Modelling the Impact and Public Health Response to COVID 19 in Uganda

Background: As of June 18, 2020, Uganda had registered 741 COVID-19 cases, and several intervention measures have been implemented to control the spread of the SARS-CoV2. With increasing risk of community transmission, it is important to forecast the extent on the COVID-19 epidemic in Uganda. This study aimed to design a predictive model to provide reliable estimates for COVID-19 in Uganda. Methods: A stochastic model of a modified SEAIR (Susceptible, Exposed, Infected Asymptomatic, Infected Symptomatic and Removed) type, which made assumptions on the local SARS-CoV2 transmission, was used to forecast the impact of COVID-19 in relation to the varying levels of compliance to mitigation measures. Results: Our results predict between 2,000 to 4,000 cases within 100 days from lifting the lockdown if no or weak interventions are in place. However, if social distancing was implemented at 40% effectiveness, to prevent interaction between susceptible and infected populations with no use of face-masks, we would expect 200 to 400 cases. The number of infections decreases with increase in the level of effectiveness of social distancing measures and the curve flattens at ≥60% level of social distancing. Wearing face-masks at ≥40% consistency and social distancing at 50% reduces the number of infections and flattens the curve. Interpretation: The prompt public health response in Uganda has so far limited COVID-19 to imported cases, comprising mainly of truck drivers. Although lifting the lockdown measures is feasible, social distancing strategies, as the most effective measure, should be maintained at high levels and supplemented with wearing face-masks. Funding Statement: None to declare Declaration of Interests: All authors have no reported conflicts of interest. Ethics Approval Statement: Not required