Temporal trend and spatial analysis of the HIV epidemic in young men who have sex with men in the second largest Brazilian Amazonian province

Abstract

After 40 years of its starting, the HIV epidemic in Brazilian Amazon region remains on an increasing trend. The young men who have sex with men (MSM) have been the most impacted by the HIV in the last decade. However, much more than attributing the risk behavior to HIV uniquely to the individual, behaviors are shaped by social determinants of health (SDH). Despite the problem, there is a scarcity of studies evaluating the impact of SDH on HIV among young MSM and none of them were done in the Northern of Brazil. Therefore, the main goal of this study was to analyse the HIV epidemic among Brazilian Amazonian young MSM using temporal trends and spatial analysis.

We conducted an ecological study using reported cases of HIV/AIDS in young MSM living in Pará, the second larger Brazilian Amazonian province, between 2007 and 2018. Data were obtained from the Information System for Notifiable Diseases. For the temporal analysis, we employed a Seasonal and Trend decomposition using Loess Forecasting model (STLF), which is a hybrid time-series forecast model, that combines the Autoregressive-Integrated Moving Average (ARIMA) forecasting model with the Seasonal-Trend by Loess (STL) decomposition method. For the spatial analysis, Moran’s spatial autocorrelation, spatial scan, and spatial regression techniques were used.