Recent empirical studies show that participants in runaway/homeless youth crisis shelters report strong positive outcomes. These studies are limited in that they examine outcomes for only short periods. While this evidence is sensible given the crisis nature of the services, it does not address how these outcomes maintain or disappear over time-knowledge critical to the development of interventions. The purpose of this study is to examine longitudinal outcomes for youth participating in runaway/homeless crisis shelters. Data were collected for consecutive intakes (n=261) beginning in October 1997 at 11 youth crisis shelters. Interviews were conducted at intake, 6-weeks, 3-months, and 6-months post-discharge. Variables included days-on-the-run, family contact and support, negative school events (suspension, expulsion, detention), employment (currently employed, recently fired), current sexual activity, current substance use, and self-esteem. To minimize potential bias, an expectation-maximization algorithm was used to impute missing data. Multiple imputations of missing data were performed and five implicative data sets created, identical analyses were conducted for each, and results regressed together to estimate standard error. Multiple t-tests compared outcomes with intake and previous responses for each time-period. Results showed that significant positive changes had occurred for 10 of the 11 outcome variables at 6-weeks post-intervention. By 6-months, however, only days-on-the-run and current substance use had maintained significant levels of improvement. Crisis shelters have long-term positive effects on youth, particularly on runaway behaviors-the outcomes these programs were designed to address. To maintain outcomes achieved short-term, coordination with other health, educational, and vocational services appears to be required.
Learning Objectives: At the conclusion of the session, the participant will be able to: 1. Describe longitudinal outcomes for youth participating in runaway/homeless crisis shelters. 2. Articulate the role of runaway/homeless crisis shelters as part of the larger continuum of services to youth. 3. Discuss the potential for using an expectation-maximization algorithm to impute missing data for service evaluation
Keywords: Runaways, Homeless
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I have a significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.
Relationship: The research was funded by a grant from Administration for Children and Families, DHHS to Dr. Pollio