Adherence to COVID-19 Preventive Measures Prior to Vaccine Interventions: A Meta-Analysis of Online Surveys in Sub-Saharan Africa

Background: Little is known regarding the observance of the non-pharmaceutical measures implemented to curtail COVID-19 transmission in Sub-Saharan Africa (SSA). Recent deployment of COVID-19 vaccines may lead to a neglect of these still needed non-pharmaceutical strategies. We investigated adherence to preventive measures during the initial six months of the COVID-19 outbreak in SSA.Methods: Between March and August 2020, online surveys were conducted in six SSA countries: Benin, Cameroon, Democratic Republic of Congo, Mozambique, Somalia and Uganda. A five-point individual adherence score was constituted by scoring respondents’ observance of the following: mask use, physical distancing, hand hygiene, coughing hygiene, and avoiding to touch one’s face. To take into account community preventive measures, we also proposed a modified adherence score. Data was analysed in two time periods: Period 1 (March-May) and Period 2 (June-August).Findings: Responses from 26,678 respondents were analysed (mean age: 31.0±11.1y; 54.1% males). Mean individual adherence score decreased from 3.80±1.37 during Period 1, to 3.57±1.43 during Period 2; p<0.001. Modified adherence scores showed a negative relationship with the prevalence of suspected COVID-19 among the respondents, although non-significant statistically (p=0.307). Predictors of increased individual adherence included: higher age (Coef=0.005; 95% CI: 0.003-0.007), female gender (Coef=0.071; 95% CI: 0.039-0.104), higher educational level (Coef=0.999; 95% CI: 0.885-1.113) and working in the healthcare sector (Coef=0.418; 95% CI: 0.380-0.456).Interpretation: Despite the advent of vaccines, decreasing adherence to non-pharmaceutical measures still constitute a risk for COVID-19 resurgence in SSA. Younger persons and those with lower education levels are target groups for improving adherence.Funding Information: The establishment of the ICPCovid website to conduct the surveys was supported by a grant from the European Research Council (ERC 671055). No funding source supported the drafting and dissemination of this meta-analysis.Declaration of Interests: The authors have no conflicts of interest to declare.Ethics Approval Statement: This ICPCovid surveys were approved by the Ethics Committee of the University of Antwerp, Belgium (Ref: 20/13/148), and national ethical committees in each participating country (see Author Statements for details). Only data from participants aged 18 years and above who provided an e-consent were retained for analysis. All data were collected anonymously and treated with absolute confidentiality.