EDITOR’S NOTE: The following is the first article in a five-piece series titled “Dirty Tricks,” covering abuse in audits and statistical extrapolation in the healthcare industry.
I’ve been helping providers fight against statistical extrapolations since 2001. It has not been a pleasant experience. In almost all cases, the auditor, a Zone Program Integrity Contractor (ZPIC), failed to meet the disclosure requirements of the Program Integrity Manual. Yet, in a number of cases, multiple violations have been ignored by the presiding administrative law judge.
Anyone who has been audited is familiar with statistical extrapolation. The auditor takes a sample of a few patients, or claims, usually 30 or so, and then reviews them to see if they are valid. This review is done in the strictest possible manner. After all, the audit contractor is going to get a percentage of the take. Imagine what would happen if every highway patrolman in your state was awarded 15 percent of the fine of every speeding ticket they wrote. The number of tickets would skyrocket!
If 75 percent of the claims in a sample are rejected, the auditor will ask that the provider return 75 percent of the money collected over a number of years. That’s the extrapolation.
In other words, the relatively small value of the rejected claims in the sample is ballooned into a gigantic number. A few thousand dollars in mistakes becomes millions in penalties.
The ground-rules for statistical audits are found in the Program Integrity Manual (PIM). It requires statistical work to be done the right way. After all, statistics is a science. There are accepted methodologies. There are procedures that must be followed. And when these procedures are not followed, the statistics are no good. This means that the extrapolation should be discarded. But it is not always easy to defeat an extrapolation. There is a presumption in favor of the contractor. Overcoming it means that the provider must take a very careful look at the contractor's statistical work and uncover evidence to show how it is invalid.
Providers should know that they have rights: the right to examine the statistical work, for example, and to verify that it was done fairly and accurately. But to do this, the provider needs two things: First, it needs to have a complete record of the statistical work; second, it needs to employ a credible scientific expert who really understands statistics, and who can see through the often amateurish and deceptive work of the contractor.
Lets divide the statistical issue into three phases. First is the discovery phase. That’s when you try to collect all of the documentation about what the contractor did. It often is not easy to see what was done. Second is the analytical phase. That’s when you look at the auditor’s statistical work and determine if the methodology used was sound. This is where you need a certified statistical expert. We use only statisticians who are elected fellows of the American Statistical Association, but for simple cases, someone with a Ph.D. in statistics from a good school should be able to do the job. Third is the presentation phase. This is when you make arguments to the administrative law judge. They must be convinced that the statistical work was so bad that the extrapolation must be discarded. There often are dirty tricks afoot in all phases.
Problems begin in the discovery phase. The provider and its expert need to get a complete picture of what was done by the contractor, but they usually face a number of obstacles designed to keep them from seeing what was done.
Dirty Trick No. 1: Don’t explain anything at all.
The contractor just says that there has been a statistical extrapolation and that a certain amount is owed. There are few real details. But the language in the letter is frightening. It all seems so official, and there may be scientific-sounding words such as “confidence interval.” But when you start to probe for details, there are none to be found.
Here it is important to understand a very important legal principle. The contractor must actually show their work, not merely attest to having done it. Imagine a trial in a courtroom. Your defendant is charged with robbing a bank. The prosecutor says they have collected your defendant's fingerprints on the safe. You ask to see the proof. The prosecutor says "I told you we found the defendant's fingerprints. There is no need for you to actually see them." How would that go?
The same happens with many audits! The contractor writes a lengthy demand letter, about 99 percent of it boilerplate, complete with detailed citations from the Code of Federal Regulations. They describe an audit and the statistical work. They say there was a random sample, they say there is a certain error rate, they say there is a valid extrapolation, but they never show it to you.
And if you demand a number of details, the auditor goes through a number of maneuvers to hide their work. For example, in some cases, providers have had to hire an attorney and use the Freedom of Information Act to request information on how the statistical work was done. Otherwise, the contractor would stiff-arm their requests. Of course, this adds a lot of time to the process. The provider sees its legal bills piling up and if their payments are being impounded, it generates even more financial pressure.
I’ve also seen cases when a contractor was asked details about how they did their calculations and they responded that it was a commercial trade secret. Considering that the provider is supposed to have the absolute right to verify the statistical work, this answer reached the very heights of arrogance and duplicity.
Dirty Trick No. 2: Make it impossible to analyze the information provided.
Another favorite trick is to pretend to provide complete disclosure of the statistical work, but to do it in a way in which critical information is hidden. Then make it extraordinarily expensive for the provider to analyze what happened. For example, the contractor can provide everything on paper, preferably faxed over and almost illegible. If there are printouts of spreadsheets, the column headings will be missing and much of the information will be illegible. This makes it virtually impossible to analyze the statistics unless you are willing to spend the money and hire a data entry service to manually recreate the electronic spreadsheet from which the list was printed.
We have gone through this before. It is expensive to do the data entry because all of the data must be re-checked. And then, after the spreadsheet is completed, there is no possibility of seeing even a single calculation. There are no formulas, no calculations. In short, there is still no way to verify the statistical work.
The result is that your statistical expert will not be able to determine much. So all of your expenses in getting this far will have led you nowhere.
There is an important legal principle involved here. It is called the "best evidence rule" and it means that an original of a document is superior evidence. If the original is available, then a secondary copy will not be accepted.
So the dirty trick here is to not provide an actual record of what was done on an electronic spreadsheet, but instead to print out an excerpt and reduce it to paper so that much of the process is hidden. In fact, from a statistical point of view, all of the vital information will be hidden.
Dirty Trick No. 3: Delay every request for information.
This is one of the favorite dirty tricks. I can almost guarantee you that if there is a 120-day waiting period for production of information, the contractor will wait until 4:45 p.m. on the 120th day before sending you the materials. This will happen even if the contractor has the information on hand from the very beginning, and even if the amount of time it would take them to complete your request is only five minutes.
This delay and obfuscation can do great harm to the provider. First, it will dramatically restrict the amount of time the provider (or their statistical expert) has to consider the information. Less time means that your expert will need to work faster, and there is an increased chance that they will miss something. Second, it will narrow the provider's options in case what the contractor provides is incomplete. And it is almost always incomplete. In fact, in my 15 years of work on these cases, not even once have I ever seen a complete production by the contractor of the details of how they did their statistical work.
So the next step will be to go to your expert and have them consider what was provided by the contractor. Then they will come back to you saying that much important information still is missing. Then your attorney will need to go back to the contractor and demand the missing information. This means more expert and legal bills to pay, and still you are no closer to seeing what was actually done by the contractor.
This behavior on the part of the contractor fits the standard profile: delay, obscure, and reluctantly provide information that turns out to be incomplete. Don't make the mistake of thinking these actions are not being undertaken on purpose. They are, and they work. The number of providers that simply give up in their appeals is increasing. They get worn down by this type of tactics.
There are many more dirty tricks. This is only the beginning. In the second article in this series we will go into even more detail about how sneaky the contractors can be. Not only do they try to hide data, but many times they simply manufacture data out of thin air. It is a sad story, but one that needs to be told. See you next time.
About the Author
Edward M. Roche is the founder of Barraclough NY LLC, a litigation support firm that helps healthcare providers fight against statistical extrapolations.
Contact the Author
Comment on this Article