Risk Adjustment: Models and Using Encounter Data

Transcription

1 Risk Adjustment: Models and Using Encounter Data May 12, 2009 Workgroup on Managed Care Reimbursement Agency for Health Care Administration Tallahassee, Florida

2 Risk Adjustment Models

3 General Risk Adjustment Design Elements Types of Risk Adjustment Approaches Additive approach Members assigned to multiple categories (demographics and disease conditions) Categorical approach Members assigned to a single category based on overall illness burden 2

4 Model Selection Considerations Specificity of the model to the population to which it is being applied Cost of the software Access to data of sufficient quality Transparency of the mechanics and results of the model Application purpose (payment or quality assessment) Usability in the current market 3

5 Model Use by State Medicaid Program ACG CDPS CRG ERG MRX Delaware Colorado New York Arizona Delaware Maryland Delaware Florida Minnesota South Carolina Tennessee Michigan New Jersey Ohio Oklahoma Oregon Pennsylvania Texas Utah Virginia Wisconsin Washington 4

6 Moving Forward The Agency continues to review risk adjustment models to identify relevant features of each model and their applicability to Florida Medicaid. Plans to use CDPS have not changed; however, we believe we should perform our due diligence by looking at other options. We will continue to partner with the plans as we move forward. 5

7 Using Encounter Data in Rate Setting

8 Why Use Encounter Data in Rate Setting? CMS preferred data source as base data Uses the actual experience for the covered populations at a detailed level Allows the State to gain a more in-depth understanding of the factors driving cost and quality within its managed care program Can achieve greater accuracy and better match the payments to the participating plans risk 7

9 Base Data for Rates Primary Uses Requires complete and accurate data so that the base experience is not over or understated for each population and service category. Should have a good benchmark to measure the quality of the paid fields reported on encounters (i.e., financial reports). Financial reports that can be helpful in understanding the quality of encounter data include claims lag reports and annual income statements. Encounters can be blended with other source data (e.g., financial reports or FFS data) for some or all service categories until encounter data are fully complete. 8

10 Risk Adjustment Primary Uses Risk adjustment can be performed using pharmacy or nonpharmacy encounters. Focus is on complete and accurate information for the required fields: Pharmacy must have good national drug code information Non-pharmacy must have good diagnosis information Other information such as procedure and revenue codes Since results are budget neutral, each plan s data quality can have an impact on capitation payments for other plans. 9

11 Utilization Studies Primary Uses Compare utilization patterns for providers, regions, etc. Compare results to FFS program for benchmarking and policy decisions 10

12 Rate Setting Additional Uses Detailed encounters provide means for making clinically-supported actuarial adjustments for: Non-emergent and preventable ER visits Preventable Inpatient stays Inappropriate and/or non-cost effective use of pharmaceuticals Pay for Performance initiatives can be supported through encounter data. Claims Distributions derived from reported encounters can assist in evaluating high cost claims, reinsurance, stop loss, or other risk sharing arrangements. 11

13 Integrating Encounter Data into Rate Setting Rate Section Base data for rates Risk Adjustment Estimate program changes Utilization comparative analyses Trend analyses Compare HMOs at detailed level etc. Clinical Quality Medical management adjustment ER - Low Acuity Non-Emergent (LANE) Hospital - Prevention Quality Indicators (PQI) Pharmacy Overall cost effectiveness etc. Rate-setting Financial Section Validation of financial reports Flexibility in scope & depth of monitoring Analyzing COB, TPL Fraud, Abuse, and Waste Provider contracting analysis etc. Encounter Data in supporting role HMO relations Policy decisions Risk pools, risk sharing Performance based contracting FFS data comparisons etc. 12

14 Questions