Sistema de suporte à decisão espacial para assistência do transtorno do espectro autista na cidade de João Pessoa-PB

Abstract

This study aimed to develop a spatial decision support system to prioritize assistance for Autism Spectrum Disorder (ASD) in João Pessoa - PB, identifying and mapping neighborhoods with the greatest need for support based on geospatial data. It is an exploratory cross-sectional, and quantitative study using data from a public service on ASD diagnoses from 2018 to 2023. The sample included 1,863 records, excluding individuals without an explicit ASD diagnosis, those who did not reside in the municipality, and records with missing information. Socioeconomic and sociodemographic variables extracted from electronic medical records were analyzed, including location, age, sex, race/skin color, income, and referral type. The analysis involved statistical tests, spatial and spatiotemporal analysis, and a fuzzy rule-based system. The study was approved by the Research Ethics Committee (approval number 7.454.640). The results indicated a progressive increase in cases, with the highest incidence in the 1 to 4-year age group (34.6% to 48.4%) and among males (80.9%). Most cases were of mixed-race individuals (60.2% to 96.6%) and from families with an income of 1 to 2 minimum wages (91.0% to 97.6%). Spatial analysis revealed significant concentrations in the North, East, and South zones, with the Spatial Incidence Ratio (SIR) ranging from 0 to 7.71 in 2018, 0 to 6.63 in 2019, 0 to 11.18 in 2020, 0 to 3.81 in 2021, 0 to 4.05 in 2022, and 0 to 3.33 in 2023. Some neighborhoods had an SIR more than 2.0 times the municipal average. The Scan statistics analysis identified spatial clusters per year. The spatiotemporal analysis showed persistent clusters, indicating significant clusters in time and space, with a likelihood ratio (LLR = 18.11) and persistence of up to 3 years. The fuzzy rule-based system generated 321 rules, suggesting the following decision regarding neighborhoods: 7 priority neighborhoods, 8 tending toward priority, 5 tending toward non-priority, and 44 non-priority neighborhoods for ASD support. This study contributes to the understanding of ASD distribution in João Pessoa through a spatial decision support system. The application of the fuzzy rule-based system allowed for the suggestion of neighborhoods based on priority for assistance, providing a foundation for public policy planning. The results indicate the need for targeted interventions in areas with greater needs, which can improve access to and effectiveness of healthcare services, promoting more efficient assistance for individuals with ASD.