Thursday, January 27, 2011

Disseminating Taxpayer Funded Research: Problems at CMS

When your tax dollars fund research, do you have a right to see the results? Do the federal agencies that fund research have an obligation to taxpayers to see that findings from the research they support are disseminated and easily accessible to the public? And what would you think about a federal agency that takes active steps to obstruct the dissemination of research they support?

There is a very important post by Dr. Chris Langston on the John A. Hartford Foundation health AGEnda blog that addresses some of these issues. Over the years, CMS has spent large amounts of money on research and demonstration projects. Goals of these projects often focus on improving the quality and reducing the cost of care provided to Medicare recipients. These projects are often awarded as grants or contracts to academic researchers or contract research organization.

Dr. Langston tried to learn about the results of a project examining the impact of care management on high cost beneficiaries. Finding information about the project and the results proved to be a massive effort, requiring perseverance and the skills of a detective.

So, one concern is that CMS does not disseminate the results of its taxpayer funded research to the public in a manner that is useful. But, I am concerned that the problem may run much deeper than this.

Several times over the years, I have been told by people involved in CMS research that CMS often has obstructed the dissemination of its funded research. I have been told that some CMS contracts include provisions requiring CMS permission to publish findings from its funded grants and contracts. I have also heard several reports of CMS refusing to grant permission for researchers to publish findings. It is widely accepted that if a pharmaceutical company acted in such a manner to prevent dissemination of research, it would constitute an ethical breach.

While I have heard this allegation several times, I have never actually seen a CMS agreement for a grant or contract, so I have never verified that such contractual language exists. It has also been several years since I have heard these stories, so I do not know for sure if these practices are continuing.

However, on the CMS website, in the guidelines for authors preparing final reports to CMS, it notes that a "limited number of reports may be selected each year for publication by CMS."

Why a limited number? What is so hard about making all reports easily accessible on its website? Further, Dr. Langston's report of how difficult it is to learn about the results of a CMS demonstration is a sentinel story suggesting that CMS has not adopted a culture of openness and dissemination when it comes to its funded research.

CMS should do the following:

  • CMS should assure the public that no contracts or grants for research or demonstration projects will include any language abridging the obligation of researchers to disseminate findings to the public.
  • CMS should then take this a step further. Contracts and grants should include language noting that public dissemination of research findings is an essential obligation of those who receive a research grant or contract. All applications should include a dissemination plan, and the quality of the dissemination plan should be scored as part of the review criteria. Grantees should provide an assurance that they will disseminate their findings irrespective of whether the project or demonstration was effective, and irrespective of whether the results are favorable or unfavorable to CMS.
  • If it is true that previous contracts and grants included language obstructing dissemination, it should immediately make clear to grantees that such language is now null and void.
  • If it is true that CMS has previously acted to prevent, discourage, or obstruct the dissemination of its funded research, it should report to the public a list of all projects on which these actions occurred
  • CMS should have an easy to find section on its website that provides access to all final reports of its funded research and demonstrations. This section should include all past and future reports.
by: [ken covinsky]

2 comments:

Christopher Langston said...

Thanks for taking up this cause Ken.

I have heard directly from the parties involved, that results from one of the Medicare Health Support demonstrations were suppressed by CMS project staff. This would be results based on data from and available to one of the participating sites "i.e., contractors." Reasons why were unclear but mostly seemed to do with an abundance of caution on the part of project staff and the politics of offending either powerful business interests, higher CMS officials, or congress.

I had thought about complaining to congressional representatives at the time, but alowed myself to be dissuaded by those who thought it more important to get along.

In the cases I mentioned in my post it is rather different - those are CMS "grantees" RTI and MPR who produce reports for CMS the nature of which are carefully contracted, not like NIH grants at all.

In any event, getting the results of demonstrations is one thing, but I think an even more basic thing is getting the proposal/contract documents. While I believe you can see a lot of information on NIH proposals pretty easily through CRISP's successor, RePORTER in the NIH universe, I have never actually seen or seen how to find, the "winning" proposals for projects that CMS funds.

I believe this stems from the nature of government contracts but it clearly deprives us all of the ability to learn as we go along trying to redesign healthcare. It may be the way CMS has opperated, but it needs to change.

ken covinsky said...

Thanks Chris. I wonder if CMS proposal and contract documents would be available through a freedom of information act request. It would be useful to do a review of language regarding dissemination and restrictions on publication.

I think the central principle, whether CMS has contracted with a research company to produce a report, or whether they have funded an academic agency to conduct research is that the activities are funded by taxpayers and should be in the public benefit-particulary for the benefit of the patients CMS serves.

If it is a report, the public should have access to it. If it is a demonstration project, or analysis of Medicare data, the results should be freely available, and the researchers should be expected to publish the findings. I would take this a step further, and argue that the data should be made available in a public use dataset, for analysis and review by other researchers.

For CMS officials to decide what the public can be allowed to see on based on political considerations is ethically questionable and a breech of their obligations as public servants.