Algorithm for predicting the most frequent causes of mortality by analyzing age and gender variables.

Abstract

High density of populations in cities, complexity of risk factors that influence health and the impact of inequalities in sanitary outcomes, call for the adoption of decisive measures to improve health, and thus avoid injuries that trigger pathological events   directly leading to the death of people. The above applies to Colombia and especially to Bogota D.C.; after the massive health crisis due to the pandemic.  Consequently, it was proposed to implement a Prediction Algorithm based on a database directly taken from Salud Data and Salud Capital, which registered 31,720 deaths in Bogota in 2016, representing a rate of 397.5 deaths per 100,000 inhabitants, leading the list of the top ten causes: ischemic heart disease, with a rate of 65.8%, chronic respiratory tract diseases, with a rate of 26.4%, and cerebrovascular diseases with a rate of 25.7% per 100,000 inhabitants. The above data have shown the need to find a death prediction system, since it was difficult to predict the number of deaths that the pandemic was going to cause. It should be understood that the causes of mortality maintain a direct relationship with the medical study, as it evolves and develops according to the processing of the data obtained from the causes of mortality. Obtaining a good prediction system based on the data obtained greatly helps the medical area to centralize more efforts to counteract diseases with a higher rate, seeking to reduce the most significant causes of mortality. The algorithm designed analyzed two variables age and gender to predict the probability of death of a person with a percentage of 94.66% accuracy.