Identifying housing instability using administrative data: Implications for research and advocacy

OBJECTIVES: Housing instability is an important social driver of health for children; however, it remains difficult to identify. This study explored whether an administrative database could be used to identify children with housing instability presenting to care at pediatric hospitals and how rates of identification have changed over time. 

METHODS: Using the Pediatric Health Information System database, this study explored identification of housing instability in children through the use of International Classification of Diseases, Ninth Revision (ICD-9) and International Classification of Diseases, Tenth Revision (ICD-10) Z codes at tertiary children's hospitals. The primary outcome was incidence of identified housing instability, either as a primary or nonprimary diagnosis, during the study period. 

RESULTS: Over a span of 20 years, there were 22 828 encounters with housing instability, 15% (3491) of which had a primary diagnosis for housing instability. The proportion of encounters identified with housing instability increased during the study period (P < .001). Non-Hispanic Black (34.5%), Hispanic (27.8%), and government-insured patients (81%) were disproportionally represented among encounters with housing instability. Additionally, there was significant regional variation in the identification of housing instability (P < .001). Nearly all encounters (98%) with a primary diagnosis of housing instability were discharged from the emergency department. 

CONCLUSIONS: These findings suggest that administrative databases may present an emerging opportunity to identify housing instability among pediatric patients. Although Z-code identification of housing instability varies widely by region, their use has increased over time, and continuing to study housing instability trends through administrative data can inform future research and advocacy.