The Ingenix settlement – you wanted details…

The phone and email has been buzzing today. So has the blog-o-sphere, at least among those bloggers who follow this. Both of us.
Today’s follow up announcement by Ingenix’ parent UHC revealed the giant health plan will pay $350 million to settle a class action lawsuit directly related to the use of the Ingenix UCR database. This brings the total (to date) cost for legal settlements to $400 million after yesterday’s NY settlement. Here’s the key language from UHC’s statement today.
“UnitedHealth Group (NYSE: UNH) announced today that it has reached an agreement to settle class action litigation related to reimbursement for out-of-network medical services. The agreement resolves class action litigation filed on behalf of the American Medical Association (AMA), health plan members, health care providers and state medical societies.
Under the terms of the proposed nationwide settlement, UnitedHealth Group and its affiliated entities will be released from claims relating to its out-of-network reimbursement policies from March 15, 1994, through the date of final court approval of the settlement. UnitedHealth Group will pay a total of $350 million to fund the settlement for health plan members and out-of-network providers in connection with out-of-network procedures performed since 1994. The agreement contains no admission of wrongdoing.”
The real problems with the Ingenix UCR database weren’t the subject of much discussion in the settlement documents or the various analyses of the settlement. But underlying the case are some significant problems with the database, how it is put together, and its uses. These issues were highlighted in the Davekos case in Massachusetts, and are discussed in the court record. Among the problems are:
– the accuracy and consistency of the underlying data is questionable. Ingenix cannot assure that the entities (health plans and claims administrators and insurance companies) that supply the data that Ingenix uses to determine what usual customary and reasonable charges are in specific areas are all using the same criteria, are coding consistently and accurately, and are aggregating the data in the same way.
– Ingenix may not always have enough charge data to provide a statistically valid sample for every CPT code. In that case, it appears that Ingenix may aggregate data from similar codes to produce a large enough sample. The potential issue with this work-around is obvious. In some instances, Ingenix actually called medical providers in specific areas where it did not have enough data to ask what they would charge for specific procedures. Thus they were combining telephonic survey data with actual charge data in their UCR databases, a commingling of very different data from very different sources with varying reliability.
– Ingenix itself defines the sociodemographic region boundary lines that are used to determine which charges are relevant for which geographic areas. In the Davekos case, the court appeared to be concerned when Ingenix could not give a defensible rationale for the logic or process they used to determine the boundaries for charge areas.
– Ingenix scrubs the data to extract charges that they decide are outliers for reasons that appear to be subjective. It also appears Ingenix removes high end values but not low end outliers. If this is the case, it would likely skew the charge data towards the lower end.
Those are some of the details behind the Ingenix UCR settlements. As to what will happen next, Bob Laszewski’s perspective provides insights as only he can.
What does this mean for you?
If you are using the Ingenix UCR database, you may want to look for other options.