Altering Administrative Segregation for Inmates and Staff: A Mixed-Methods Analysis of the Effects of Living and Working in Restrictive Housing

The effects of isolation through restrictive housing placements are at the forefront of national discussions on crime and punishment. The attention is well-deserved, as some have argued that the practice leads to significant harm to the mental health of inmates. However, not all research
documents negative outcomes associated with the practice. The present study seeks to advance our understanding of the impact of restrictive housing on the mental health of inmates and staff.
Specifically, the project will 1) determine whether a change in policy to a contingency-management approach to restrictive housing improved the mental and physical health outcomes of inmates and staff, 2) determine whether inmates in restrictive housing experience a change in mental health outcomes as compared to inmates not placed in restrictive housing and, 3) determine whether working in restrictive housing differentially impacts the mental and physical health of officers relative to officers who work in nonrestrictive housing settings. The present study capitalizes on a partnership between Arizona State University (ASU) and the Arizona Department of Corrections (ADC). Three project phases carry out the above research aims.
The first phase uses administrative data to determine whether the policy change improved staff and inmate outcomes. An interrupted time series analysis analyzes whether the patterns of inmate and staff mental and physical health were altered upon the introduction of the policy in March of 2014. The second phase uses semi-structured interviews with 200 male inmates who have been placed in restrictive housing as a form of administrative segregation. These inmates are interviewed at intake and followed up at 6 months and 12 months later, with an explicit focus on how their mental health
may change over this time period. Multilevel modeling will be used to determine changes in mental health and whether these changes are different as compared to a group of 100 inmates not placed in restrictive housing. The third phase uses semi-structured interviews with 100 officers working in
restrictive housing and 100 officers working in other security settings. Quantitative and qualitative analyses are performed to determine whether working in restrictive housing produces mental and physical distress above and beyond that experienced working in other units.

Taken altogether, the study takes place in a large and racially and ethnically diverse prison system, and is bolstered by quasi-experimental mixed methods, a prospective longitudinal design, relatively large sample size and follow-up periods, comparison groups, and aggregate and individual-level
analyses. ca/ncf