A relação das políticas públicas de saúde, trabalho, educação e participação da sociedade civil sobre o índice de reincidência penitenciária : um estudo em presídio do estado do Rio Grande do Sul.

Abstract

Introduction: With the third largest prison population in the world, Brazil lacks effective strategies for the reintegration of formerly incarcerated individuals into society. Prison recidivism emerges not only as a legal issue but also as an indicator of social vulnerabilities and failures in public assistance policies, being closely related to the Social Determinants of Health (SDH). General objective of the dissertation: To analyze the rate of prison recidivism and its relationship with resocialization projects, particularly regarding public policies in health, work, education, and civil society participation aimed at the incarcerated population. Abstract of Manuscript I: This study aimed to evaluate the impact of public policies in health, education, and work on social inclusion and prison recidivism, using the Social Determinants of Health as an interpretative framework. A mixed-methods design was employed. The quantitative phase consisted of a retrospective observational study using data from the Penitentiary Information System (2019–2023), including 1,349 formerly incarcerated individuals. The outcome investigated was prison recidivism, defined as return to the prison system after release. The Chi-square test was used to assess associations between recidivism and categorical variables, identifying statistically significant associations with work during incarceration, receiving visits, and age group. These associations were then quantified using Odds Ratios. The qualitative phase involved two focus groups (n=25) with inmates from the State Prison of Lajeado, with textual analysis conducted using Iramuteq and a thematic approach grounded in the Social Determinants of Health. Data integration occurred at the interpretative stage through convergence of quantitative and qualitative findings. The recidivism rate was 86.5%. Working during incarceration (OR=2.44), receiving visits (OR=2.20), and being older than 34 years (OR=1.71) were protective factors associated with lower recidivism. Focus groups highlighted the pursuit of life transformation, emphasizing professional insertion and educational development. These factors act as structural elements of protection; however, stigma and post release barriers, particularly regarding access to employment, limit reintegration. Reducing recidivism requires intersectoral policies and strengthening of post-release social support networks. Abstract of article II: This study aimed to analyze the structural determinants of crime in Brazil, particularly the lack of access to education, poverty, unfavorable living conditions, lack of opportunities, and insecurity, articulated with social inequalities and institutional weaknesses. This is a qualitative and analytical study based on a narrative literature review. The findings reveal a complex scenario marked by increasing violence and overburdening of the prison system. Socioeconomic inequality and weaknesses in public policies related to education, health, and social assistance were identified as central factors in the cycle of crime. Education emerges as a key strategy for prevention and social inclusion, acting as a tool for reducing violence. Addressing crime requires integrated public policies combining prevention, repression, and control, along with investment in education and resocialization programs. Technical Product: The objective was to develop a tool based on machine learning algorithms to predict the risk of prison recidivism by identifying patterns in data from incarcerated individuals, aiming to optimize screening processes and support resocialization strategies within the prison system. The results demonstrated the feasibility of using predictive models as decision-support tools, without replacing professional judgment. Improving data quality and integration is recommended to enhance model accuracy. General conclusion of the dissertation: The high prison recidivism rate (86.5%) highlights structural weaknesses in post-release support, despite the presence of protective factors such as prison labor, receiving visits, and being over 34 years old. The mixed-methods analysis, grounded in the Social Determinants of Health, emphasized that social support networks, access to education, and addressing stigma are central elements for resocialization. Qualitative findings reinforce that individual efforts are insufficient in the face of systemic barriers. Machine learning proved to be a promising decision-support tool, although improvements in information systems are necessary. Reducing prison recidivism requires humanized management, technological investment, and strengthening of intersectoral public policies in the post-release context.