Comparative Case Study of Two Risk Adjusted Plans
by Todd M. Husty, D.O., FACEP – President and Medical Director MARSI
Prologue
The two physician groups referenced in this document provide services to Medicare Advantage patients within the same town, have the same demographics and actually operate within a few miles of each other. MARSI reviewed each groups’ operations and performed chart audits for each group. While the audit results were dramatically different, their operations bore many similarities – with the exception of one critical area.
Physician Group I
I just returned from perhaps my most interesting experience as a physician educator in 20 years. I was training a group of physicians, in separate sessions of approximately 40 physicians each, on documentation and coding. Usually, physicians view this subject as a necessary evil – and probably more evil than necessary.
But, these physicians were very attentive. They were listening. They asked well-informed and important questions. Not one of them complained or said “What do you want me to do, write on charts or take care of patients?”
When I was finished, many of them came by and thanked me for my presentation with comments like “That was very helpful, I had forgotten some of those things” or “I wasn’t doing some of those things correctly. Thank you.” I can honestly say that’s never happened to me before with such a large group.
This organization is a physician group owned by a Hospital. The physicians are employed through this organization. They include approximately 85 primary care physicians. These PCPs have approximately 15,000 Medicare Advantage patients.
Four years ago, the organization hired a CPA who had previously worked for a large healthcare organization with a much larger number of Medicare Advantage patients. She recognized the inner workings of Medicare Advantage contracts, with HCC/Risk Adjustment as an ICD-9-based reimbursement system which depends on physician documentation and coding for the majority of its revenue.
She enacted, with the help of upper-level management support, significant changes in the process. This included physician contracting and incentives, physician education, and a 100% review of all documentation (including lab, x-ray, consultations, hospital admissions and other diagnostic services). This review included a process of quarterly face-to-face meetings with each physician in order to assist them in responding to requests for clarification of documentation.
After three years of operation, her team of nurse reviewers, coders and financial specialists brought their MRA or RAPS score up from approximately 1.0 to 1.4. Not only that, but the group felt strongly that the physicians had a higher satisfaction level because they actually understood more about documentation and coding. They decided that they wanted to have an audit to make sure that they were doing things appropriately, especially in light of their success.
After a nationwide search, the organization decided upon a contract with Medical Audit Resource Services, Inc. (MARSI) to do an audit. The instructions were to evaluate coding for accuracy and the documentation to see if it supports the assignment of the codes. We were to look for both over coding and under coding, including missed opportunities.
MARSI has audited tens of thousands of records for HCC/Risk Adjustment. We are accustomed to finding a high percentage of both over codes and under codes. By high percentage we mean 20 to 60%.
Our review of this organization was very refreshing. We found minimal over coding which, quite frankly, were in gray areas that are debat¬able. There was no pattern to the over codes; they were random.
They were less than CMS‘s “acceptable” error rate of 2.5% and, therefore, they were certainly far less than the average.
We did find opportunities for improvement but they were far fewer than with the majority of our other risk adjustment reviews. We believe that this physician organization would have a MRA or RAPS score of 1.5 or 1.6, if they did everything right – which is almost impossible with such a complex system.
The organization does incentivize the physicians. They do better if they have proper documentation and proper coding. But, the incen¬tives also reward for a higher quality of care. When I discussed examples with the physicians of their documentation and coding, it was evident that they all realized that they were being rewarded for earlier detection and better treatment of disease states.
The physicians are “given” a pool of revenue (a share of the total revenue) that increases with more accurate diagnoses. The incentives diminish rapidly with each admission to the hospital, the most expensive part of healthcare. It is clear that taking better care of patients, keeping them out of the ER and preventing admissions whenever possible, results in more shared revenue. The reimbursement structure is aligned with the practice of medicine.
Historically, pooled revenue was a disincentive for quality care. The pool was based on a fixed amount per covered life. The incentive was to not do tests or order consults. The difference here, with HCC/Risk Adjustment, is that revenue increases with better detec¬tion of disease states.
Utilizing diagnostic studies, and using but not over-utilizing consul¬tants, results in more identified diagnoses and more revenue. Inter¬estingly, even hospitalization results in a long-term increase in revenue if it is for new conditions. Obviously, if a patient has a previously identified condition, it is important for their physician to provide the highest quality of care and promptly respond to any changes.
HCC/Risk Adjustment is a near-perfect system if there is complete and total physician buy-in and participation. That is a lot to ask or to expect but, with incentives properly aligned and an ongoing support structure, any organization of physicians can show similar success.
Physician Group II
Having had an eye opening experience auditing a physician group that had instituted their own thorough, 100% review and follow up process for their Medicare Advantage patients, we became excited when we were provided the opportunity to audit a similar physician group with the same geographic and demographic basis. Both groups had provided physician education. Physicians in each group had the opportunity to prosper by finding more diseases, documenting, coding correctly and keeping patients healthy and out of the hospital. The main difference between the two groups was the presence or absence of robust chart audit and follow up feedback and education based upon the results.
Our experience is that physicians do not exhibit a permanent change in behavior when utilizing an education process that only occurs once or twice. We have been convinced that it usually requires an ongoing process that continuously educates.
In our experience, most physicians are haphazard in their coding accuracy. They routinely are both over coding and under coding. Also, documentation is often lacking in specificity or sufficiency until they receive continuous education.
We view this second audit as a comparison of training concepts . . . a test of less than constant audit process.
Initially, the first two physicians we reviewed were excellent making us wonder if, perhaps, our experience had been skewed. After the first two shining stars, the other physicians demonstrated poor to atrocious results. The group had enormous need for improvement in documentation and coding.
The audit of charts for this group revealed that the majority of physicians were not documenting sufficiently resulting in a high number of under codes. MARSI did not select charts using a statistically relevant random sample. For the most part, the physician group selected the charts so we cannot extrapolate results. However, the over and under codes, on a combined basis, occurred at an average rate that was in excess of two per chart.
Conclusion
The process of reviewing charts and using a system to query physicians provided the first physician group with the critical information that they needed to ensure coding was accurate and complete.
The methods that the first group used enabled continuous feedback with physicians as well as a mechanism for improving the physicians’ understanding of documentation requirements. Ultimately, they enjoyed a higher MRA score and both patients and physicians felt that a better level of care was achieved.
Since RADV audits only review for overcodes, the plans that contract with Group II are at great risk for substantial re-payments. Perhaps what may be worse are the under coded amounts due to this group.
Depending on the contract language, the physicians and/or the group may be at risk for re-payment. Certainly, if they are able to keep their contracts, future contracts will likely include documentation and coding accuracy provisions. Plans will also be compelled to require the performance and documentation of quality measures in their provider contracts
Educating and training physicians in the performance of quality measures, properly documenting the conditions identified and then accurately assigning and submitting the ICD-9 codes that describe those conditions remains at the heart of any risk adjusted program. For training to be successful it must be on going and repetitive and the physicians should have adequate incentives. This is a proven formula as evidenced again by these two groups of physicians both located within the same city.
