On December 10^th, 2019, the Department of Health and Human Services Office of Inspector General (OIG) released a report, which evaluated how Medicare Advantage Organizations (MAOs) used chart reviews to increase risk adjustment payments for Medicare Advantage (MA) beneficiaries in the 2017 payment year (2016 Dates of Service [DOS]). While the Centers for Medicare and Medicaid Services (CMS) did not completely agree with all of OIG’s findings, they did concur with OIG’s recommendations to provide additional oversight of MAOs.

Additionally, while the OIG study did have limitations, the findings are still highly relevant, and come at a time when CMS continues to push forward with Risk Adjustment Data Validation (RADV) audits to ensure that all diagnosis data submitted to CMS by MAOs is accurate and complete. (The payment year [PY] 2014 RADV audit started last year, and another 2015 RADV recently started.) It is important to note that CMS considers the use of RADV audits as the best approach for ensuring MAOs are documenting diagnoses appropriately.

Important Details on Risk Adjustment in Medicare Advantage

CMS risk-adjusts payment using diagnoses submitted by MAOs and pays a higher capitated payment to MAOs that report a higher level of illness burden for members. MAOs submit these diagnoses through two submission processes: Risk Adjustment Processing System (RAPS) and Encounter Data Processing System (EDPS).

To be eligible for risk adjustment, a diagnosis must be documented in a medical record as a result of a face-to-face visit. Currently, CMS allows plans to submit chart review diagnoses either linked or unlinked. A linked chart review diagnosis can be traced back to a previously accepted service record or original encounter. An unlinked chart review diagnosis occurs when MAOs cannot identify the specific service or encounter associated to the diagnosis. It is important to note that, at this time, CMS still considers both linked and unlinked chart reviews as acceptable methods to submit risk adjustment-eligible diagnoses.

Key Takeaways from the OIG Report

Below are some of the major findings in the OIG evaluation of chart reviews in MA:

• MAOs almost always used chart reviews to add, rather than to delete, diagnoses. Over 99% of the chart reviews in the OIG study added diagnoses.
  • In CMS’ response letter to the OIG study, CMS did state that, “chart review records are intended for the submission of additional diagnosis codes for risk adjustment. Based on their reviews of medical records, MAOs may also use chart review records to delete previously submitted diagnoses codes that are not supported by those medical records; if they identify unsupported codes, MAOs must delete them.”
    
• Although limited to a small number of beneficiaries, almost half of MAOs reviewed in the study received payments from unlinked chart reviews where there was no single record of service being provided to the beneficiary. Furthermore, more than one-third of MAOs linked some, but not all, of chart reviews to original service records.
  
• CMS estimated that $2.7 billion in risk-adjusted payments were paid for chart review diagnoses that MAOs did not link to any service provided to the beneficiary. However, the OIG study did not take into account the blended payment model (RAPS/EDPS) with this calculation. The actual value of these payments using the blended model is approximately $675 million.

Implications for Medicare Advantage Organizations

As a result of these findings, OIG made three recommendations to CMS with which CMS agreed. Below are the recommendations along with implications for MAOs:

Provide targeted oversight of MAOs that received risk-adjusted payments resulting from unlinked chart reviews for beneficiaries with no service records in the 2016 encounter data.

Implications: MAOs that submitted chart review diagnoses, but did not submit any other service records for the beneficiary, were reimbursed by CMS to care for the member; however, no evidence of services or treatments that MAOs provided to these beneficiaries was submitted to CMS.

What MAOs Can Do: Evaluate overall data integrity and perform an in-depth analysis of the process for submitting complete RAPS and encounter data to ensure that all appropriate service records are reported to CMS. Improved data quality can also improve the rate of linking chart review diagnoses to original encounters. This is especially important as CMS continues to rely on encounter data more than RAPS submissions for risk adjustment payments, which was announced the latest CMS Advance Notice.

Conduct audits that validate diagnoses on chart reviews in MA encounter data. 

Implications: 2015 was the first year in which CMS started to use encounter data to calculate risk adjustment payments. CMS has stated that they will include diagnoses from chart reviews in RADV audits.

What MAOs Can Do: Perform an over-read validation audit on a sample of chart review diagnoses submitted to CMS that resulted in payment. If any chart review diagnoses fail validation, MAOs can submit these as deletes to CMS (even if the period is closed). In general, MAOs should consider the use of chart reviews to submit both adds and deletes to CMS—i.e., incorporating a “two-way look” for HCCs, especially on claims.

Further, MAOs should implement an HCC compliance program and/or a rigorous QA/over-read process as part of an overall risk adjustment strategy to ensure that diagnoses from encounters, as well as diagnoses from chart reviews, are accurate and valid. This is especially important for MAOs who use a third-party vendor for chart review coding.

A comprehensive compliance program must include ongoing evaluation throughout the year that identifies and validates diagnoses that are at high risk of being submitted erroneously without the proper documentation, as well as the performing of outlier audits.

Reassess the risks and benefits of allowing chart reviews that are not linked to service records to be used as sources of diagnoses for risk adjustment.

Implications: At this time, CMS will continue to allow the use of chart review records both linked and unlinked as sources for risk adjustment payment. However, this may not always be the case, which would result in a reduction of payments for MAOs that submit a high number of unlinked chart review records.

What MAOs Can Do: As best practice, MAOs should continue to submit both linked and unlinked chart review records to CMS. The linking of chart review records to the original encounter can be complex and burdensome to an organization’s submission process infrastructure. However, MAOs should evaluate current chart review processes, specifically around the linking process and logic, to identify whether the linking logic is too stringent and to submit more chart review diagnoses as linked records. As an example, MAOs can identify cases where previously unlinked chart review records can be linked back to the original encounter by analyzing the use of rendering versus billing provider and DOS ranges in the linking process. If MAOs use a third-party vendor for chart review, there must be adequate vendor oversight and QA of the submitted codes.

Lastly, MAOs should start to implement and invest in prospective provider programs (including concurrent chart review) to capture all relevant member diagnoses at the point-of-care and ensure data is both documented in the medical record and submitted  appropriately on the claim/encounter that is submitted to the health plan.

Conclusion

Historically, MAOs have used risk adjustment programs like chart reviews to enhance risk-adjusted payments from CMS by supplementing submissions to CMS with more complete diagnoses data from chart review programs. MAOs should evaluate current risk adjustment programs to ensure payments from CMS are both complete, and more importantly, accurate. Accuracy in risk adjustment should include “looking both ways” by using chart reviews to add and verify/delete diagnoses from previously submitted encounters.

Please contact Jeff De Los Reyes, leader of Risk Adjustment and Healthcare Analytics Advisory practice at Gorman Health Group, at jdelosreyes@ghgadvisors.com for more information.