What, Us Measure?

By Kim Bellard, May 18, 2011

I read with some interest an article in the Wall Street Journal outlining Wellpoint’s new approach to hospital payments, in which future payment increases would be tied to performance on 51 quality measures.  As interesting as that is in itself, what fascinated me even more was the reaction of Chip Kahn, the President of the Federation of American Hospitals.  Mr. Kahn did not appear too excited about the approach.  As he told the Journal: “We don’t have good outcomes measures yet.”

I don’t mean to take a shot at Chip Kahn or the Federation.  They are strong advocates for their members.  To its credit, the Federation has actively been involved in quality efforts for many years, including the Hospital Quality Alliance.  But, seriously – we pay hospitals some $800 billion per year, and we don’t have good outcome measures yet?  What are we waiting for?

A recent study by the Beryl Institute, whose mission is to improve the patient experience, indicated that 31% of hospital executives listed quality/patient safely as their organization’s top priority over the next three years, followed by 21% who cited patient experience/satisfaction.  The Beryl Institute seems to view this as good news, and I suppose it is good that no other single priority topped these two, but I have to wonder: so, almost half of respondents did think something else was their top priority?  

“Adverse events” – injuries caused by medical errors -- have been one of the open secrets in the health care system, and a recent study indicated that the problem may be as much as ten times worse than thought – impacting up to one third of hospitalized patients.   With those kinds of problems, if I ran a hospital, I might not want to track outcomes either, or at least not report them.

Wellpoint isn’t alone in targeting hospital quality.  Many health plans have implemented some version of pay-for-performance, although they tend to be paid as incentives rather than core to reimbursement.  Last month CMS announced the final rules for its Value Based Purchasing Program, which begins in 2012.  Medicare will also pay based on performance, initially as incentives but rapidly moving to reductions for hospitals that do not perform well.  The CMS measures are oriented towards process and patient satisfaction indicators, whereas 55% of Wellpoint’s measures are based on health outcomes, 35% on patient safety measures, and 10% on patient satisfaction.

Of course, quality measures are not just a problem for hospital performance.  Hospital quality measures are more evolved than physician quality measures, and both are far ahead of other parts of the health care system.  As Kenneth Kizer, founder of the National Quality Forum (NQF), said: “There are many areas of medicine where there simply are no measures – or there are, but they aren’t as good as they should be.” 

There certainly is no shortage of organizations working on the issue.  In addition to NQF, one could cite The Leapfrog Group, NCQA, CMS, the Physician Consortium for Performance Improvement, and several medical specialty associations, among others.   Still, the sense from the provider community seems to be that we’re not quite there, certainly not to the point where consumers should be making judgments based on the various indicators, nor having reimbursements materially impacted by performance on the measures.  They like the measures to be voluntary, although the recent CMS report on the Physician Reporting Quality System indicate only one in five health care professionals who are eligible to participate actually do – and only slightly more than half of those earned satisfactory scores which merited bonuses. 

It’s too bad, because it appears that the “standard” measures that consumers tend to look at to evaluate physicians, such as education or board certification, don’t appear to actually distinguish quality performance very well, according to a study by Rachel Reid and colleagues.  As the authors concluded:  “Few characteristics of individual physicians were associated with higher performance on measures of quality, and observed associations were small in magnitude. Publicly available characteristics of individual physicians are poor proxies for performance on clinical quality measures.”

Similarly, a study by Lauren Nicholas and colleagues published last fall indicated that the “process” measures currently reported by CMS on its Hospital Compare website don’t correlate with actual patient outcomes, such as mortality rates or surgical complications. 

How in the world did we get to the point of spending so much money on health care without even being able to measure if we’re doing it well?

I’ve complained about the lack of data in health care in previous blogs (such as in Gambling on Health Care), but it still disturbs me.  Maybe when – or if – we get to a world of electronic medical records and fully realized health information exchange we’ll have a better job of getting the right data; that is the point of “meaningful use.”  Still, I go back to another quote from Kenneth Kizer: “Nothing makes a performance measure better than when it starts being used.”  We need to start using the existing measures more now.

Putting performance data out there is half the battle, and I applaud the many people and organizations working to make that happen, hopefully sooner rather than later.  The other half of the battle – and one that, frankly, worries me even more -- is getting people to use it.

Comments

[B Messner]

Is anyone aware of what the actual 51 quality indicators are?

[Doug Meil]

The NQF website is: http://www.qualityforum.org/Home.aspx