Evaluation of a national program to distribute free face masks in Uganda: Evidence from Mbale District.

backgroundCOVID-19 posed a major threat to countries around the world, but many nations in sub-Saharan Africa avoided large-scale outbreaks. In Uganda, the government first enacted strict lockdowns but later focused on public health policies like masking and distancing. The government also embarked on an ambitious campaign to deliver a free face mask to all Ugandan citizens (approx. 30 million masks). We test whether mask distribution, and public education and encouragement of mask use by community health volunteers, affected mask behavior.

methodsWe collected data about mask behavior before and after masks were distributed in the Mbale district of Uganda. Trained enumerators directly observed mask wearing in public places and asked about mask use via phone surveys. We compared observed and self-reported mask behavior before and after masks were distributed. We also tested whether training volunteers from randomly selected villages to educate the public about COVID-19 and masks affected behavior, attitudes, and knowledge among mask recipients.

resultsWe collected 6,381 direct observations of mask use at baseline (February 2021) and 19,855 observations at endline (April 2021). We conducted a listing of 9,410 households eligible for phone surveys and randomly selected 399 individuals (4.2%) at baseline and 640 (6.8%) at endline. Fewer than 1% of individuals were observed wearing masks at baseline: 0.9% were seen with a mask and 0.5% wore masks over mouth and nose. Mask wearing significantly increased at endline but remained low: 1.8% of people were observed with masks and 1.1% were seen wearing masks correctly after the distribution campaign. At the same time, a high proportion of people reported using masks: 63.0% of people reported using masks at baseline and 65.3% at endline when walking around their villages. When respondents were asked about mask use in public places, 94.7% reported using masks at baseline and 97.4% reported using masks at endline. We found no differences in knowledge, behavior, or attitudes among mask recipients in villages where volunteers were tasked with conveying information about COVID-19 and masks during distribution.

conclusionMask use remained low in Mbale district of Uganda during study observation period even after free masks were distributed. Encouraging new health behaviors may need to involve more intensive interventions that include reminders and address social norms.