Assessing predictive factors of COVID-19 outcomes: a case-control study in the metropolitan region of São Paulo (Brazil)

Abstract

Background and objectives: the aim of this retrospective cohort study was to search individual, sociodemographic and environmental predictors of COVID-19 outcomes. Materials and methods: a convenience sample of 1036 COVID-19 confirmed patients (3–99 years, mean 59 years; 482 females) who sought treatment at the emergency units of the public health system of Diadema (Brazil; March-October 2020) was included. Primary data were collected from medical records: sex, age, occupation/education, onset of symptoms, presence of chronic diseases/treatment and outcome (death and non-death). Secondary socioeconomic and environmental data were provided by the Department of Health. Results: the mean time spent between COVID-19 symptom onset and admission to the health system was 7.4 days. Principal component analysis summarized secondary sociodemographic data, and a Poisson regression model showed that the time between symptom onset and health system admission was higher for younger people and those from the least advantaged regions (availability of electricity, a sewage network, a water supply and garbage collection). A multiple logistic regression model showed an association of age (OR = 1.08; 1.05–1.1), diabetes (OR = 1.9; 1.1–3.4) and obesity (OR = 2.9; 1.1–7.6) with death outcome, while hypertension and sex showed no significant association. Conclusion: the identification of vulnerable groups may help the development of health strategies for the prevention and treatment of COVID-19.