Returning Home–Ohio (RHO) aimed to reduce recidivism, homelessness, and use of multiple systems and services among incarcerated persons with disabilities who had been released from prison and were returning to the community. RHO linked prisoners who had disabilities and a risk or history of housing instability with supportive housing and referrals to services.
To be eligible for the program, participants had to have been disabled (broadly defined to include developmental disorders, severe addiction, and behavioral health problems), close to their Ohio Department of Rehabilitation and Correction (ODRC) release date, and either homeless at the time of their arrest or at risk for homelessness upon release.
The ODRC funded and partnered with the Coalition for Supportive Housing (CSH) between 2006 and 2007 to design the RHO program. Nine service providers across five Ohio cities provided supportive housing services, which included a combination of permanent affordable housing and supportive services. The five participating communities were Dayton, Cincinnati, Cleveland, Columbus, and Toledo. While each provider served a different target population and implemented housing and services slightly differently, the core components of the program across sites included housing placement and a referral to at least one service related to education, mental health, substance use, or another category such as community service, access to a gym, anger management, and financial literacy training.
Individual providers had specialized experience providing housing support for different populations, including veterans, women, individuals with severe mental illness, and indigent populations. Some providers had access to multiple, scattered units with private landlords; others owned or managed one affordable housing property where RHO participants were provided housing in units among non-RHO participants.
The staff at the CSH were trained in coordinated case management, mental health and health services, substance use treatment, and vocational and employment services. The ODRC facilitated the recruitment process of program participants.
Past research has found that providing supportive housing to individuals with homelessness and mental health histories reduces the chances of recidivism, dependence on corrections, use of emergency services, and shelter systems (Fontaine et al. 2012).
Although the evaluation of the RHO pilot program by Fontaine and colleagues (2012) found that program participants had lower probability of rearrest and spent longer time in the community, they had statistically significantly more rearrests at the 1-year follow up, compared with those who did not participate.
Fontaine and colleagues (2012) found that that individuals in the Returning Home-Ohio (RHO) treatment group had a lower probability of rearrest at the 1-year follow up. This difference was statistically significant.
There was no statistically significant difference between the RHO treatment and comparison groups on the probability of re-incarceration.
Days to Rearrest
The results showed that the length of time from release to rearrest was greater for individuals in the RHO treatment group than the comparison group. This difference was statistically significant.
Number of Rearrests
However, the results showed that the individuals in the RHO treatment group had more rearrests than the control group. This difference was statistically significant.
Fontaine and colleagues (2012) used a quasi-experimental design to evaluate the impact of participation in the Returning Home-Ohio (RHO) supportive housing program on recidivism outcomes.
The treatment group (N=121) consisted of individuals who participated in the program, while the comparison group (N=118) was composed of individuals who were referred to the program, but were not provided housing in the community. Individual providers had the discretion to accept or reject eligible participants who were referred to them.
The average age of individuals in the treatment group was 42 years old, and 77 percent were male. Similarly, the comparison group was 42 years old and 79 percent were male. The racial demographic breakdown was 41 percent white and 59 percent nonwhite for the treatment group, and 60 percent white and 40 percent nonwhite for the comparison group, which was a statistically significant difference. The treatment group served an average of 907 days in prison and had an average of 1.8 prior convictions, while the comparison group served an average of 1,290 days in prison and had an average of 1.6 previous incarcerations. Two statistically significant differences between the groups indicated program participants may have been at a higher risk for recidivism than comparison individuals: the treatment group had a higher proportion of members with drug or alcohol abuse and a higher average security level in prison.
Recidivism data was drawn from the Ohio Department of Rehabilitation and Correction (ODRC). Each individual was tracked for 1 year after release. Logistic regression was used to examine differences between the treatment and comparison groups in terms of the proportion of each group that was rearrested and re-incarcerated. A Cox proportional hazards model was used to analyze the time until first rearrest, and a zero-inflated negative binomial regression was used to estimate differences in number of rearrests. In each model, the study authors controlled for basic demographic variables, including mental health, disability, homelessness, incarceration, security, risk, and post-release supervision, which included controls for the three variables noted above that differed between groups. Program referral was initially slow, so recruitment into the comparison group occurred more slowly than recruitment into the treatment group. Because this led to a significantly longer average time in the community for the treatment group, the follow-up period was set at 1 year for every individual. There was no subgroup analysis conducted in the study.
The study authors noted that because the nine treatment providers involved in the program focused on different target populations and used different housing models, it was difficult to identify the specific aspects of the program or characteristics of program participants associated with positive or negative recidivism outcomes. In addition, because the individual providers had the discretion to accept or reject eligible participants, the study authors also noted the potential selection bias: providers may have cherry-picked participants whom they felt would do well in the program.
Fontaine and colleagues (2012) conducted process and cost evaluations using data collected from semistructured interviews with program staff and stakeholders and the program’s administrative databases. Overall, the cost–benefit analysis found that the Returning Home-Ohio (RHO) pilot program was not cost beneficial. Participation in the program increased costs by more than $9,500 per individual.
Evidence-Base (Studies Reviewed)
These sources were used in the development of the program profile:Study 1
Fontaine, Jocelyn, Douglas Gilchrist-Scott, John Roman, Samuel Taxy, and Caterina Roman 2012. Supportive Housing for Returning Prisoners: Outcomes and Impacts of the Returning Home – Ohio Pilot Project.
Washington, D.C.: The Urban Institute.https://shnny.org/uploads/RHO-Urban-Institute-2012.pdf
These sources were used in the development of the program profile:
Fontaine, Jocelyn, C.A. Nadeau, Caterina Roman, and John Roman. 2009. Evaluation of the Ohio Department of Rehabilitation and Correction and Corporation for Supportive Housing’s Pilot Program: Interim Report
. Washington, D.C.: The Urban Institute.
Markman, J.A., Jocelyn Fontaine, John Roman, and C.A. Nadeau. 2010. Evaluation of the Ohio Department of Rehabilitation and Correction and Corporation for Supportive Housing’s Pilot Program: Interim Re-Arrest Analysis
. Washington, D.C.: The Urban Institute.