Abstract Background Previous literature shows systematic differences in health according to socioeconomic status (SES). However, there is no clear evidence that the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection might be different across SES in Portugal. This work identifies the coronavirus disease 2019 (COVID-19) worst-affected municipalities at four different time points in Portugal measured by prevalence of cases, and seeks to determine if these worst-affected areas are associated with SES. Methods The worst-affected areas were defined using the spatial scan statistic for the cumulative number of cases per municipality. The likelihood of being in a worst-affected area was then modelled using logistic regressions, as a function of area-based SES and health services supply. The analyses were repeated at four different time points of the COVID-19 pandemic: 1 April, 1 May, 1 June, and 1 July, corresponding to two moments before and during the confinement period and two moments thereafter. Results Twenty municipalities were identified as worst-affected areas in all four time points, most in the coastal area in the Northern part of the country. The areas of lower unemployment were less likely to be a worst-affected area on the 1 April [adjusted odds ratio (AOR) = 0.36 (0.14–0.91)], 1 May [AOR = 0.03 (0.00–0.41)] and 1 July [AOR = 0.40 (0.16–1.05)]. Conclusion This study shows a relationship between being in a worst-affected area and unemployment. Governments and public health authorities should formulate measures and be prepared to protect the most vulnerable groups.
【초록키워드】 COVID-19, coronavirus disease, SARS-CoV-2, coronavirus, Infection, Prevalence, Health, Portugal, Evidence, Odds ratio, confinement period, Health Service, acute respiratory syndrome, measure, cumulative, logistic regressions, repeated, country, AOR, likelihood, PROTECT, Result, defined, identify, less, determine, analysis, groups, different time point, public health authority, 【제목키워드】 coronavirus, Pandemics, ecological,