In previous blogs we have explained the importance of using Statistical Extrapolation to win your RAC or MIC appeals.
Here are the top five things to avoid in order to overturn the statistical extrapolation.
DO NOT Base your extrapolation argument on a sample size that is too small
Although it is true that in most cases, the sample size used is too small, this argument alone generally will not prevail in the appeal.
- What often happens is that the trier of fact will simply verify if the evidence that the statistical methodology used was generally consistent with the Medicare Program Integrity Manual (MPIM).
- There is an unfortunate passage in the MPIM which states that any sample size can be valid as long as the underlying methodology is sound.
And the unfortunate result is that the widespread use of poor and inadequate samples in the extrapolation often leads to results not in your favor.
DO NOT Accept print-outs instead of real electronic spreadsheets
It’s common for contractors to send print-outs with tables of data to back up their statistical work. Don’t accept them for the extrapolation.
- Try to get the original electronic spreadsheet because this allows you to verify the statistical work, and to see the details of how it was done.
- It also will allow you to run various tests to determine the quality of the work.
Always use the “best business record” rule to argue for the original sheets.
DO NOT Neglect to obtain the full universe file
Audit Contractors generally select a “frame” from the larger universe of all of your claims. From this “frame”, they will select their sample for analysis.
- Without the entire universe file containing all claims you have filed, it’s impossible for you to run several important statistical tests that will help uncover any flaws in the contractor’s statistical work.
DO NOT Accept the argument that poor precision works in your favor for the extrapolation
Poor precision in statistics is a sign of bad work. It is often argued that poor precision works in favor of the provider because the lower side of the confidence interval is chosen for the over payment demand number.
- Using a confidence interval to adjust for poor precision is not good statistics practice.
- Insist that the contractor meet the Federal precision standards that were published in the Federal Register.
DO NOT Ignore all aspects of the extrapolation’s statistical methodology
The statistical work is much more than simply the sample taking and calculation of the extrapolation. ALL of it has to be checked.
- In addition, there are a number of PIM rules that must be complied with. Check what was done against each and every one of the PIM rules and that they were followed in every detail.
Breaking a single PIM rule is often is overlooked, but if there are a substantial number of violations, then it helps build the case that the statistical work is not to be trusted and the appeal will be decided in your favor.