Changes in guideline-recommended medication possession after implementing Kendra's law in New York

This study examined changes in possession of guideline-recommended medication among three groups of New York State Medicaid enrollees with severe mental illness: those who received an involuntary outpatient commitment order, voluntary enhanced services, or neither of these interventions.

An observational study was conducted with New York State Medicaid claims data for enrollees with bipolar, schizophrenia, or schizoaffective disorders in New York City, Long Island, and the Hudson River regions from 2000 to 2005 (N=7,762). With adjustment for clinical and demographic characteristics, logistic regression models predicted the probability of a monthly medication possession ratio (MPR) ≥ 80% for medications recommended by expert guidelines or by the U.S. Food and Drug Administration for the indicated psychiatric diagnosis. Separate models were fit by region and for patients who ever received assisted outpatient treatment (AOT), voluntary enhanced services but never AOT, or neither treatment.

In all three regions, for all three groups, the predicted probability of an MPR ≥ 80% improved over time (AOT improved by 31-40 percentage points, followed by enhanced services, which improved by 15-22 points, and "neither treatment," improving 8-19 points). Some regional differences in MPR trajectories were observed.

After New York implemented AOT and increased community resources for enhanced services, guideline-recommended medication possession improved among Medicaid enrollees with severe mental illness--even among those who never received these interventions or services. However, further study is needed to understand why there were different regional trajectories and why some groups did not gain similarly across regions.