Hospitalizations For Abortion Complications in Northeast Brazil, 2014-2024: A Temporal Analysis.
TL;DR
Despite the overall reduction, regional, age-related, and racial disparities persist, underscoring the need for reproductive health policies targeted at the most vulnerable groups.
Abstract
Objective
To analyze the profile and temporal trend of hospitalizations due to abortion complications in the Northeast region of Brazil from 2014 to 2024.
Methods
A time-series study using data from the Unified Health System's Hospital Information System (SIH). Hospitalizations of women of childbearing age (10 to 49 years) were analyzed. Hospitalization rates were calculated per 10,000 women of childbearing age, and the temporal trend was analyzed using the Prais-Winsten regression model with robust variance, allowing for the estimation of slope coefficients (β), coefficients of determination (R²), and statistical significance (p < 0.05) using R software.
Results
A total of 685,569 hospitalizations due to abortion complications were identified. A statistically significant downward trend was observed in most states, except in Piauí (p-value = 0.106). The highest mean hospitalization rates occurred in Bahia (38.1 ± 2.6 per 10,000 women) and Piauí (38.1 ± 2.0 per 10,000 women), while the lowest was in Pernambuco (30.9 ± 3.2 per 10,000 women). Hospitalizations were most frequent in the 20-29 age group (58.3/10,000; p < 0.0001), with significant reductions among the 10-29 age groups (β ranging from - 0.217 to - 2.164) and an increase among women aged 40-49 (β = 0.171; p = 0.0079). Analysis by race/color revealed significant disparities: while hospitalization rates among White women showed a stable trend (β = 0.116; p = 0.322), there was a significant increase in records for Brown (β = 3.259; p = 0.0017) and Black women (β = 0.347; p = 0.0011). For analytical purposes, these two groups (Brown and Black) are presented separately in the Results to maintain statistical precision but are combined into the single category of 'Black Women' in the Discussion to address structural inequalities, following IBGE conventions. Concurrently, a marked reduction was observed in cases lacking race/color information (β = -3.702; p = 0.0010), which may indicate progressive improvement in the quality of SIH/UHS data.
Conclusion
Despite the overall reduction, regional, age-related, and racial disparities persist, underscoring the need for reproductive health policies targeted at the most vulnerable groups.