Spatial-temporal analysis of hospitalizations with death caused by oral cancer in Brazil and its correlation with the expansion of healthcare coverage

ABSTRACT

Background: Oral cancer (OC) is a growing public health problem worldwide. In Brazil, the National Oral Health Policy, implemented in 2004, expanded access to oral health services and prioritized OC care. However, it is not known whether this expansion resulted in a reduction in hospital admissions with death. This study aimed to analyze the proportion of hospital admissions who progressed to death due to OC in Brazil from 2007 to 2019 and its correlation with the coverage of health services. Material and Methods: This study is an ecological, longitudinal, and analytical study of hospital admissions with death due to OC recorded in the Brazilian Hospital Information System. The following analyses were performed descriptive, spatial (choropleth maps and Moran index), and negative binomial regression, with a hierarchical approach, estimating crude and adjusted regression coefficients (β) and respective 95% confidence intervals (95% CI) (alpha=5%). Results: In 2019, Moran's index (I) of spatial autocorrelation showed a negative association between hospital admissions with death and dentist surgeon/inhabitant rate (I=-0.176), physician/inhabitant rate (I=-0.157), family health strategy (FHS) coverage (I=-0.080), oral health team (OHT) coverage (I= -0.129), dental specialty centers (DSC)/inhabitant rate (I= -0.200), and oncology bed/inhabitant rate (I= -0.101). In the adjusted regression analysis, the proportion of hospitalizations with deaths caused by OC was higher in Brazilian states with a lower medical ̸inhabitant ratio (β= -0.014; p=0.040), a lower dentists/inhabitant ratio (β= -0.720; p=0.045), a lower number of DSC (β= -0.004; p<0.000), a lower amount paid per hospitalization (β= -10.350; p<0.001), and a lower number of biopsies (β= -0.00008; p=0.010). The proportion of hospitalizations that progressed to death showed a positive association with the number of days of hospitalization (β= 0.00002; p=0.002). (AU)