Spatial distribution and spatiotemporal clustering of AIDS notifications among men in Brazil, 2014–2023: an ecological study

This study aimed to analyse the spatial distribution and spatiotemporal clustering of AIDS notifications among men in all 5570 Brazilian municipalities between 2014 and 2023. Using an ecological design, data on AIDS notifications among men (≥ 13 years) were retrieved from Brazil’s Notifiable Diseases Information System database. Spatial analysis included global autocorrelation via Moran’s I and the identification of local clusters via the Local Indicators of Spatial Association applied to Empirical Bayes smoothed rates. Retrospective and prospective (active) space–time scan statistics were applied to detect clusters of elevated AIDS notification rates. Additionally, the Spatial Variation in Temporal Trends (SVTT) method was used to assess differences in notification trends within and outside these clusters. The global Moran’s I statistic indicated positive spatial autocorrelation for the entire period ( p < 0.01). High-high clusters were predominantly located in the North, Northeast and South Regions, whereas low-low clusters were concentrated in the interior of the Northeast and Southeast Regions. Twenty-seven retrospective clusters (maximum RR: 3.12; p < 0.01) were identified, with the highest relative risks observed in the North Region. Prospective analysis revealed 24 active clusters (RR: 1.22–3.22; p < 0.01) in predominantly urban areas, especially in the Amazon region, including Manaus and Belém. SVTT analysis identified six clusters where the temporal trend in notification rates differed significantly from the trend observed outside them ( p < 0.01), with inside trends of up to 24.971% annually; this trend was not necessarily accompanied by a correspondingly high absolute notification rate or relative risk. These findings highlight substantial geographic inequalities in AIDS notification rates among men in Brazil. Metropolitan and peri-urban areas of the Amazon show disproportionately elevated notification rates. These patterns may reflect inequalities in access to HIV diagnosis, delayed diagnosis of infection, and differences in access to or adherence to antiretroviral therapy, although none of these mechanisms were directly measured in this study. Spatial surveillance may support the identification of territories where barriers to diagnosis and care require closer public health attention.