Análise temporal das doenças relacionadas ao saneamento básico no Brasil do século XXI: iniquidades raciais e o impacto da Covid-19

Abstract

Intestinal infectious diseases represent a persistent challenge to public health in Brazil, with a higher prevalence in areas with deficient basic sanitation. Epidemiological surveillance, when integrated with robust analytical capacity, becomes indispensable for the formulation of evidence-based public policies and for achieving effective results. This study, using statistical tools for time series analysis, investigated patterns and trends in hospitalization rates for Diseases Related to Inadequate Environmental Sanitation (DRSAI) in Brazil, from the 21st century onwards. Specific objectives included understanding the temporal dynamics of the racial aspect as a determinant factor of health inequities and developing a time series model to predict unregistered hospitalizations for DRSAI during the Covid-19 pandemic. The comparison between hospitalization rates for whites and non-whites revealed that, although both series showed a decreasing trend, the racial disparity remained constant. On average, non-white individuals had twice the risk of hospitalization due to DRSAI compared to white individuals. The results showed a 96.6% reduction in public spending on hospitalizations for DRSAI, from 294 million in 1998 to 10 million in 2022, accompanied by a change in the age profile of deaths, in addition to the increase in lethality, especially in recent years. Predictive modeling, using the SARIMA (4,1,0) (3,1,0)3 model, provided a more accurate view of the pandemic's impact on the epidemiological surveillance system of diseases due to lack of sanitation. The Covid-19 pandemic significantly affected the data, resulting in estimated underreporting of about one third of the rates predicted by the model. It is concluded that time series analysis is a valuable tool for identifying DRSAI patterns, evaluating the effect of interventions, quantifying determinants of health inequities, and delineating vulnerable populations. Predictive modeling can support action planning, especially in atypical events such as health emergencies.