MRD Inaccuracies Poised to Mislead Health Plan Shoppers

MRD_inaccuracies_mislead_hp_shoppers-vericred.com

 

A few weeks ago, I shared an analysis we developed at Vericred that explored the accuracy with which hospital websites disclose the health plans where they participate in-network. The results were very revealing, so we decided to extend that analysis to examine the quality of Machine Readable Directories (“MRDs”) mandated by CMS for plans sold through Healthcare.gov. MRDs are used on healthcare.gov to enable shopping for a health plan based on the doctor(s) or hospital(s) you want to keep.

 

We learned that MRDs are not much better than hospital websites as a source of whether or not a hospital is truly in-network.

 

We analyzed the MRDs from the four carriers offering Individual under 65 (ACA or “Obamacare”) plans around Columbus Ohio on healthcare.gov.  We examined the accuracy with which each carrier included, or excluded, the 29 hospitals in their respective MRDs. As the source of truth, we again used each carrier’s consumer facing provider directory.

 

As one of the carrier’s MRDs was noted as “Medicare” we excluded that carrier’s MRD from the analysis. Even after removing one of the four carriers, the overall results were as bad, if not worse, than the hospital websites.

 

In the aggregate, the three remaining MRDs were incorrect approximately 24% of the time, either including a hospital as in-network when it was not, or vice versa.  No MRD was perfect, with error rates ranging from 14% to 41%.

 

Just under 5% of the time, these MRDs incorrectly excluded in-network hospitals; a false negative that could result in an individual selecting an alternative plan with their hospital, at perhaps a higher premium.

 

mrd machine readable directories error rates

 

More disconcerting was that nearly 20% of the time, the MRDs showed a hospital as accepting a plan when in fact they did not, creating a false positive. In this case someone shopping for a plan may enroll with the expectation that their local, or preferred, hospital will be in-network when in fact it is not.  Those anticipating a hospital stay, for the delivery of a baby for example, may not be able to use their chosen hospital despite doing their diligence.

 

The conclusion is the same as in my last analysis.  Before enrolling in a plan based on a particular provider choice, make sure you check with your prospective carrier.  And even then, there is no guarantee that your provider will remain in network throughout the year, so be sure to double-check again before going to the hospital.