Text Size
Sunday, March 29, 2015
wns_logo2.png

Participate in the Project

Welcome to the website for the Wakely National Risk Adjustment Reporting Project. The objective of the project is to provide interim information for the risk adjustment program under the Affordable Care Act. This information is critical for health plan issuers for purposes of pricing health plans and estimating the ACA individual and small group risk adjustment transfer payments. Please review the tabs below for detailed information on this project.

Overview

Health plans participating in the project will receive the following benefits:

  1. Estimated 2014 risk adjustment payables and receivables
  2. Estimates of health plan specific risk scores by state and market for purposes of pricing products in 2015.
  3. Data diagnostics to assess potential data issues early, so they can be corrected.
  4. Detailed reports that show the key drivers behind risk scores.
  5. Guidance for estimating the pricing impact of high risk pool members and previously uninsured members entering the individual insurance market.
  6. A deeper understanding of the ACA risk adjustment program.

Participation in the project will not require the transfer of any detailed data to Wakely or any other outside parties. Participating companies will only provide Wakely with summary level results after running the Wakely Risk Adjustment Reporting model code on their data. Wakely has implemented very strict controls and security to ensure that even the summary level results by health insurance company will not be shared. The only information that will be shared with other companies is aggregate results by state and market, across all health insurance companies. In fact, this aggregate information is the primary reason for companies to participate in this project.

It is important to the success of this project that we receive notification of intent to participate from as many health plans as possible, as early as possible.  It is important to note that intent to participate is not a binding agreement to participate, it only represents a health plan's intention at the time the notice is provided.  Actual participation is contingent on results being submitted to Wakely and payment. 

To protect the confidentiality of health plan information and to ensure that the modeling is credible, Wakely has set a minimum participation requirement.  We will require participation from at least 75% of the state and market and at least three health plans each representing at least 5% of the market.  If high risk pool data is available, it will be included in the numerator and denominator for purposes of meeting the minimum participation requirements.  If it is not available, it will not be included.  If either or both of the market minimums are not met and would likely not be met, we will contact all companies that had agreed to participate in that state / market and see if they would be willing to waive the minimum requirements and proceed with the simulation.  We are using NAIC data from the Supplemental Healthcare Exhibit to determine enrollment by state and market.

How to Participate

We have tried to make it easy to participate in the Wakely National Risk Adjustment Reporting Project. Just follow these steps:

  1. Register with this website: To do this, please click on the Login button in the top menu.  Once you register, Wakely will list your company name on the Participants tab.
  2. Download & run the model: Click on the Participation Guide tab for specific guidance.
  3. Upload report: You are able to upload data as soon as you register with the site. The file to be uploaded is the WNRAR report that is output during the course of running the risk adjustment model. To upload data securely to your own private folder on our server, please go to the Data & Results page (you can also go there from the top menu).
  4. Access Results: Once we have compiled the results, we will notify registrants via email that the results are available on the website. You will be able to view and access results using the same process as in #4.

Technical Details

Wakely has issued several papers (see Resources tab) related to the risk adjustment and reinsurance provisions of the Affordable Care Act (ACA) . We believe it is critical for health plans to identify and correct data issues prior to submitting final data for 2014 and to have estimated risk scores available in early 2014 to price individual and small group products for 2015.

The following notes are intended to provide a high level discussion of the proposed reports that will be returned to the participants of the project. The focus of the 2014 results is a) pricing for 2015 products, and b) estimating risk transfers from the ACA risk adjustment program. A total of 9 versions of results will be provided for each of the 2014 runs. The result versions are described further below.

Unadjusted HHS (3 versions of results)

  1. Three versions of results will be provided.
    1. Version 1: This version will use all available 2014 information for a member in order to calculate the risk scores. The primary purpose of this version is to allow a comparison to 2013 partial year ‘unadjusted HHS’ version of results.
    2. Version 2: A second version will use only post-renewal experience (per HHS rules). This version is intended to match the HHS Edge server calculation.
    3. Version 3: A third version will be the same as the second version, except that no diagnoses exclusion criteria will be applied. The usefulness of the third version is to allow issuers to identify any potential issues with data coding, submission or market average risk
  2. Note that it is difficult to get a good sense of the ultimate relative morbidity (or relative risk scores) from the ‘Unadjusted HHS’ results due to:
    1. Partial eligibility: Partially eligible members, particularly within a small data window, can distort relative risk comparisons (for example, if one issuer has more or less partially eligible members compared to the market average). The calculated relativity may or may not hold up well as experience completes through the year.
    2. Renewal patterns: The potential impact of the renewal pattern across issuers is likely very significant, since in 2014 only the post-renewal experience is to be used towards calculation of a risk score. This is expected to bias the risk scores, such that a partial year comparison will likely not hold up well as experience completes through the year.
  3. For the reasons outlined in (c) above, we provide additional versions of results as described below.

