Evaluation of the effect of Arizona's crisis response system on hospitalizations for behavioral health conditions

Abstract

Background: US federal and state governments are investing in comprehensive behavioral health crisis system reforms. These reforms aim to prevent suicide, limit police involvement in crises, and reduce reliance on hospitals and emergency departments. Arizona's reform is considered a model for other states. Aims: We evaluated whether Arizona's crisis system reform was associated with reductions in behavioral health-related hospitalizations. Method: We used a comparative interrupted time series design to assess Arizona's implementation of their crisis response system in 2015. We used 2011-2018 Healthcare Cost and Utilization Project (HCUP) State Inpatient Dataset (SID) data and selected Colorado, Kentucky, Michigan, New Jersey, New Mexico, and Washington as comparison states. Results: Arizona's annual behavioral health-related hospital discharge rate per 100,000 population decreased from 686.3 in 2011 to 673.7 in 2014 and increased from 759.3 in 2016 to 955.7 in 2019. The comparative interrupted time series analyses revealed that implementation of Arizona's crisis system was not associated with a change in the rate of behavioral health hospitalizations. Limitations: There may be some unmeasured, time-varying factors related to the rate of behavioral health-related hospitalizations between Arizona and our comparison states that we are not accounting for. Also, hospitals switched from using ICD-9 to ICD-10 codes in 2015, the same year as Arizona implemented their crisis system. Conclusions: More research is needed to confirm whether and how comprehensive crisis response systems impact behavioral health-related hospitalizations.