BY ZACK GERBARG, MD, CPC AND GREG SINAIKO, MPH
Many providers and Medicare Advantage plans are just now waking up to the fact that the Centers for Medicare & Medicaid Services has radically changed the rules about how health plan premiums are calculated. The change is dramatic and, by not acting now, health plans may lose out on millions of premium dollars to which they otherwise would be entitled. Premiums for healthy members are decreasing while premiums for members with significant illnesses are increasing. If health plans do not capture the data to document important diagnoses, CMS will calculate premiums at the healthy member rates. The result will be that some sicker members who require additional care will not have an adequate premium to support their care. How did this come to be?
Background
Previously, patient demographics were the major driver of Medicare health plan premiums. For example, a 70-year-old male member living in a particular county
might generate $650/month of premium, regardless of his overall health status. Starting in 2003, CMS began to phase in a severitybased premium payment system. For 2006, 75 percent of the premium will be calculated by the member’s medical severity and only 25 percent will be based on demographics.
What this means is that rather than a $650/month premium ($7,800/year), a healthy 70-year-old man might only generate a $400/month premium ($4,800/year); whereas, if that member has congestive heart failure and chronic obstructive pulmonary disease (COPD), the premium might be $850/month ($10,200/year). The annual difference could be $5,400 for that one member.
Health plans that have attracted relatively healthy Medicare Advantage members or who do not capture and submit important diagnoses will see a sharp decrease in their annual premium. For most Medicare Advantage plans, the new premium calculation will result in a premium decrease, but we have seen this trend reversed by taking specific action.
Providers’ Role in This Process Claims and encounters from physicians and hospitals are the major sources of data that CMS uses to determine patient severity. If providers do not document and then submit claims and encounters with the members’ appropriate diagnoses, the risk score will reflect a healthier population than actually exists and the premium will be lower.
CMS currently requires that the member’s diagnoses be submitted each calendar year and be based on a hospitalization or a face-to-face visit with a physician. Typically, hospitals do a better job of diagnosis documentation and coding because they need this data to ensure accurate DRG payments. Most hospitals have certified coders on staff to help them achieve this, but
other hospitals face serious challenges if complete data were not captured in the first place. Each significant diagnosis is grouped into one of about 100 Hierarchical Care Categories (HCCs) and each HCC is assigned a weight that is used to determine the risk adjustment score for a member. For example, the diagnosis for uncomplicated diabetes is coded using ICD-9 code 250.00; and this corresponds to HCC 19, which has a weight of 0.200. Congestive heart failure is ICD-9 code 428.0, which corresponds to HCC 80 and has a weight of 0.417. Using CMS tables and the published calculation formula, the risk scores and premium dollars can be determined. Members with more severe and multiple medical conditions will have higher risk scores. CMS allows Medicare Advantage plans to submit diagnosis codes retrospectively and will pay the increased premium. The window of opportunity for submitting diagnoses is up to 16 months after the end of the calendar year.
What this means is that plans have until May 2006 to submit diagnoses from dates of service in calendar year 2004 and then CMS recalculates the 2005 premium. Diagnoses from 2005, which impact the 2006 premium, can be submitted any time up until May 2007. The earlier health plans submit the data, the sooner they are paid the increased amount.
What is the Challenge?
CMS performed a test audit of medical records for six Medicare Advantage health plans and found approximately 50 percent of the diagnoses submitted were not accurate, either because there was no documentation in the medical record to support the diagnosis or because the wrong code had been selected.
That is a frightening result. When audits are performed with certified coders, with additional expertise in diagnosis coding, quality checks will typically show errors of roughly 2 percent.
Part of this improved result can be attributed to the use of laptop-based software that helps coders prevent making many common errors. In addition, most experienced professionals will put in place a careful screening process and apply extensive coder education and oversight. Indeed, medical record audits by knowledgeable certified professional coders have consistently yielded a significantly positive ROI. Results over a population of 10,000 or more members have consistently shown that 10 percent to 20 percent of primary care physician (PCP) medical records yield a diagnosis that is unique that corresponds to an HCC.
While there is a wide range for each individual physician, this overall result has been constant in more than five different geographic parts of the country. Audits of specialty medical records usually yield higher weight diagnoses than PCP medical records.
Benefits and Risks
Approximately 70 percent of the diagnosis data collected for Medicare risk adjustment comes from physician claims and encounters. What physicians document in their medical records is the most important component of accurate diagnosis coding. When auditing diagnoses submitted for severity adjustment, CMS requests the specific note in the medical record in that calendar year that supports the diagnosis. If physicians and their staff can be educated to document and code correctly, then health plans can feel confident that the data they receive are accurate and complete.
The goal in any HCC coding effort must always be accurate documentation and coding. It is relatively easy for physicians and their staff to misinterpret coding rules, which can be fairly complex. For example, certi- fied coders know they cannot submit a code for a diagnosis just because it is noted in the problem list in the medical record—a problem list is not considered a valid source for diagnosis coding.
Clearly, health plans receive a higher premium if they capture data on all the illnesses for their members. Providers will benefit directly if they are capitated at a percent of premium.
A health plan that loses Medicare premium revenue may attempt to reduce the rates paid to the plan’s providers. Another benefit is that more accurate information about the membership population is valuable in supporting case management and disease management initiatives.
Summary of Key Issues
Here is a summary of the central facts for health plan executives to know regarding Medicare risk adjustment: s diagnosis data submitted to CMS have a multimillion dollar impact on health plan premiums; s diagnoses submitted for dates of service in a calendar year determine the premium for the following year. s CMS allows a window of 16 months from the end of a calendar year to submit any and all diagnoses for dates of service in that calendar year from valid sources of diagnosis data. Then CMS pays any premium increases retrospectively—this is formula driven; s diagnoses submitted to CMS should be supported by medical record documentation that will withstand an audit; and s claims and encounters from physicians and hospitals are the key sources of diagnosis data.
Conclusion
There are various strategies that Medicare Advantage health plans can employ to ensure accurate and complete diagnosis documentation, coding, and data capture. Medical record audits by knowledgeable coders and physician education are two approaches that have consistently yielded significant ROI. The benefits of undertaking the challenge of improving diagnosis accuracy can be measured in millions of premium dollars.
Dr. Zachary Gerbarg is principal of Eagle Medical Management, a medical management and executive leadership consulting firm, and Greg Sinaiko is president of The Coding Source, a national provider of health care coding services. More information is available at http://www.eaglemedicalmanagement.com or http://www.thecodingsource.com.
Case Study
One of our clients serves approximately 20,000 Medicare Advantage members. Based on the claims that were collected and submitted to CMS, the average risk score was 0.922 for this Medicare population. Through medical record audits by certified professional coders and targeted physician education, we were able to identify and capture additional diagnoses that had been missed.
This result was an improved risk score of 0.964, which increased the 2005 premium by nearly $4 million (note premium was only 50 percent risk adjusted in 2005); the return on investment (ROI) was approximately 4:1. The same result for 2006 with a 75 percent risk adjustment will be roughly $5.5 million with an ROI of approximately 6:1; and at 100 percent risk adjustment in 2007, the premium increase will be roughly $7 million with an ROI of approximately 8:1.
COPYRIGHT © BY THE BUREAU OF NATIONAL AFFAIRS, INC., WASHINGTON, D.C.
No comments:
Post a Comment