A New Method for Estimating Incidence of First Psychotic Diagnosis in a Medicaid Population.

Objective: Early intervention programs for first-episode psychosis (FEP) require population-based methods to identify individuals with FEP. This study adapted a previously published method to estimate incidence of first psychotic diagnosis in a state Medicaid program. Secondary aims were to examine demographic and service patterns associated with a first psychotic diagnosis in Medicaid.

Methods: A retrospective, population-based study of New York State Medicaid data was conducted to identify first occurrence of psychotic diagnosis among persons ages 15-35 between January 1, 2013, and December 31, 2017 (N=31,606). Age-stratified incidence rates (IRs) were calculated by demographic characteristics, first-diagnosis type, and service-related characteristics. Review of charts from OnTrackNY and Medicaid managed care organizations (MCOs) was conducted to confirm identified cases. Initial IRs and confirmation rates were used to estimate adjusted IRs.

Results: Age-stratified IRs varied by demographic, diagnostic, and service-related characteristic. IRs of FEP were higher for persons ages 15 to 25 relative to persons ages 26-35 if the first provider was an acute behavioral health emergency or inpatient setting (rate ratio=1.286; 95% confidence interval=1.24-1.33). Case confirmation rates were 90% for OnTrack NY and 53% for the MCOs. Adjusted annual IR of first diagnosis of psychosis was 272 per 100,000.

Conclusions: Incidence of first psychotic diagnosis in this Medicaid population was higher than previously found in insured populations. Future work will focus on algorithm refinements and piloting outreach. Administrative data algorithms may be useful to providers, Medicaid MCOs, and state Medicaid authorities to support case finding and early intervention.