Incarceration's Syndemic Effects: Tobacco Smoking, Alcohol Use and Cardiovascular Disease

The United States continues to have the highest incarceration rate in the world. Chronic stress induced by incarceration is associated with risk factors for cardiovascular disease (CVD) and exacerbation of existing health disorders. Additionally, justice-involved persons smoke  tobacco and use alcohol at higher rates than the general population; interventions to address either behavior during incarceration have demonstrated limited efficacy. Incarceration-associated impacts, moreover, are heterogeneous; Black men are incarcerated at far greater rates than their White or Latino counterparts, leading to greater stressors among them and their communities. Consequently, we are seeing a syndemic of tobacco smoking, alcohol use, and cardiovascular disease (CVD) in Black men involved with the criminal justice system and their
communities that urgently requires a public health response. However, greater understanding of how these factors act together to impact CVD in communities of justice-involved persons is needed, which can in turn inform estimates of how criminal justice reform may positively impact this syndemic. 

Two challenges exist in understanding how chronic diseases and substance use disorders are exacerbated among communities of justice-involved persons are impacted by incarceration. First, the effects of syndemic conditions may be realized over time periods that are longer than the duration of most empirical studies. Second, the social network communities of justice-involved persons might not be involved in the criminal justice system themselves and therefore may not be considered in health studies on incarceration. Approaches that combine empirical data with complex systems science modeling methods can help quantify the impacts of a syndemic that are harder to estimate through purely empirical approaches. Computational modeling has unique strengths: (1) allowing investigation of the impact of policy interventions prior to their real-world implementation; (2) providing rigorous methods to quantify uncertainties given that data on policy parameters are often sparse, uncertain or unavailable. In particular, agent-based models, a dynamic systems modeling technique, provide flexibility in modeling individual persons and members of their community networks as "agents", and the social network structures that connect them as "ties". The coevolution of agents and networks provides practical insight on the impacts of health policy implementations.

In this project, an agent-based network model, parameterized with incarceration and recidivism patterns as measured from cohort data for Black men in Chicago, will be developed. Risk factors for tobacco smoking, alcohol use, and chronic stress will be estimated using existing clinical laboratory data from project mentors (Drs. Kahler, Martin, Cioe). Social network parameters on incarcerated Black men will be obtained from an ongoing NIDA study (PI Schneider, Co-I Khanna), the modeling component of which has been led by PL Khanna to simulate the impact of incarceration and recidivism on HIV outcomes among Black men and their sexual partners and to investigate networks of stimulant-using men who have sex with men (MSM) populations. A synthetic population, generated in a virtual computational laboratory, will be simulated using powerful computing supercomputing resources at Brown University to investigate the following aims:

Aim 1: Quantify the impact of incarceration on (a) smoking and alcohol use and (b) cardiovascular disease among Black men and persons in their social networks. We hypothesize that simulated data will show that length and frequency of incarceration events and lifetime number of incarceration episodes are: (Hypothesis 1a) associated with greater tobacco and alcohol use and (Hypothesis 1b) increased CVD incidence among justice-involved persons and their social network communities. The models developed in the proposed project will allow for investigating causal mechanisms between incarceration, smoking, alcohol use, and CVD risk factors.

Aim 2: Predict how various decarceration policies (e.g., fewer and/or shorter detentions) are likely to impact smoking and alcohol use and CVD incidence among justice-involved persons and their social network communities. Hypothesis 2: Given the evidence in the literature, we hypothesize that incarceration is causally linked with smoking and alcohol use and CVD risk factors. The models developed here will allow us to measure the strength of a causal relationship, if it exists, and study how decarceration policies might decrease smoking and alcohol use and CVD onset.

The proposed project will provide data on the population-level impacts of potential policy reforms, a context in which health data are difficult to obtain and have broad-reaching policy implications.