240 likes | 413 Views
Rapid Re-Housing Research Evidence and Beyond. Jamie Taylor Cloudburst Consulting Group Connecticut Coalition to End Homelessness Training Institute May 8, 2014. 1) RRH - Results from across the nation 2 ) RRH - The Philadelphia Story 3 ) RRH – Promising Practices
E N D
Rapid Re-Housing Research Evidence and Beyond Jamie Taylor Cloudburst Consulting Group Connecticut Coalition to End Homelessness Training Institute May 8, 2014
1) RRH - Results from across the nation 2) RRH - The Philadelphia Story 3) RRH – Promising Practices 4) RRH – Local RRH Evaluation Objectives – Rapid Re-Housing Overview
Did RRH help decrease risk of homelessness in CT 2010–2013? Source: HUD CoC Reports
NAEH Evaluation of 7 CoCPrograms, Average RRH cost = $4,000/family
Was not in a Rapid Re-Housing program Had a history of homelessness Went to a “temporary” destination Was Non-Hispanic / Non-Latino Was Non-White Had a disabling condition at program exit Program was in a non-rural county Was male Was unaccompanied Was not with a teenage male Source: Jason Rodriguez, GA Dept of Community Affairs Georgia Study of Reoccurrence Rates – Rigorous method to control for differences, found factors most correlated with a return to homelessness Research question: Which client, program, geographical characteristics exert greatest influence on the likelihood that someone returns to homelessness? Results: 9000 enrollments, 28% return to homelessness. Return Risk Factors: Key Finding: Exits from Shelter 4.7 times; Tran. Housing 4.0 times more likely to return to homelessness than exits from Rapid Re-Housing
RESEARCH AIM for RRH Policy: Research for RRH policy goal is to estimate whether RRH is the specific element responsible for decreasing homelessness. Counterfactual: What would have happened to RRH households if there was no RRH? WHY RESEARCH DESIGN IS NECESSARY: When households who participate in RRH are different from households who do not, need to control for differences using research design. Differences in RRH and non-RRH households show up as confounders: i.e. RRH enrollment strategies differences by case manager, by program; length of RRH assistance; Housing market variability Gold Standard = Random Control Trial = assess causal effect of RRH RESEARCH DESIGN WITHOUT RCT With no RCT, matching methods can be used to create comparison groups that look alike, controlling for confounding differences. Propensity score matching now widely applied, probability of participation estimated using observable variables., Research Aims for Rapid Re-Housing Can we answer the counterfactual?
Does Rapid Re-Housing improve housing stability for formerly homeless households by decreasing the risk of a return to homelessness? Does RRH help to improve household income? Was the HPRP RRH policy effective in decreasing the risk of homelessness? Specific Research Questions for Philadelphia Rapid Re-Housing Study
Dataset: All Households that entered Philadelphia shelters 10/2009-5/2012 Propensity Score Match
PSM Result–households in each group similar, standard means balanced RRH Treatment……….1169 households Non-RRH Control…...1286 households Each variable included in PSM represents HMIS data indicator correlated with risk of homelessness. (Disabling condition excluded based on high correlation with SSI-SSDI)Standard means comparison, t-tests performed on PSM matched groups, strong PSM model, households similar
PSM Analysis: Return to Homelessness Results • Odds ratio: The odds of returning to homelessness were 42% higher for households that did not receive RRH compared to households that did receive RRH
Washington State Evaluation – Robust matching model RRH and employment Washington State 2010 Evaluation - Rapid Re-Housing Impacts on Employment* *RRH clients were 1.25 times more likely to be employed, and, on average, earned $422 more annually than their counterparts who did not receive RRH.
