Earlier this week, the GAO released a new report that looks closely at Section 1115 Medicaid demonstration evaluations. The states examined are Arizona, Arkansas, California, Indiana, Kansas, Maryland, Massachusetts, and New York. The report was requested by Republican Members of Congress from the committees of jurisdiction.

The report uncovered a number of problems including that final evaluations are often not done and/or released and a waiver is extended anyway. Also, the GAO notes that a federally-commissioned evaluation that is being conducted by Mathematica has 15 quick response briefs that have been completed over the past few years but have not been publicly released.

I find the report’s findings troubling but not surprising. It has been clear for some time that evaluations of Section 1115 waivers are not adequate. There is some good work going on in this space at the state level, for example in Michigan and Iowa, but as the report makes clear, state’s evaluations are often incomplete and not rigorous enough.

The GAO notes that the state of Indiana did not release its data to a federally-commissioned evaluator citing concerns about protecting the privacy of the individual data. I don’t buy that excuse for not handing over the data. The Indiana example cited by GAO underscores the high stakes politics of waiver evaluation. The state clearly wanted to stymie efforts by the Obama Administration to have an independent evaluation conducted by not turning over its data. I very much doubt the current Administration will ensure that the public ever sees an independent, rigorous evaluation of the Indiana model.

Moreover, it is very troubling that many reports are not being publicly released after a sizable contract has been awarded. More sunshine and data are needed to assess Medicaid waivers — especially as they are clearly the vehicle the Trump Administration is now using to pursue its ideological objectives for Medicaid.