ABSTRACT
BACKGROUND: Coronavirus disease 2019 (COVID-19) has spread worldwide, causing a high burden of morbidity and mortality, and has affected the various health service systems in the world, demanding disease monitoring and control strategies. The objective of this study was to identify risk areas using spatiotemporal models and determine the COVID-19 time trend in a federative unit of northeastern Brazil. METHODS: An ecological study using spatial analysis techniques and time series was carried out in the state of Maranhão, Brazil. All new cases of COVID-19 registered in the state from March 2020 to August 2021 were included. Incidence rates were calculated and spatially distributed by area, while the spatiotemporal risk territories were identified using scan statistics. The COVID-19 time trend was determined using Prais-Winsten regressions. RESULTS: Four spatiotemporal clusters with high relative risks for the disease were identified in seven health regions located in the southwest/northwest, north and east of Maranhão. The COVID-19 time trend was stable during the analysed period, with higher rates in the regions of Santa Inês in the first and second waves and Balsas in the second wave. CONCLUSIONS: The heterogeneously distributed spatiotemporal risk areas and the stable COVID-19 time trend can assist in the management of health systems and services, facilitating the planning and implementation of actions toward the mitigation, surveillance and control of the disease.
ABSTRACT
BACKGROUND: The detection of spatiotemporal clusters of deaths by coronavirus disease 2019 (COVID-19) is essential for health systems and services, as it contributes to the allocation of resources and helps in effective decision making aimed at disease control and surveillance. Thus we aim to analyse the spatiotemporal distribution and describe sociodemographic and clinical and operational characteristics of COVID-19-related deaths in a Brazilian state. METHODS: A descriptive and ecological study was carried out in the state of Maranhão. The study population consisted of deaths by COVID-19 in the period from 29 March to 31 July 2020. The detection of spatiotemporal clusters was performed by spatiotemporal scan analysis. RESULTS: A total of 3001 deaths were analysed with an average age of 69 y, predominantly in males, of brown ethnicity, with arterial hypertension and diabetes, diagnosed mainly by reverse transcription polymerase chain reaction in public laboratories. The crude mortality rates the municipalities ranged from 0.00 to 102.24 deaths per 100 000 inhabitants and three spatiotemporal clusters of high relative risk were detected, with a mortality rate ranging from 20.25 to 91.49 deaths per 100 000 inhabitants per month. The headquarters was the metropolitan region of São Luís and municipalities with better socio-economic and health development. CONCLUSIONS: The heterogeneous spatiotemporal distribution and the sociodemographic and clinical and operational characteristics of deaths by COVID-19 point to the need for interventions.