Trust Youth Evaluation

The Trust Youth program involves a “cash plus” model, providing unconditional direct cash transfers (“cash”) along with connections to optional complementary supports and services (“plus”) over a 24-month period. Both the direct cash transfers and supportive programming were co-designed with young people with lived experience of homelessness and for the purpose of helping young people achieve safe and stable housing and get on paths to thriving.

2022-04-04

2024-04-12

Homelessness
Number of days of safe and stable housing

HOUSING SITUATION – Survey items on participants' housing situation will capture information on a young person’s housing situation and current or recent homelessness or housing instability. The first five items capture information on the number of moves as a measure of housing instability and confidence in maintaining safe and stable housing over the coming month. These items were created by the Chapin Hall research team. The next set of items (24 total) capture longitudinal information on the respondent’s recent (last 14 nights) housing/homelessness based on sleeping arrangements; these items comprise the NYC Pathways Study/Trust Youth Evaluation’s primary outcome measure of degree of homelessness and housing instability. The Chapin Hall team constructed these items by drawing on extant literature and other measures on housing and homelessness situations (Morton et al. 2019; Tsemberis et al., 2007). Specifically, these items mimic the information that can be drawn from a Residential Timeline Follow-Back Inventory (RTFBI; Tsemberis et al., 2007) through a self-administered survey. Three items tap into housing security and include items from the US Census Bureau’s Household Pulse Survey Phase 3 Questionnaire. The items capture information that can be used for calculating overcrowding. These items were based on questions from the American Community Survey and measurement guidance from HUD (Blake et al., 2007). Four items capture information on the safety (from Redline et al., 2021) and quality of the respondent’s current housing. Two items measure spatial mobility over the course of the evaluation. Except for the safety item, these items were created by the Chapin Hall team.

Additionally, through data sharing agreements (DSAs) with the NYC Department of Homelessness Services and the Department of Youth and Community Development, we will also receive data on young people's use of shelters/crisis emergency services. This will allow for analysis of number of days using these services as another measure of reductions in homelessness (shelter stays are considered a form of homelessness by the US Department of Housing and Urban Development).

Education/training enrollment and participation

Employment status and earnings

Sleep amount and quality

Food insecurity

Use of other public assistance benefits and services

Debt

Access to financial services

Mental health

Self-efficacy/mastery/empowerment

Loneliness/social isolation

Social supports

Personal time use

Independent living preparedness

Flourishing

Thriving against one's personal goals

General health

Substance use

EDUCATION & EMPLOYMENT (5 -15 items) – Items measure the education, employment, and earnings draw or adapt from the Santa Clara Basic Income for Former Foster Youth evaluation baseline survey and the US Census Bureau’s Household Pulse Survey Phase 2 (2020). A global single-item measure of job satisfaction is also included (Judge et al., 2020).

SLEEP (5 items) – Items assess sleep quality from the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) (Cella et al., 2019) The PROMIS 4-item Short Form was used by Redline et al. (2021) in a recent study of youth experiencing homelessness.

FOOD INSECURITY (6 items) - This section includes items that focus on food hardships from the U.S. Household Food Security Survey Module short form (Blumberg et al., 1999)

BENEFITS, PROGRAMS, & SERVICES (14 - 22 items) - The section on benefits, programs, and services was created by the Chapin Hall research team for use in this study with input from NYC Government partners. The section on access to essential documents was adapted from the Youth Villages Transitional Living, 12-Month Survey.

FINANCES (6 - 13 items) – This section includes information on income adapted from the US Department of Housing & Urban Development (HUD) HMIS data standards. Savings items were adapted from the Youth Villages Transitional Living, 12-Month Survey. Debt items were adapted from the Santa Clara County and Youth Villages survey instruments. Financial inclusion items (e.g., having access to a financial account and services) came from the World Bank's FINDEX survey.

SOCIAL & EMOTIONAL WELL-BEING (20 items) – This section includes the Patient Health Questionnaire-4 (PHQ-4), the Pearlin Mastery Scale, and the Flourishing Scale. All are widely used and validated scales. Depression and anxiety items we assessed with the PHQ-4. The depression items are the briefest validated mental health screening scale for depressive disorder and have shown sensitivity to change (Kroenke et al., 2003; Löwe et al., 2005; Staples et al., 2019). The anxiety items are the briefest validated mental health screening scale for General Anxiety Disorder and are sensitive to change over time (Kroenke et al., 2007; Staples et al., 2019). Then one item was included from the NIH Toolbox 6-item loneliness measure on the perception that one is alone, lonely, or socially isolated from others. The Pearlin Mastery Scale is a 7-item scale that measures an individual’s level of mastery, which is a psychological resource that has been defined as “the extent to which one regards one’s life-chances as being under one’s own control in contrast to being fatalistically ruled” (Pearlin & Schooler, 1978, p.5). The Flourishing Scale is a 8-item scale describing positive relationships, to feelings of competence, to having meaning and purpose in life (Diener et al., 2010).

