Bridging Disciplinary Boundaries (January 11 - 14, 2007)

An estimated one in eight youth have run from home at least one time in adolescence (Whitbeck & Simons, 1990). However common it may be, running away puts youth at serious risk of victimization, sexual exploitation, substance abuse, offending, and presence in places where criminal activity occurs (Biehal & Wade, 1999; U.S. Department of Justice, 2002). Youth in substitute care are particularly vulnerable, as they have already experienced a disruption in living arrangements stemming, in the majority of cases, from documented abuse or neglect. Nonetheless, data that help elucidate which personal, family, and placement factors predict running from substitute care are sparse (Angenant, et al., 1991; Biehal & Wade, 2000; Courtney & Barth, 1996; Miller, et al., 1990; Nesmith, 2002).

This presentation describes runaway events of at least 24 hours in duration (n = 14,282) among youth in out-of-home care in a Midwestern state between 1993 and 2003. We analyze trends in the likelihood of runaway over time and present the findings of multivariate analyses (Cox proportional hazards models and repeated events hazard models) of the predictors of runaway behavior among youth in out-of-home care over this period. Our data come from state management information systems containing information on child maltreatment investigations, placement histories of children in out of home care, and Medicaid Paid Claims, which describe mental health and substance abuse diagnoses of study subjects. Data from these three data systems were linked to provide a longitudinal perspective on the experiences of the study subjects before, during, and after their time in out-of-home care.

The number of runs per care year among children in out-of-home care over the study period increased from a low of .18 in 1994 to .51 in 2003, indicating an increase in the risk of runaway over time. Factors associated with the likelihood that youth would run included: age, race, gender, mental illness, substance abuse, and developmental disability; placement instability; type of out-of-home placement; and the presence of a youth's sibling in the home. Some findings suggest potential strategies for preventing runaway. The focus of runaway prevention efforts should clearly be directed at older youth; over 90% of runaways are 12 or older and most youth who run away entered care as adolescents. Better assessment and treatment of mental health and substance abuse among foster youth may reduce the risk of runaway. Efforts to reduce placement instability could also prevent runs. Youth placed in kinship care and with their siblings were less likely to run, suggesting the importance of maintaining family ties as a prevention strategy. Lastly, the timing of runaway events is informative. Among youth in out-of-home care for the first time, the probability of runaway is very low throughout their time in care. In contrast, the probability that a youth who has run once will run again is relatively high (e.g., over 20% run again with 30 days). Prevention efforts should not ignore youth in care for the first time, but the population that has already run is an important target for intervention.