Preliminary first year findings from the evaluation of FIMC in Southwest Washington

The Fully Integrated Managed Care (FIMC) approach to delivering Medicaid-funded physical and behavioral health care was implemented in the Southwest Washington region (Clark and Skamania counties) in April 2016. The FIMC model integrates purchasing of Medicaid physical and behavioral health services within a single accountable managed care organization (MCO), in contrast to the presence of separate medical MCOs and behavioral health organizations (BHOs) operating in the balance of the state. The Health Care Authority has contracted with the DSHS Research and Data Analysis Division to partner with the University of Washington in the evaluation of the FIMC model. This document summarizes preliminary first year evaluation findings based on outcome data through CY 2016, including nine months of beneficiary experience under the FIMC model.

Evaluation Approach

The evaluation of the FIMC payment model tests whether the integrated purchasing of physical and behavioral health care:

  • Improves access to and engagement in needed behavioral health services;
  • Improves coordination and quality of physical and behavioral health care;
  • Reduces potentially avoidable use of emergency department (ED), inpatient, and crisis services;
  • Improves beneficiary level of functioning and quality of life, as indicated by improved labor market outcomes, increased housing stability, and reduced criminal justice involvement; and
  • Reduces disparities in access, quality, utilization, and social outcomes between Medicaid
  • beneficiaries with behavioral health treatment needs, relative to other Medicaid beneficiaries.

Preliminary findings are based on statistical tests of the relative change from CY 2015 to CY 2016 in 19 outcome metrics for adult Medicaid beneficiaries in Southwest Washington, relative to the experience of Medicaid beneficiaries in the balance of the state. Final evaluation analyses will use a regression-based difference-of-difference approach.

Summary of Preliminary Findings

Of the 19 outcome measures analyzed:

  • 10 showed statistically significant relative improvement for Medicaid beneficiaries residing in the FIMC region,
  • 8 showed no significant difference between the FIMC region and balance of state, and
  • 1 showed a statistically significant relative decline in the FIMC region.

For the one metric where the relative change in performance in the FIMC region was negative and statistically significant (ED utilization per 1,000 member months), the performance level in the FIMC region in CY 2016 was second best among all ACH regions.

Subgroup analyses focused on the experience of Medicaid beneficiaries with serious mental illness or co-occurring mental illness and substance use disorder showed a similar pattern of improvement in outcomes from CY 2015 to CY 2016 for persons residing in the FIMC region, relative to the experience in the balance of the state.

Summary of Changes in Outcome Metrics in the FIMC Region Relative to the Balance of State, CY 2015 to CY 2016

Metrics showing statistically significant improvement for adult Medicaid beneficiaries residing in the FIMC region, relative to the balance of state:

  • Adults’ Access to Preventive/Ambulatory Health Services
  • Cervical Cancer Screening
  • Chlamydia Screening in Women
  • Comprehensive Diabetes Care – Hemoglobin A1c Testing
  • Antidepressant Medication Management – Continuation Phase Treatment
  • Follow-up after ED Visit for AOD Dependence-Within 7 Days
  • Follow-up after ED Visit for AOD Dependence-Within 30 Days
  • Percent Homeless – Narrow Definition
  • Percent Homeless – Broad Definition
  • Percent Arrested

Metrics showing no significant difference between the FIMC region and balance of state:

  • Breast Cancer Screening
  • All-Cause 30-Day Readmission
  • Antidepressant Medication Management – Acute Phase Treatment
  • Adherence to Antipsychotics for Persons with Schizophrenia
  • Diabetes Screening for People With Schizophrenia or Bipolar Disorder
  • Inpatient Utilization per 1000 Coverage Months
  • Balance of Home and Community-Based Services and Nursing Facility Utilization
  • Percent Employed

Metrics showing statistically significant decline in the FIMC region, relative to the balance of state:

  • Outpatient ED Utilization per 1000 Coverage Months