Evaluating sustained reach and effectiveness of collaborative care models: A Cross-sectional study of the New York State Collaborative Care Medicaid Program

BackgroundLittle data exists on collaborative care (CC) sustainability. ObjectiveDescribe and determine predictors of long-term CC reach and effectiveness Design: Cross-sectional observational study of the NY State CC Medicaid Program (CCMP), involving technical assistance (TA), quality monitoring, and fee-for-quality Medicaid reimbursement codes for implementing CC. We included clinics participating in CCMP from 2012-2019 with [≥]0.5 full time equivalent (FTE) care manager and available 2021 and/or 2021 data. Main MeasuresClinic (size, type, region, enrollment year); and CC program (care manager FTE, caseload/care manager FTE [target 100-150], screening [proportion of clinic screened for depression], engagement [proportion of CC patients contacted/engaged monthly], and psychiatrist consultations in unremitted patients) characteristics. Outcomes were reach (proportion of screen-detected depressed patients enrolled in CC) and effectiveness (proportion of CC enrolled patients achieving remission or 50% reduction in depressive symptoms [Target 50-60%]). We used multilevel negative binomial regression models, adjusting for clustering by healthcare system and county. ResultsOf eligible 160 clinics, 71.2% were Federally Qualified Health Centers (FQHCs); the median caseload/care manager FTE was 55.1, reach 13.0% and effectiveness 42.0%. In multivariable analyses, key CC factors associated with reach included engagement (adjusted Rate Ratio [aRR]=3.99 [1.82, 8.76]), care manager FTE (aRR=1.06 [1.02, 1.10]), and caseload/care manager FTE (aRR=1.23 [1.17, 1.29]); smaller clinic size (aRR=0.60 [0.53, 0.69]), earlier adoption (aRR=0.40 [0.23,0.69] in 2017-2019 vs. 2012-2014), and academic/private clinics (vs. FHQC) (aRR=0.66 [0.45, 0.96]) were also predictive. Caseload/care manager FTE (aRR=1.04 [1.01, 1.07]), psychiatry consultations (aRR=1.55 [1.19, 2.00]), and FQHCs (aRR=1.19 [1.02, 1.40]) were associated with greater CC effectiveness. ConclusionDespite ongoing fiscal and TA, CC clinics particularly struggle to achieve long-term reach. While majority FQHCs limit generalizability, we provide several targets for selecting ideal settings for CC, optimizing the pace of sustainability and considering de-implementation efforts when futile. Primary Funding SourceAgency for Healthcare Research