Background: In Ethiopia, the first case of COVID-19 was reported on March 15, 2020. As of 8 December 2020, the number of people infected with COVID-19 had reached 114,266, of whom 84,948 have recovered and 1,766 have died. The government of Ethiopia declared a state of emergency and detailed COVID-19 pandemic prevention guideline to combat the spread of the pandemic. Unintended consequences of COVID-19 responses are becoming visible across sub-Saharan Africa. This study aims to understand the impact of COVID-19 and government response measures on poor and vulnerable groups in urban areas in Ethiopia.

Methods: This study used a mixed method approach with a panel study design. Semi-structured interviews were conducted among randomly selected 450 households (i.e. 45 per city) who are enrolled in urban productive safety net program (UPSNP), small scale business (SSB) and IDP/refugee populations in 10 cities (Mekelle, Bahir Dar, Semera, Logiya, Addis Ababa, Adama, Ble Hora, Dire dawa, Jigiga, and Gambela). For the qualitative study we conducted a total of 50 diary-style interviews (an average of five per city) with selected respondents from urban productive SafteyNet beneficiaries, SSB owner respondents, IDPs/refugees, and from the ‘special population group’ (i.e. daily labourers, shoeshines, waiters, porters, and commercial sex workers). Moreover, In-depth interviews were conducted with a total of 40 respondents (i.e. 4 per city) representing health care providers, relevant government offices NGOs and civil society organizations. All the interviews were conducted through telephone calls.

Results: There appeared to be fatigue with respect to the disease and there are misconceptions related to COVID-19, particularly in terms of the severity of its effects and even its existence (and people who are asymptomatic), which seemed to have contributed to the limited practising of preventive measures. Respondents without access to clean water nearby were struggling to wash their hands and keep up with the required level of hygiene. Most respondents struggled with increased food prices and even though they managed to consume three meals a day. Most respondents or members of their families who needed medical attention reported having been able to access medical treatment. However, some people still chose not to visit health facilities due to fear of exposure to COVID-19, or of being sent to a quarantine centre. During school closure, children mostly spent their time playing around the neighbourhood, helping their family, and watching TV. Only a small proportion spent their time studying. Overall, support was reported to be available during the COVID-19 pandemic, but was perceived to be inadequate.

Conclusions: Misconceptions about COVID-19 and compliance to restrictions must be addressed with awareness-raising activities and more tailored interventions for certain groups. Most of respondents had access to water supply from nearby or inside their resident; however, sporadic availability of a water supply remained major challenge to access water for household use and to practice proper handwashing. Maternal health service utilisation does not seem to be significantly impacted, although our sample for this group was small, and thus the results need to be interpreted with caution.