Epidemiological Profile and Factors Associated with Work Absence Due to Illness Among Brazilian University Public Servants: A Historical Series Study and Comparison of Predictive Models

• Grant number: 25/02773-4
• Support Opportunities: Regular Research Grants
• Start date: July 01, 2025
• End date: June 30, 2027
• Field of knowledge: Health Sciences - Collective Health - Epidemiology
• Principal Investigator: Adriano Dias
• Grantee: Adriano Dias
• Host Institution: Faculdade de Medicina (FMB). Universidade Estadual Paulista (UNESP). Campus de Botucatu. Botucatu , SP, Brazil
• Associated researchers: João Marcos Bernardes ; Juan Gómez Salgado ; Juan Ramon Lacalle-Remigio

Abstract

Absenteeism is a significant indicator in workers' professional trajectories, affecting productivity and reflecting health-related issues. Sickness absences, especially long-term or recurrent ones, are linked to occupational and non-occupational diseases, contributing to premature labor market exit. Among the leading causes, mental health conditions stand out as major obstacles to work reintegration. In Brazil, the aging workforce within the public sector raises concerns, as older employees face increased health risks and prolonged sickness absences, highlighting a critical knowledge gap in the epidemiology of absenteeism among senior public servants.This study aims to describe the epidemiological profile of public servants at a state university in São Paulo who were absent from work due to illness between 2010 and 2022. It seeks to analyze approximately 19,000 workers, including over 6,000 aged 60 and older, regarding the duration of sickness absences, associated diagnoses, and outcomes (full return, work readaptation, or retirement). Additionally, it will investigate factors associated with long-term absenteeism and retirement.This historical cohort study will use secondary data on approximately 64,000 sickness absence episodes recorded in a university occupational health database. Descriptive and analytical approaches will be employed, including exploratory epidemiological profiling and two case-control studies on long-term absenteeism and post-absence outcomes. Data will be processed, cleaned, and analyzed using IBM® SPSS and R software. Traditional statistical models, such as logistic regression, will be compared with machine learning techniques, particularly Targeted Maximum Likelihood Estimation, to estimate causal associations between sociodemographic, occupational, and health-related factors and absenteeism outcomes.While not aimed at immediate practical solutions, this study has significant potential to contribute to occupational health and public sector workforce management. Findings will provide evidence-based insights for policymakers to develop strategies promoting worker health, reducing long-term absenteeism, and ensuring workforce sustainability (AU)