The proposed project, which responds to NOT-AG-21-033 (Health Disparities and Alzheimer’s Disease), is the
first longitudinal and epidemiological examination of justice-involvement and dementia. Americans who
experience justice-involvement, moving from the community to incarceration and from incarceration back to the
community, are aging rapidly. Moreover, justice-involvement disproportionately impacts persons of color and of
low socioeconomic status, who are also at high risk for all forms of dementia yet are frequently excluded from
dementia-related research. This study will contribute to our understanding of health disparities in dementia by
evaluating the inter-relationship among incarceration, dementia, and a number of important social risk factors
and multi-morbidity factors over time in a national sample. We will optimize the Veterans Health
Administration’s (VHA) capacity to access Incarceration History Files from the Centers for Medicare and
Medicaid Services (CMS). We propose to leverage a longitudinal “prison release” cohort which we created for
a different NIH-funded study and supplement (R01 MH117604 and S1). This cohort includes >30,000
individuals enrolled in both Medicare and VHA, who reentered the community at age ≥50 (between 2008-2018)
following an incarceration. We will enhance this cohort to include all veterans who were age ≥50 between 2008
to 2022, including those incarcerated at any point during that timeframe and who were released, those who
remained incarcerated (no community reentry), and those never incarcerated, for a total of more than 46,000
justice-involved veterans among 10.5 million veterans. Using this rich, longitudinal dataset, we have the
unprecedented opportunity to develop social risk factor profiles and multi-morbidity profiles associated with risk
of incarceration among those with dementia (Aim 1) and risk of dementia among those released from
incarceration (Aim 2). Access to VA and Medicare data provides more complete outcome ascertainment, thus
increasing the opportunity to determine if these profiles differ according to dementia subtypes (e.g.,
Alzheimer’s Disease, vascular, frontotemporal, mild cognitive impairment (MCI)). We will also make novel use
of transition models to evaluate probabilities of experiencing distinct pathways (e.g., pathway to entering
incarceration, to being released from incarceration, to developing dementia) and determine precursors that
predict these pathways (Aim 3). We will evaluate all Aims within health disparity populations such as non-
Hispanic Black, Hispanic, native/indigenous groups and those experiencing indicators of poverty (e.g.,
homelessness and socioeconomic disadvantage). This study has substantial public health significance. It will
inform strategies to mitigate risk of patients with dementia entering incarceration, will delineate those at highest
risk of developing dementia after incarceration so as to inform prison-to-community transitional care planning,
and will identify prime intervention points where optimizing dementia care could reduce health disparities
between those with and without justice-involvement and even within the justice-involved.
