This study tests a typology of family homelessness based on patterns of public shelter utilization and examines whether family characteristics are associated with those patterns.

The study highlighted in the article examined patterns of shelter use among families in four geographic areas: Philadelphia, New York City, Columbus, and Massachusetts. The researchers found three distinct patterns of shelter use. The majority of families have a "transitional" homeless episode, remaining in shelter for a relatively brief period of time (one to five months) before exiting and not returning to shelter. Approximately 20 percent are long term stayers in shelter or transitional housing (ranging from 144 days in Columbus to 552 in New York City). Finally, a very small number of families have "episodic" homeless episodes, with repeated entries and exits into the homeless shelter system during the study period.

The researchers examined families' interaction with intensive social service programs (including inpatient mental health, substance abuse treatment, and foster care placement) to determine whether differences existed between the three groups. Approximately a quarter of all families in the shelter system had intensive social service program use, but the rate was highest among those in the episodic group and lowest among long term stayers. In Massachusetts, for example, a third of families in the episodic group had at least one intensive social service use compared to 17 percent of those in the long term stayers group and 29 percent of those in the transitional homeless group. The authors explore the utilization of resources to support families in the three subgroups and propose shifts in how resources are allocated.

The results suggest that policy and program factors, rather than family characteristics, are responsible for long shelter stays. An alternative conceptual framework for providing emergency assistance to homeless families is discussed.