of greater linkages to international settings. Additional analyses from subsequent waves could confirm if these findings persist or changed as SARS-CoV-2 became endemic in Zambia. During the beginning of a COVID-19 wave in Zambia, surveillance and response activities might focus on urban population centres.