Exploring the association between precipitation and population cases of ocular toxoplasmosis in Colombia

Abstract

Background Previous studies suggest a relationship between precipitation and ocular toxoplasmosis (OT) reactivation and congenital toxoplasmosis infection. We aimed to investigate the relationship between precipitation and the frequency of new OT cases in Colombia from 2015 to 2019. Methodology This retrospective cohort study analyzed data obtained from a claims-based database created by the Colombian Ministry of Health and national registries of precipitation of the Institute of Hydrology, Meteorology, and Environmental Studies. We estimated the daily number of OT cases, interpolating data from the average number of annual cases from 2015 to 2019. Then, we compared exposures (mean daily precipitation) in the case period in which the events (interpolated OT new cases) occurred by a quasi-Poisson regression, combined with a distributed lag non-linear model to estimate the non-linear and lag–response curve. Principal findings In the 5-year analysis, there were 1,741 new OT cases. Most of the cases occurred in 2019, followed by 2015 and 2018. New OT cases among departments were significantly different (PConclusions Precipitation influenced the RR for new OT cases. However, varying trends among geographical regions (departments) lead us to hypothesize that other sociodemographic, behavioral, and environmental variables, such as wind and water contamination, could influence the RR. Author summary We analyzed data obtained from the Colombian Ministry of Health and the national meteorology center to determine the effect of precipitation on the new cases of ocular toxoplasmosis (OT). The number of interpolated daily cases was estimated and compared with the exposure (precipitation) by a quasi-Poisson regression, combined with a distributed lag non-linear model to estimate the lag, non-linear response curve, and a Pearson correlation test. In the 5-year study period, 1,741 new cases of OT were reported. We found significant differences in the trends among all departments, with most departments showing decreasing cumulative exposure-response curves. However, in Chocó, Bogotá, Cesar, Cauca, and Guajira, when a certain amount of precipitation accumulates, the relative risk (RR) increases, contrary to the pattern observed in the rest of the country. The response curves at one-day lag showed that precipitation influences the RR; however, trends vary by department. We found a positive correlation between the number of cases and precipitation (β = 0.03; P< 0.05), indicating that precipitation affects the RR of new cases of OT. However, inconsistent trends exist across geographical regions, leading us to hypothesize that other sociodemographic, behavioral, and environmental variables, such as wind and water contamination, might influence the RR.