CMS provides quarterly benchmarks to ACO's.  These benchmarks are used to determine the beneficiary costs for the ACO, essentially the bottom line as to whether the ACO is generating savings.  Clinigence users with access to the claims pivot tables can generate similar but not exactly the same numbers.  CMS recommends that ACOs use the monthly CCLF data to identify areas for savings and to look for trends but not to use it as a source of truth as there are a number of shortcomings in the data they provide.  

The following is a list of inconsistencies in the data that CMS uses in their calculations and what CMS provides to the ACO:

  • CMS numbers are generated off of all attributed patients but they only send the data for patients who have gone through the claims data opt-out and request process.  The impact of this will vary wildly depending on the percentage of attributed patients for whom data is being shared.  The best practice to manage this differential is to stay on top of the list of attributed beneficiaries and move them through the process as quickly as possible.
  • CMS includes substance abuse costs but does not share this data with the ACO's.  The impact of this will make CMS' numbers higher and is estimated to be between 1% and 5% of the total claims costs.
  • Prior to 2015 CMS did not provide the beneficiary enrollment status.  This status is used to calculate "patient years".  For example if a patient was only enrolled for 6 months of the year they will only be counted as one half of a patient year.  The impact of this will make CMS' numbers higher and is estimated to be between 2% and 6% of the total claims costs.
  • CMS may include some data in their calculations prior to sending it to the ACO.  The data CMS sends is about 45 days behind.  For example, data sent in mid-April has a most of the claims data through the end of February.  However, other February data will continue to trickle in many months later.  Eventually this data is used by both Clinigence and CMS but the timing is not 100% aligned.