RRH Promising Practice: King County RRH Pilot • Goal – To move 350 homeless families in King County into rental housing by December 31, 2014 • Assessment: Short-term financial assistance and temporary housing-focused supports, including employment and training services, • RRH funding: $3.1 million over 2014. Funders and planning partners include King County DCHS, City of Seattle Human Services Department, United Way of King County, Building Changes and the Seattle and King County Housing Authorities. • RRH partnerships: Employment Navigator program. The navigators will provide critical supports to assist in gaining employment. Families may continue working with the employment navigator after rapid re-housing assistance
RRH Promising Practice: Massachusetts Fireman Foundation Secure Jobs Pilot • Goal – Offer employment assistance to families transitioning from shelter into housing with Rapid Re-housing • Assessment: Participating agencies enrolled 506 formerly homeless parents in the Secure Jobs program from a pool of 5,400 Massachusetts families receiving rental subsidies • RRH funding: Fireman Foundation awarded $1.5 million in grants to encourage housing, employment, and other agencies to work together provide comprehensive services to help low-income families regain financial independence and stay out of the shelter system. • RRH partnerships: Collaboration with workforce-training organizations with employer partners. Secure Jobs participants employed by large retailers, hospitals and nursing facilities, hotels and hospitality industries, social service agencies, and manufacturing,
RRH Promising Practice: Tacoma Housing Authority • Goal – Serve Homeless households with children. Housing Authority launching pad for family success • Assessment: Tailor the availability, type, amount, and duration of assistance to the need for family housing • RRH funding: Use Tacoma Housing Authority Moving to Work flexible demonstration status (HUD) for RRH assistance • 2013 - $80.00 for 19 families • 2015 - $650,000 • 2017 - $1million • RRH partnerships: Schools and the child welfare system
RRH Promising Practice: Utah - The Road Home • Goal – Exit family households out of shelter to stable housing as soon as possible • Assessment: Of659 families entered Salt Lake County shelter2013 572 families movedout: 62% of all families move out with RRH 5%families moved into supportive housing, 33% of families moved out of shelter with no financialassistance • Reassessment: Progressive Engagement • RRH funding: Utah uses state TANF $$ for first four months of RRH,then ESG and other RRH funding if household still needs RRH • RRH partnerships: TANF, State Department of Workforce Services to increase employment income
Recommendations for the Hennepin County Family Shelter System 2013 Summary of Recommended Practices • Collaboration and communication are key to providing not only a positive environment for families experiencing homelessness, but also provide better outcomes for families. • Streamlining the movement for a family from the point in time in which they seek out shelter to the point that they are stably housed reduces inefficiency and better serves our community. • Using existing resources provides the largest area of opportunity to make immediate changes and see an immediate reduction in family shelter use. • Targeting services based on individualized needs of the family is a more efficient use of resources, and provides the best outcomes for families.
Maybe….RRH housing case management services access landlord partnerships, find new viable housing opportunities not previously on the radar for very poor households with housing barriers Maybe….time-limited housing stabilization assistance provides a self-determination boost, motivating efforts to do “whatever it takes” to stay out of homelessness Maybe… RRH works on the same fundamental principle as Housing First - -CLIENT CHOICE. By putting housing first in the service equation, clients access all three critical aspects of self-determination: autonomy, competence, and connectedness RRH appears to effectively decrease risk of a return to homelessness. Why?
Variable influencing factors in every RRH region: Housing market – % affordable rents Network of Landlord partnerships Capacity to leverage TANF / HOME/ other Rental Assistance Funds ESG funding levels Belief in RRH approach Coordinated Assessment Tools Mass movement out of state or HMIS region Growingneed for additional RRH research evidence AND additional investment in affordable housing. RRH does not end poverty. Multiple factors in every region impact RRH outcomes
1. Define Rapid Re-Housing Success in own community 2. Use HMIS data indicators Return to Homelessness by cohort/group Length of stay/time homeless Reduction in shelter households over time Average Shelter costs per day Average RRH assistance costs per day 3. Establish comparison group using matching method 4. Analyze Data – Courageously accept data shortcomings 5. Add results to emerging RRH evidence Local HMIS RRH Evaluation – Five Steps
Strong Performance Measurement Driver Diagram Mapping out a theory of change is key to monitoring RRH performance and continuous quality improvement. Three questions: 1) What is the aim of your RRH intervention? • What are you seeking to improve? 2) What are the necessary conditions for achieving RRH aim • What strategies will be necessary to achieve your RRH aim? • How will you know you are successful with each strategy? 3) What will it take to implement each primary strategy?
Driver Diagram for Expanded RRH Theory of Change - Change Metrics
Thank you! Jamie Taylor Cloudburst Consulting Group jamie.taylor@cloudburstgroup.com Phone # 860-716-7392