Effect of an integrated housing intervention for people involved in the criminal-legal system who have housing instability

This study demonstrates that an integrated housing, medical, and behavioral health intervention for individuals involved in the criminal-legal system significantly improved housing stability, particularly for older participants and those initially unstably housed, although high-severity substance use remained a barrier to success.

The Gist

Imagine trying to build a sturdy house on a foundation that is constantly shaking, crumbling, or even missing entirely. That is the reality for many people who have been through the criminal-legal system (prison, jail, or probation). They face a perfect storm of challenges: they often struggle with addiction, mental health issues, and a lack of money, all while society puts up "No Vacancy" signs because of their past records.

This paper describes a project called Project CHANGE, which tried to fix this broken foundation by offering a "all-in-one" solution. Instead of making people jump through hoops to get housing, then separately find a doctor, then separately find a therapist, the project brought all these services together under one roof.

Here is the story of what they did and what they found, explained simply:

The Problem: A Maze with No Exit

Think of the current system like a giant, confusing maze. If you just got out of prison and are homeless, you have to find a shelter, then apply for a job, then find a doctor, then find a therapist, and then try to get an apartment. But every door is locked with a different key (like a clean drug test for the apartment, or a referral slip for the doctor). Most people get stuck in the maze.

The researchers knew that for people with criminal records, addiction, and mental health struggles, this maze is nearly impossible to navigate alone.

The Solution: The "Swiss Army Knife" Approach

Project CHANGE decided to stop making people navigate the maze alone. Instead, they created a mobile, integrated team that acted like a Swiss Army knife for their lives.

• The Team: They had doctors, housing experts, job coaches, and mental health counselors all working together in one place (a mobile clinic).
• The Strategy: They didn't say, "Get sober first, then we'll give you a house." They said, "Here is a house, here is a doctor, and here is help with your addiction, all at the same time."
• The Intensity: They didn't just check in once a year. They were like a dedicated support squad, meeting with participants an average of 10 times over 6 to 12 months. They helped with everything from filling out paperwork to finding a job to managing medication.

The Experiment: Who Did They Help?

Between 2019 and 2023, they enrolled 187 people in New Haven, Connecticut. These were folks who had been in prison or jail, were struggling with housing (many were literally sleeping on the street), and dealing with substance use or mental health issues.

The Starting Line:

• 43% were living in unstable situations (couch-surfing, unsafe housing).
• 37% were literally homeless (sleeping outside or in shelters).
• Everyone was dealing with significant health and social hurdles.

The Finish Line: What Happened?

After following these individuals for up to a year, the results were a mix of hope and reality:

1. The Good News: About 26% of the people actually moved up the ladder. They went from being homeless to having a stable home, or from a shelter to a permanent apartment.
2. The "No Worsening" News: Another 65% didn't get better, but they didn't get worse either. They stayed in their current situation. In a world where things often spiral downward for this group, simply staying put is a victory.
3. The Overall Shift: The percentage of people with a stable home went from 43% at the start to 58% at the end.

The Secret Sauce: Who Succeeded?

The researchers looked at the data to see what made the difference. It was like looking at a recipe to see which ingredient made the cake rise.

• Time is the Secret Ingredient: The longer someone stayed in the program, the more likely they were to get a home. Every extra month in the program increased their odds of success. It takes time to rebuild a life.
• Age Matters: Older participants had an easier time finding housing. Perhaps they had more life experience or fewer chaotic variables in their lives.
• The "Starting Point" Factor: If someone already had some kind of housing (even if it was unstable) when they started, they were 7 times more likely to end up with a stable home. It's easier to fix a shaky roof than to build a house from scratch on an empty lot.
• The Big Hurdle: People with severe substance use struggled the most. They were much less likely to get housed. This isn't because the program failed them, but because addiction is a massive, heavy anchor that makes moving forward incredibly difficult.

The Takeaway: It's Hard Work, But It Works

The authors admit that this program was expensive and exhausting to run. It required a massive amount of human energy (dozens of meetings per person). It wasn't a quick fix; it was a marathon.

However, the study proves that when you stop treating housing, health, and justice as separate problems and start treating them as one big, tangled knot, you can untangle it.

The Bottom Line:
You can't expect someone to climb a mountain if you only give them a map but no boots, no water, and no guide. Project CHANGE gave them the boots, the water, and the guide. While not everyone reached the summit, many more made it up the mountain than would have without that support. The paper argues that we need more of these "all-in-one" teams, especially for the people who are fighting the hardest battles against addiction and the justice system.

Technical Summary

1. Problem Statement

Individuals involved in the criminal-legal system face disproportionate challenges in securing and maintaining permanent housing compared to the general population. They are ten times more likely to experience homelessness and twice as likely to move homes frequently. These challenges are compounded by:

• Co-occurring conditions: High rates of substance use, psychiatric disorders, and social isolation.
• Structural barriers: Federal and local policies (e.g., restrictions on Section 8 vouchers for those with drug or sex offenses) and strained personal relationships.
• Service fragmentation: Medical care, housing assistance, and behavioral health services are traditionally delivered in "silos," creating gaps for individuals with complex, intersecting needs.
• Negative outcomes: Housing instability is linked to reincarceration, reduced engagement in medical care, and compounded trauma.

