Tag Archives: risk adjustment

Transition from RAPS to EDS data decreases Medicare Advantage risk scores

Milliman consultants Deana Bell, David Koenig, and Charlie Mills performed a study of how the transition from Risk Adjustment Processing System (RAPS) data to Encounter Data System (EDS) data is affecting payment year (PY) 2016 risk scores and revenue for Medicare Advantage organizations (MAOs). Fifteen MAOs participated in the study, reflecting a cross section of small- and medium-sized organizations and representing over 900,000 members in 154 plans. The consultants offer perspective in their article “Impact of the transition from RAPS to EDS on Medicare Advantage risk scores.”

Overall, the study found that the median percentage difference between PY 2016 risk scores based on RAPS and the EDS-based risk scores is 4.0%. The percentage difference is larger for special needs plans (SNPs) and smaller for general enrollment plans as shown in Figure 1. The prior year’s diagnoses make up a larger component of SNP members’ risk scores, compared to general enrollment plans, so the risk score impact for SNP plans is larger.

[The authors] have not attempted to quantify what portion of the difference between RAPS and EDS is due to incompleteness of the EDS submissions, issues with CMS’s return files (revised MAO-004 files), changes to filtering logic, and the effect of claims coding errors.

As an illustration, the potential Part C PY 2016 revenue using the median difference of -4% between RAPS and EDS results in a reduction of approximately $40 per member per year, assuming approximately $800 in Part C risk-adjusted revenue and a 1.0 RAPS-only risk score. To the extent that this -4% gap persists in future years, the revenue impact will grow because the EDS-based risk score will make up an increasing portion of the final risk score (e.g., with the 25% EDS weight in PY 2017, the per member reduction would be about $100 per year).

This article is the second in a series of articles on the transition to EDS. For more information about the EDS and RAPS data used in MA risk scores, read “Medicare Advantage and the Encounter Data Processing System: Be prepared.”

Milliman Advanced Risk Adjustment™ (MARA™) software tops the competition in risk scoring study

laurent-dianeMilliman’s popular Milliman Advanced Risk Adjusters™ (MARA™) software topped the competition in the latest study by the Society of Actuaries (SOA), “Accuracy of Claims-Based Risk Scoring Models.” Milliman achieved the highest overall performance among concurrent models, beating out 11 other vendors. MARA’s prospective models scored similarly well, ranking at or near the top in all four primary metrics measured.

MARA’s performance is proof of what we’ve known all along—the Milliman Advanced Risk Adjusters software is a leader in claims-based risk scoring models, and one of the most accurate tools on the market.

Since the passage of the Patient Protection and Affordable Care Act (ACA), risk scoring models play a central role in predicting or explaining healthcare expenditures. As in each of the previous SOA studies, coefficient of determination, or R2, was used as one measure of predictive accuracy and indicates how well data fit a statistical model. With an R2 of 55.4%, MARA’s concurrent diagnosis and pharmacy model (CxAdjuster) leads the competition with the highest observed R2 for uncensored costs. For costs censored at $250,000, the same model achieved the highest R2 to occur in the study, at 66.7%.

We were especially pleased that in an exploration of ensemble methods of modeling, more weight was put on the MARA model than any other. To me that indicates that giving MARA the most credence resulted in the best performing composite prediction.

MARA models are valued by clients because they produce separate risk scores for major categories of service, including inpatient, outpatient, emergency room, physician, pharmacy, and other medical care. Beyond risk scores, all MARA models provide a probability of inpatient admission and emergency room visits, plus the contribution of clinical risk drivers for more than 1,000 medical conditions.

The study, released Monday October 25, is published by the Society of Actuaries.

Encounter data submission consideration for Medicare Advantage organizations

The Centers for Medicare and Medicaid Services (CMS) is shifting to the Encounter Data System (EDS) as the sole basis for risk scores of Medicare Advantage organizations (MAOs). EDS data will have a substantial influence on risk scores and revenue as early as payment year (PY) 2016.

In this article, Milliman consultants offer perspective on several analyses MAOs should consider to prepare data submissions for CMS. The authors also discuss some challenges MAOs may face when verifying that the data submitted through the EDS are complete and accurate and that all appropriate diagnosis codes are being accepted for risk adjustment by CMS.

