- Statistical sampling determines error rates and overpayment demands
- Court cases support statistical sampling for calculating overpayments
- It is a bad strategy to focus on the claims where errors were identified
- Focus should be on validity and confidence level of sampling methodology used
The HHS OIG depends on statistical sampling to gather the necessary supporting documentation for determining the error rates and penalty levels for providers. Other oversight bodies also use statistical sampling that results in overpayment demands, including Medicare/Medicaid contractors and State Medicaid Fraud Control Units. The method used in applying the appropriate statistical formulas to calculate estimated overpayments vary based on different risk factors. Numerous administrative appeal decisions and Federal court cases have concluded that statistical sampling is an appropriate way to calculate overpayment. They have also upheld the right of providers to appeal the process when the statistical sampling method employed was flawed or inaccurate for some reason. When employing statistical sampling as the basis for assigning overpayments and penalties, the process should meet four standards: (a) statistically valid, (b) efficiency of effort, (c) being representative of the larger group, and (d) producing a valid estimate of any overpayment. Noteworthy also is that the OIG will mandate a sampling level of claims for review by an Independent Review Organization in their Corporate Integrity Agreements.
When moving to defend against the level of overpayments in a demand letter, it is a mistake to focus on reviewing the same claims where errors were identified. That is not a good use of resources. It is important to understand that government agencies and their contractors don’t always adhere to proper statistical sampling methodology, and it is not uncommon for there to be mistakes in calculation. Those assigned to do the work may not be the best qualified or may be sloppy in taking shortcuts in drawing the sample. Those making the demands must maintain documentation and data related to every sample so that it can be reproduced in evidence.
The first step in assessing the true exposure to overpayments is to engage a credible independent statistical expert to determine if the sampling methodology employed produced a high confidence level in the validity and reliability of results. When confidence levels in the sample drop, the overpayment credibility estimates plummet. If flaws are found, then the next step is to have the expert take a RAT-STAT sample (the method adopted by the OIG) and, from that cell sample, take a smaller “probe sample.” This probe sample will give you a clear idea of error rates but will not be statistically valid for projection to the universe. This is important as it is not disclosable, whereas a representative sample of the universe would be disclosable. It will provide a reasonably reliable estimate of the real error rates and, if lower than that in the demand letter, will provide the basis for challenging their result. At that point, the balance of the cell can be reviewed for a final exact error rate amount. All of this provides the best possible position to properly resolve the issue. On the other hand, if the result indicates a higher level of errors, then moving to a settlement is the answer.
A final note is that it is not advisable to act with the demanding party based upon the first step without doing the follow-up statistical analysis. If this is done, the agency will be embarrassed, defensive, and angry. They will likely bring out their “A-Team” to redo the review with a no- mercy attitude. However, if their errors are identified along with the actual error rates being presented for settlement, it is far more likely they will accept a half loaf than be forced to admit their mistakes.
Keep up-to-date with Strategic Management Services by following us on LinkedIn.Subscribe to blog