Human behaviour data analysis and noncommunicable diseases: a systematic mapping study

Abstract

ABSTRACT Noncommunicable diseases (NCDs) or chronic diseases are responsible for 41 million deaths each year, equivalent to 71% of all worldwide deaths. Many technologies are used to aid the treatment of NCDs, and data analysis has been used as an approach to improve the understanding of human behaviour related to risk factors. This study aims to distinguish how human behaviour data analysis has been applied to support the treatment and prevention of NCDs, what technologies are currently used, and what gaps are still left unexplored. We conducted a systematic mapping study to analyse academic articles published from 2010 to September 2021. A filtering process mitigated article bias by reviewing, analysing, and classifying 41 works from 12,395 collected. The main results obtained presented that 43% applied data analysis in depression, 17% applied for general NCDs, and 12% for diabetes. Whereas, machine learning represents 60% of technologies found in the articles, mobile devices 58%, and wearables 29%. This study proposes two taxonomies obtained from the analysis of the selected articles that allow systematised guides to access the knowledge produced in the study. In addition, the taxonomies link technologies used to identify human behaviour with associated NCDs.