HEALTH AND SUBSTANCE USE (11 items) – This section includes an item on general health from the US Census Bureau’s Household Pulse Survey and the World Health Organization’s widely used and validated The ASSIST was designed to be culturally neutral and useable across a variety of cultures to screen for use of the following substances: tobacco products, alcohol, cannabis, cocaine, amphetamine-type stimulants (ATS), sedatives and sleeping pills (benzodiazepines), hallucinogens, opioids, and ‘other’ drugs (Humeniuk et al., 2008).
12. SOCIAL SUPPORTS (8 items) – NIH Toolbox Emotional Support 8-items that measures the perception that people in one’s social network are available to listen to one’s problems with empathy, caring, and understanding (Salsman et al., 2013). 13. PERSONAL TIME USE (7 items) - The personal time use items were adapted by the Chapin Hall research team from a similar module in the Santa Clara County Basic Income for Former Foster Youth Baseline Survey Instrument.

INDEPENDENT LIVING PREPAREDNESS (2-3 items)- This section includes two items capturing young people’s perceptions of their own level of independent living preparedness and the independent living skills they believe they need. These items come from the Point Source Youth 2019 Baseline Survey for its Rapid Rehousing Program evaluation.

Young adults experiencing homelessness in NYC will be recruited into the evaluation through homelessness services and randomly selected to either continue receiving services-as-usual while participating in the longitudinal data collection over 2.5 years OR to also receive an invitation to participate in the Trust Youth cash plus program while participating in the longitudinal data collection over 2.5 years. As such, the experimental evaluation aims to estimate the effects of the Trust Youth program over and above young people's access to services-as-usual.

Services-as-usual include shelter, counseling, case management, life skills program, and connection to other housing resources (e.g., housing vouchers or supportive housing) and a range of other services that young people routinely have access to through the drop-in center and shelter programs from which they will be recruited and from NYC’s service array for young people experiencing homelessness more broadly.

All young people participating in the study, irrespective of random assignment, will receive cash stipends for participating in data collection activities over 2.5 years ($50 for 45-60 minute surveys administered every 6-12 months and $20 for 10-15 minute surveys administered monthly in-between the 45-60 minute survey months).

Not available

Mathematica, recruitment, data collection, and random assignment contractor to Chapin Hall for this study, will conduct random assignment within approximately one week of respondents having completed the baseline survey to determine the subset of study participants who will be randomly invited to participate in the Trust Youth “cash plus” program. To reach the predetermined group sizes of 30 intervention group and 40 control group participants, the random assignment module will be programmed to use a biased coin design approach. The survey and random assignment module will be programmed in Confirmit survey software. The study participant will not experience or be aware of the random assignment procedure while it occurs, and only those randomly assigned to Trust Youth will be notified that this random assignment process occurred to identify invitees to participate in Trust Youth in addition to participation in the longitudinal data collection and services-as-usual.

Individual (young adult)

No

0

70 (initially for the pilot stage; subject to funding, we anticipate a second round of recruitment to begin approximately 6 months later for an additional 60 participants for a total sample size of 130. We will update the registration information accordingly).

30 in the Trust Youth cash plus program; 40 in the services-as-usual comparison

The activities to evaluate the Trust Youth program described in this protocol are part of a multistage research, evaluation, and program development process, consistent with the Medical Research Council (MRC) Framework for Developing and Evaluating Complex Interventions (Craig et al., 2008). Consistent with the MRC Framework, we take a developmental approach to evaluation. In this project stage, the project focuses on pilot testing the program, evaluating its processes and implementation to inform potential program improvements, and gaining better information on the direction and magnitude of effects to inform future sample size calculations for a more fully powered evaluation in the future. In alignment with the MRC Framework, this initial trial will help to pilot andd refine sampling and randomization procedures in advance of a second round of recruitment, explore the direction and magnitude of likely program effects, and help test/inform sample size calculations and outcome measures for a better-powered impact study before the second round of recruitment begins to achieve a fully powered study. While this preliminary randomized trial of Trust Youth is intentionally not fully powered to estimate statistically significant effects of modest sizes or by subgroups, this sample size is sufficient to estimate overall treatment effects if we find a similar effect size (34% mean difference) for housing stability that was observed from a randomized trial of a “Housing First” rental subsidy and wraparound services intervention for young adults with mental illness in Canada (Kozloff et al., 2016). This trial is the closest comparison in the literature available from which to draw an effect size assumption and used the same primary outcome measure of housing stability that we plan to use. We assume 95% confidence intervals, two-sided tests, 80% power, 20% attrition, and SD = 0.43. Using the same assumptions, a sample size of 130 (60 in Trust Youth [TY] program and 70 in services-as-usual [SAU]) would be sufficient to demonstrate statistically significant results with a 19% mean difference. A larger sample size also provides a better buffer in the event of higher levels of evaluation dropout, missing data, or unexpected levels of variation between outcomes. REFERENCES Craig, P., Dieppe, P., Macintyre, S., Michie, S., Nazareth, I., & Petticrew, M. (2008). Developing and evaluating complex interventions: the new Medical Research Council guidance. BMJ, 337. Kozloff, N., Adair, C. E., Lazgare, L. I. P., Poremski, D., Cheung, A. H., Sandu, R., & Stergiopoulos, V. (2016). " Housing first" for homeless youth with mental illness. Pediatrics, 138(4), e20161514.