Performance measurement has major problems

3

Category : Medical Rants

I started writing about performance measurement and associated problems for 10 years. For the first few years, we in the blogosphere seemed to be shouting in the wilderness.

When I first joined the ACP Board of Regents, the general consensus favored P4P. We who questioned the value of performance measures were told that “the train has left the station”.

Over the past 7 years, many leaders in medicine have seen what the blogs saw first, we have too many bad measures, and too often performance measurement has significant unintended consequences.

Now even MedPac has major concerns about the proliferation of performance measures.

Over the past few years the Commission has become increasingly concerned that Medicare’s current quality measurement approach is becoming “over-built,” and is relying on too many clinical process measures that are, at best, weakly correlated with health outcomes. Depending on a large number of process measures reinforces undesirable payment incentives in FFS Medicare to increase the volume of services and is overly burdensome on providers to report, while yielding limited information to support clinical improvement or beneficiary choice. Instead the Commission has urged more focused attention on a small number of population-level outcome measures, such as potentially avoidable hospital admissions, emergency department visits, and
readmissions.

When MedPac complains to CMS we must have reached a “tipping point”. The current approach to performance measures is actually harmful. The ACP Performance Measurement Committee is endorsing approximately 20% of proposed measures. Most measures have inadequate data supporting them. Too many proposed measures read like expert opinions.

Bravo to MedPac for making this letter public. Now we can only hope that CMS will listen.

Comments (3)

Music to my ears. I have been part of the vox clamatis in deserto crowd, long aware that the quality metrics are selected for reportability rather than causal relation to quality, that systems are easily gamed, that the numbers do not allow legitimate application at the clinician or even practice level but need to be population based, and this approach runs the risk of replacing a quality culture with a market culture.

I am not counting on CMS listening or understanding.

I’m somewhat confused by the ACP’s position on P4P.

It seems like they’re saying it’s valid when the performance measurements are more carefully designed and are supported by the data.

But isn’t there a wealth of literature showing, as you said, “P4P does not have a data driven rationale,” it does not improve care, and that the whole concept should be abandoned?

Thanks

And now the ACP has come out with a strong endorsement of HHS’s plans to tie physician payment more closely to “quality measures.”

Post a comment