A group of about a dozen state hospital associations are fighting to make sure that a value-based incentive is included in any health reform proposal. The attached statement is their attempt at defining how value should be rewarded and our comments.
It seems like everyone is getting into the act to define value. This group of state hospital associations include Iowa, Maine, Minnesota, Montana, Oregon, South Dakota and Washington. The coalition is now defining how "value" should be rewarded.There are two proposals in this recommendation one is that hospitals get paid more if they are in an HHR or Hospital Referral Region that is more cost efficient. HHRs are reported by the Dartmouth Atlas, and are determined by geographic regions not by individual hospital performance.
I blogged a couple weeks ago on the first recommendation but reiterate that;
"The fact that a region or state has lower quality or higher cost doesn't
mean that any individual provider is of low value. So if you have a high
quality, efficient doctor in a particular region, how do you justify cutting
his/her payment because all the other doctors in the region are, on average,
delivering lower value than the providers in other regions? This doesn't
work well in markets where there are multiple providers -- you want to
differentially reward those providers, not reward or penalize each of them
based on their collective performance.
A better approach is to reward each hospital by it's HSA (hospital service area performance as reported by the Dartmouth atlas). This is specific to each hospital in a market and measures total medicare reimbursement per enrollee for the market(see my blog on this)
The second recommendation by the coalition is a payment system based on the clinical indicators each hospital reports to Medicare. Of course these are not outcomes measures but process indicators. We may be able to do better than that. USA Today reported http://www.usatoday.com/news/health/2009-07-09-baylor-heart_N.htm on a study that CMS just released which took a look at clinical performance data on hospitals including mortality and hospital readmission rates. These outcomes are what we should be focused on improving and paying for.The CMS study is reported on "hospital compare" at http://www.cms.hhs.gov/HospitalQualityInits/11_HospitalCompare.asp
Finally, the problem with using Dartmouth data is the most recent data is from 2006.If we are going to move in this direction we must have more timely data. My suggestion is six months updates with re- calculation of performance payments based on these updates.
