Timing, Compliance and Impact of COVID-19 Containment Measures in Portugal is the title of the study published on 3rd MAy 2020 by a group of researchers of the NOVA NAtional School of Public Health - Vasco Ricoca Peixoto, André Vieira, Pedro Aguiar, Paulo Sousa and Alexandre Abrantes.
Portugal acted early to contain the COVID-19 pandemic, when it only had 62 cases per million population and almost no deaths. This epidemiological situation was much less serious than that registered in Spain, Italy or the United Kingdom when they took similar containment measures.
The Portuguese population adopted the containment and mitigation measures early and effectively, reducing drastically its mobility, including that related with retail and leisure (-83%), parks and similar (-80%) and transport (-79%). The Spaniards also adopted the measures early and effectively. Citizens of Italy and the United Kingdom were late in accepting the restricted mobility rules and had lower compliance than in Portugal and Spain.
Between April 1 and 15, 2020, Portugal registered 25% less deaths, 53% less inpatients in ICU, and 12% less hospital inpatients, and 23% fewer cases that would have been expected without containment and mitigation measures. There are strong correlation coefficients between Oxford Stringency Scores and the nº of COVID-19 daily deaths (R=0,896) and the nº of COVID-19 daily ICU inpatients (R=0.937). These results and other observed causality criteria suggest that the containment measures may be contributing to the reduction of mortality and serious morbidity due to COVID-19 (patients that require hospital or ICU admission).
To evaluate the moment in which each country adopted to contain and mitigate the effects of COVID-19, we compared the respective mortality and morbidity indicators when each country reached the threshold score of 60 and 80 of the Oxford Stringency Index. To measure the compliance of the population of each country with the respective containment and mitigation measures we used the mobility data shared by Google and Apple, comparing mobility of its users during the pandemic with that of a baseline period registered between January 3 and February 6. Finally, to estimate the potential impact of the containment and mitigation measures on mortality and morbidity due do COVID-19, we compared the official daily mortality and morbidity data, and the nº of hospital and ICU daily inpatients with estimated values, if those measures had not been enacted. Estimations were based on an exponential adjustment of the time series for deaths, nº of hospital inpatients and ICU inpatients due to COVID-19. The new cases time series was adjusted with an ARIMA model.