2. Methodology

Study Design & Setting:

• Intervention: Project CHANGE (Comprehensive Housing and Addiction Treatment Network of Greater New Haven), an integrated intervention co-locating medical care, housing assistance, behavioral health, and addiction treatment.
• Location: New Haven County, Connecticut, an area with high rates of overdose, HIV, incarceration, and poverty.
• Participants: 187 individuals enrolled between June 2019 and November 2023.
  • Inclusion Criteria: Age ≥18, living in New Haven, justice-involved (released, on supervision, or at risk), experiencing homelessness/housing insecurity, and having co-occurring substance use or psychiatric disorders.
  • Exclusion: 4 participants were removed due to inappropriate conduct.

Intervention Delivery:

• Team: Interdisciplinary team including clinicians, housing/vocational case managers, and syringe services program (SSP) staff.
• Approach: Person-centered, bilingual, trauma-informed, low-barrier care.
• Services: Intensive Housing Case Managers (baseline assessment, shelter help, referrals) and Vocational Case Managers (education/employment resources).
• Follow-up: Participants were followed for 12 months (6 months for those enrolled after Jan 2023 due to budget constraints) with quarterly/bi-annual interviews and check-ins.

Data Collection & Measures:

• Primary Outcome: Housing status at final follow-up (categorized as: Housed, Sheltered/Institutional, or Homeless) based on the Government Performance and Results Act (GPRA) Section C.
• Covariates:
  • Demographics: Age, sex, race, education, income.
  • Health: Addiction Severity Index (ASI), NMASSIST (substance use severity), CES-D/PHQ-9 (depression), PTSD Symptom Scale.
  • Social: MOS Social Support Survey, quality of life (SF-12).
• Statistical Analysis:
  • Descriptive statistics for baseline characteristics.
  • Bivariate analyses (chi-square, t-tests) for changes over time.
  • Multivariate Logistic Regression: To identify correlates of being "housed" at the final follow-up. Models controlled for baseline housing status, age, substance use severity, time enrolled (person-months), and social support.

3. Key Contributions

• Integrated Model Validation: Demonstrates the efficacy of a "Housing First" approach that integrates medical, behavioral, and housing services without requiring sobriety as a precondition.
• Specific Population Focus: Addresses the unique intersection of criminal-legal involvement, housing instability, and high-severity substance use, a group often excluded from standard housing programs.
• Resource Quantification: Provides empirical data on the intensity of care required, tracking the number of service encounters (average 10 per participant) necessary to achieve outcomes.

4. Key Results

Baseline Characteristics:

• Mean age: 44.5 years; 59% Male, 58% White, 37% Black.
• Housing Status: 37% literally homeless, 19% sheltered/institutionalized, 43% housed but insecure.
• Health: 62% had severe substance use; 69.5% screened positive for clinical depression.

Housing Outcomes:

• Improvement: 49 participants (26.2%) improved their housing status (e.g., homeless to housed).
• Stability: 121 participants (64.7%) maintained their housing status (did not worsen).
• Overall Shift: The proportion of participants housed in the past 30 days increased from 43% at baseline to 58% at final follow-up.

Predictors of Being Housed (Multivariate Analysis):

• Positive Predictors:
  • Time Enrolled: Each additional month in the program increased odds of being housed by 10% (AOR 1.1; 95% CI 1.1–1.2).
  • Age: Each additional year of age increased odds by 10% (AOR 1.1; 95% CI 1.0–1.1).
  • Baseline Housing: Participants who were housed (even if unstable) at baseline had 7.2 times higher odds of being housed at follow-up compared to those who were homeless (AOR 7.2; 95% CI 3.3–16.0).
• Negative Predictors:
  • Substance Use Severity: High-risk substance use at baseline was associated with a 70% reduction in the odds of being housed (AOR 0.3; 95% CI 0.1–0.6).

5. Significance and Implications

• Policy Relevance: The study supports "Housing First" models over "Treatment First" linear models, which often fail individuals with complex psychiatric and substance use disorders. It highlights the need for concurrent delivery of medical, housing, and case management services.
• Resource Intensity: The program was time- and cost-intensive, requiring an average of 10 service encounters per participant over 6–12 months. This underscores the high resource demand needed to sustain success for this high-need population.
• Equity and Barriers: The findings reveal that while integrated care works, high-severity substance use remains a significant barrier to housing stability. The authors note that recent federal policy shifts threatening funding for harm reduction and Housing First programs could undermine these gains.
• Limitations: The study was conducted in a high-resource setting (New England) with Medicaid expansion, potentially limiting generalizability to regions with fewer safety nets. Some attrition occurred, though methods were used to minimize bias.

Conclusion:
Project CHANGE demonstrates that integrated, intensive housing and health interventions can successfully improve housing outcomes for justice-involved individuals. However, success is heavily dependent on the duration of engagement and is significantly hindered by high-severity substance use, necessitating tailored, long-term support systems.