Experimental Design
Eligibility for Referral
The study will maintain the program’s existing eligibility, namely clients must (1) be homeless, (2) be a single adult, and (3) score within the RRH range on the VI-SPDAT.
1) The criterion of being “homeless” is met if the client falls within the definition of HUD category 1 homelessness or HUD category 4 homelessness.
a. HUD category 1 homelessness includes: Individual or family who lacks a fixed, regular, and adequate nighttime residence, meaning: (i) Has a primary nighttime residence that is a public or private place not meant for human habitation; (ii) Is living in a publicly or privately operated shelter designated to provide temporary living arrangements (including congregate shelters, transitional housing, and hotels and motels paid for by charitable organizations or by federal, state and local government programs); or (iii) Is exiting an institution where (s)he has resided for 90 days or less and who resided in an emergency shelter or place not meant for human habitation immediately before entering that institution.
b. HUD category 4 homelessness includes: Any individual or family who: (i) Is fleeing, or is attempting to flee, domestic violence; (ii) Has no other residence; and (iii) Lacks the resources or support networks to obtain other permanent housing
2) The criterion of being a single adult is met if the client lives in a household without children.
3) The criterion of being within the RRH range is met if a client scores between 4-7 on their most recent VI-SPDAT.
Program Referral
The existing enrollment process consists of the County’s Office of Supportive Housing (OSH) identifying potentially eligible clients within the existing HMIS registry and referring them to housing providers like HomeFirst. Historically, OSH has imposed the same eligibility criteria as listed above when referring single adults to RRH programs. However, OSH prioritized individuals in the RRH range, referring first those who scored a 7, then those who scored a 6, and so on and so forth. Furthermore, they prioritized within VI-SPDAT score, referring first those with the highest “risk score” within the pool of those who scored a 7. Moreover, as an ultimate tiebreaker, OSH referred those with the most days homeless among clients with otherwise equal scores.
Because this study seeks to identify for whom RRH is most effective, the research team requests that the enrollment process be changed in two important ways.
1) Instead of referring one client with the highest VI-SPDAT score as a new slot becomes available, OSH will refer a list of clients across the eligible VI-SPDAT range, referring an equal number of clients who scored a 7, 6, 5, and 4 (prioritized within each score as OSH sees fit).
2) When the list of clients is depleted and program slots begin to turnover, referrals will cycle through 7’s, 6’s, 5’s, and 4’s, maintain an approximately equal number of clients from each score range in the program at all times.
These changes allow for homeless individuals with scores across the entire RRH range to receive services. Ultimately, this new referral process will allow the research team to not only analyze the impact of RRH, but also examine for which VI-SPDAT score RRH was most effective.
Intake and Randomization
Given HomeFirst’s static capacity of 60 clients, the first list of referrals should contain at least 170 homeless individuals (42-43 from each eligible score) with the goal of enrolling 15 from each score into the intervention group and approximately 15 from each score into the comparison group.
When HomeFirst receives this list of referrals from OSH, potential clients should be divided into equal group for case managers to contact. When attempting to contact clients, case managers need to make three reasonable attempts (according to OSH standard of reasonable attempt) to connect with a potential client before removing them from the list. Each case manager should have a variety of clients (i.e. each case manager should be in charge of contacting, enrolling, and ultimately serving approximately five 7’s, five, 6’s, five 5’s, and five 4’s). PROVIDER IS FREE TO CONTACT CLIENTS IN WHATEVER IS MOST EFFECTIVE. If case managers were charged with serving only one score group it would be difficult to parse out the effect of the program on a particular score group from the effect of a particular case manager. The goal of this study is to evaluate the impact of the program on individuals across the score range, not the impact of a case manager on a score group.
When the service provider contacts a potential participant, they will proceed through the intake process on an Android tablet loaded with the SurveyCTO app to accomplish four goals
1) Confirm Eligibility: determine if the client is still eligible by documenting major life changes since the time of the VI-SPDAT scoring date. This step is included to confirm that the client is still homeless and still a part of a household without children, SPECIFICALLY NO CUSTODIAL RIGHTS and, therefore, still eligible. If the client is ineligible, they will not be consented or randomized.
2) Consent: explain the nature of the study and consent the client on the tablet. Explain that if a client consents they are not guaranteed a program slot. If clients do not consent to be in the study, they will be referred to other programs and informed that they can call back if they change their mind. Because the funds are being used for participants in the study, clients who do not consent will not be offered RRH slots dedicated to this study. If a client does not consent, no additional information will be collected.
3) Record information: prior to proceed with the take questions, the case worker will enter the client’s HMIS ID and verify the client’s social security number, name and spelling), and date of birth in HMIS on another window on the tablet. The VI-SPDAT will not be re-administered, though a short intake form will be completed by all study participants to collect baseline information such as:
a. Housing status at entry
b. Zip Code of last permanent address
c. Length of stay on prior living situation
d. Night before measure of homelessness
e. Date homelessness started
f. In the last three years, how many times have you been homeless?
g. Income from any source
h. Non-cash benefit sources
i. Health insurance coverage
j. Primary language
4) Randomization: if clients consent to participate in the study, proceed to randomization page of tablet and inform the client on whether they are in the control group and will have access to usual care or in the treatment group and will receive RRH through this study. If the client is in the treatment group the case manager will proceed forward with the program intake form.
5) REFERRAL/ENROLMENT
This process must be completed every time a new slot becomes available. Clients in the control group will be eligible for any other usual service provided through the County outside of this study. When new program slots within the study become available, the control group will not be contacted again. Individuals can only be randomized once. At the close of the study, all study participants will be eligible for any program they would normally be eligible for in absence of the study.