Physico-mathematical prediction of the biannual dynamics of malaria in Antioquia, Colombia

Abstract

Introduction: Malaria is an endemic disease in Colombia. Due to its biological and pathophysiological characteristics and socioeconomic impact, it is positioned as a disease of public health interest.

Objective: To predict the number of reported cases of malaria for the second semester of 2020 in Antioquia department using the theory of probability and probabilistic random walk.

Methods: A retrospective, longitudinal, observational study was conducted. The geometric dynamics of the biannual development of malaria epidemic in Antioquia department from 2008 to 2020 was analyzed using the data from the National Surveillance System in Public Health as a probabilistic random walk. Regarding the consecutive biannual reductions and increments, the most probable number of infected individuals was determined for the second semester of 2020.

Results: The predictive value was 3433. When it was compared to the values reported by the National Surveillance System in Public Health, a 95.4% accuracy rate was obtained.

Conclusions: By predicting with high accuracy the number of infected individuals for the second semester of 2020 in Antioquia department, it is evident the methodology potential as a public health surveillance tool and as a means to support public policy decision making.