Best Estimate of Morbidity (4 versions of results)

  1. In this version of results, each member’s risk score is completed to reflect a full year’s worth ofThe reason is that all 2014 members are expected to be eligible for full year enrollment into ACA compliant plans in 2015. An exception is policies that were originally cancelled in late 2013, however are now ‘in continuation’. Therefore we report an additional version of results, in which we account for the expectation that these continuation policies are expected to be eligible for risk transfer payments for only part of the year in 2015. Members that are known to have termed at the time that the data is run will receive a zero weight (since they will not be part of the 2015 risk pool).
    1. Version 1: Member risk scores adjusted to reflect a full year of experience, using the regression completion method.
    2. Version 2: Member risk scores adjusted to reflect a full year of experience, using the HCC completion method.
    3. Version 3: Member risk scores adjusted to reflect a full year of experience (except for continuation members), using the regression completion method. Continuation member risk scores are adjusted to represent the number of months that these members are expected to be eligible for risk transfer payments in 2015.
    4. Version 4: Member risk scores adjusted to reflect a full year of experience (except for continuation members), using the HCC completion method. Continuation member risk scores are adjusted to represent the number of months that these members are expected to be eligible for risk transfer payments in 2015. 

 Best Estimate of Risk Transfer (2 versions of results)

  1. In this version of results, each member’s risk score is completed to reflect a full year of experience (similar to the best estimate of morbidity version), using all available 2014 information on the member. However the risk score is then further adjusted to reflect the number of member’s post-renewal eligibility (i.e. eligibility that counts towards risk transfer payments) in 2014. This process is intended to get us a better indication of the full year risk transfer, and adjusts the scores for bias related to partial eligibility and renewal patterns.
    1. Version 1: This version uses all available 2014 information on a member to develop a full year risk scores using the regression completion approach. The score is then adjusted to reflect the risk transfer eligible months.
    2. Version 2: This version uses all available 2014 information on a member to develop a full year risk scores using the HCC completion approach. The score is then adjusted to reflect the risk transfer eligible months.

The results will be presented with sufficient detail such that pricing actuaries will be able to consider and appropriately reflect the impact of allowable rate variation due to age. We will include issues specific to each state / market in our communication of the average results.

Data provided to Wakely by health plans will not contain any financial expenditure information. Wakely will compile the results at a state and market level and provide those results compared to health plan results to each health plan that participated in a particular state / market.

Information provided to each health plan will include the following:

  1. Risk Score for health plan (individual, small group)
  2. Average risk score for state for the markets the health plan participates in (individual, small group). 
  3. Prevalence report showing proportion of member months by risk adjustment model category (age, gender, and condition categories such as diabetes, asthma, etc.). Risk weights will be included on the prevalence report and calculated 'driver' values will be included showing which categories are most significant in driving differences from the state / market average.
  4. It is important in a distributed approach to develop and review key data diagnostics. Such measures will be output by the software and include:
    1. Average number of months of eligibility during the data collection period,
    2. Number of unique members,
    3. Proportion of eligible members with a medical claim,
    4. Average unique diagnosis codes per claimant,
    5. Proportion of members/member months that were not grouped into a medical condition category.
    6. The output reports showing the above health plan values and state / market values will be provided as part of the reporting package

Wakely will review the results for each health plan for reasonability and only include health plan results in the state / market total if reasonable. We will contact health plans that submit unexpected results and work with them to identify and resolve any issues.

Participation Guide

The instructions for running the model for 2014 Test Run of the project are included below. If you are having difficulty accessing the pdf file, you can also link directly to the document by clicking here

FAQ

Frequently asked questions are answered below. If you are having difficulty accessing the pdf file, you can also link directly to the document by clicking here.

Status

This information is available only to participants in the project. You are not logged in. Please login to access this information.

RAPID

Wakely Consulting Group (WCG) is developing an approach to efficiently run an independent validation of the key EDGE RA and RI calculations. WCG receives accepted EDGE records and standard outbound reports (such as RIDE, RACSD, and RARSD). WCG applies a rigorously developed routine of algorithms and models to a) run an independent version of the RA and RI calculations, b) compare to EDGE outbound detailed / summary files, and c) develop reports that aid to uncover sources of discrepancies.

Contact us to learn more about RAPID, or to schedule a demonstration/discussion.
Syed Mehmud: This email address is being protected from spambots. You need JavaScript enabled to view it.