Here’s an excerpt:

To provide an effective review of an MAO’s diagnosis code submissions, the following analytics can be undertaken:

• Calculation of risk scores from each diagnosis source: RAPS, EDS, and source claims/ chart review data
• Plan-level and member-level comparisons of risk scores based on each diagnosis source
• Analysis of submission gaps
• Analysis of coding gaps

To perform the EDS submission review, a possible first step is to create a “plan report card,” which summarizes the risk scores under accepted RAPS and EDS submissions and the risk scores based on all diagnosis sources (claims and chart review data) after applying the MAO’s specific RAPS filtering logic and the EDS filtering logic released by CMS.

Figure 1 provides an example of a potential Plan Report Card for PY 2016 EDS submission review. In this example, there is a 4.1% gap between the EDS risk scores and the risk scores after the MAO applied the EDS filtering logic to the source claims data. Also, based on the CMS return data, the EDS risk scores are four points lower than the RAPS scores. This indicates that the EDS submissions may be incomplete and that there are diagnoses in the source claims data that the CMS filtering logic has rejected.


If submissions to CMS contain all necessary data elements to successfully pass the filtering logic, the risk scores based on RAPS and EDS return data should match the risk scores calculated from the source claims and chart review data. In addition, if the RAPS and EDS filtering logic are the same, the RAPS and EDS risk scores should also be the same. However, there can be gaps between what is submitted and accepted by CMS and the claims and chart review data because of:

• Incomplete data submissions (e.g., claims being inadvertently filtered out or dropped, missing chart review data)
• Inaccurate data submissions (e.g., the wrong medical codes, such as incorrect bill type, being used in the submissions)
• CMS system errors (e.g., failure to match diagnosis data with the correct member)
• Other potential process errors

Furthermore, comparison of the RAPS and EDS risk scores will indicate whether the MAO’s revenue is being adversely affected by the move from RAPS to EDS. A focused look at the MAO’s own coding practices as they compare to Medicare FFS coding standards and EDS filter criteria can identify the coding gaps that may drive lower risk scores under EDS.

Maximize ACA risk adjustment with EDGE server action plan

Effective management of information entered into an External Data Gathering Environment (EDGE) server may save health plans millions of dollars in risk adjustment transfer payments. In this paper, Milliman’s Jason Petroske and Alan Vandagriff outline best practices that issuers should consider as part of their annual EDGE server submission cycles to maximize risk adjustment results.

Here’s an excerpt:

Complete and accurate data is a critical element in capturing—and, more importantly, in receiving compensation for—a health plan’s true level of risk. While navigating the first two years of EDGE submissions, we have mapped out a comprehensive action plan focused on three main areas that any issuer can integrate into its data management framework:

Establish a robust review and reconciliation process: Create a continuous process for reviewing and reconciling EDGE submissions to internal data sources. Identify key metrics for data completeness and use the test environment to ensure each EDGE submission passes these standards before finalizing in production.

Prioritize error corrections: Not all errors are created equal, so have a strategic plan for correcting errors and improving data quality. Understand the economics of risk adjustment to help effectively deploy and allocate resources.

Track data quality and establish benchmarks: Track and benchmark data quality and submission results over time. Look for patterns in errors or outliers from prior submissions as these can be signals of systemic weaknesses in the overall data management process.

Evaluating risk adjustment transfer payments

Risk adjustment transfer payments continue to have financial implications on insurers in the commercial individual and small group marketplaces. In this analysis, Milliman consultants provide an overview of 2015 transfer payments, comparing them against 2014 results. The authors also explore the following conclusions from their report.

• Total risk adjustment transfer payments at the national level remained at about 10% of premium in the individual market and 6% of premium in the small group market.
• Roughly one in four issuers offering plans in a given state or market in both 2014 and 2015 switched between payer and receiver status.
• Statewide risk scores rose more year-over-year than the movements in market demographics and average plan benefit richness would have suggested.
• Where available, the interim risk adjustment report did not provide a reliable indication of the ultimate value of the 2015 risk score.

Development and implementation of functional-based risk adjustment for Medicaid Managed Long Term Services and Supports

As the number of Medicaid Managed Long Term Services and Supports (MLTSS) programs increases, significant momentum is also building around the development of tools to adjust managed care organization (MCO) payments using the functional, medical, and behavioral needs of their members. These tools match payment to risk and align MCO and MLTSS program incentives more effectively. While the planning, development, and implementation needs of a functional-based risk adjustment (FBRA) mechanism are significant, the improvements realized in MLTSS programs are worth the effort.

Milliman actuary Michael Cook provides perspective in his article “Functional-based risk adjustment for Medicaid Managed Long Term Services and Supports: Part 2.” This article is the second in a two-part series on functional-based risk adjustment (FBRA). To read part one, click here.