Meaningless they are, and utilized everyday by payers nevertheless.

You’re splitting hairs - it’s a proxy of course.

‘Those with sicker patients will have poorer ratings, and as a result, more and more sick patients will have fewer and fewer options, as these patients become a liability. They will inevitably lower a doctor’s “quality score” because of the inherent complexity, and quality measures that are contraindicated.’

Quality measures that don’t take into account the characteristics of the patient population are pretty meaningless. Take them into account and they have meaning.

First, “Number of procedures done” is not a quality indicator. It is a statistical predictor of outcomes, which ARE quality indicators. Yes it is an important factor, but it is not a quality indicator. Read the report. Saying that “number of procedures done” is a quality indicator is analgous to saying that the amount of Windex purchased is an indicator of cleanliness.

Second, I have no problem with assessing the quality of hospitals on ANY of these factors. What I have a problem is is PENALIZING hospitals by reducing an already insufficient reimbursement based SOLELY on indicators that are not directly correlated to outcomes.

When I go to the hospital, I want to have an uncomoplicated course. Good communication, cleanliness, are nice, and as a consumer I might make choices based on these, assuming outcomes are the same. But the clean hospital with excellent communication skills might just be the one most likely to fail to provide adequate care, resulting in my death. So I will choose clinical outcomes over cleanliness and communications anyday.

Of course, clinical outcomes measures are HARD to make. Much harder than suveying patients and asking them if they thought their doctor or nurse were nice, or if the floors were mopped adequately. Much harder than counting the number of pneumococcal vaccine injections given to those with unrelated conditions.

All of the cognitive work that goes into determining clinical outcomes is ignored by these measures. It is ineveitable that if someting like this is imposed on hospitals, practioners are next. I am already judged as a practitioner, not based on outcomes, or cognitive skill, but on the percentage of patients receiving pneumovax, receiving statins, recieving flu vax, receiving mammograms and DREs. No quality measurement is made of my cognitive skill: how well sick patients are managed, or those with complic ated problems are managed.

It will soon by that the “best” doctors are those with loads of healthy patients and a good system for ensuring that all of the “quality measures” are met. Those with sicker patients will have poorer ratings, and as a result, more and more sick patients will have fewer and fewer options, as these patients become a liability. They will inevitably lower a doctor’s “quality score” because of the inherent complexity, and quality measures that are contraindicated. Dotors will favor healthier patients to get better quality scores, supply will decrease, and access with it.

I think you need to do some reading on this - surgeons who do only a few of a certain procedure often have worse outcomes than those who do an ‘optimum’ number. It is an important factor. And as I said, mammography quality has been hugely variable, and has benefited from guidelines and policing, including federal sanctions.

You dismissed the result of poor cleanliness earlier. As for doctor and nurse communications, this is the subject of increasing training and research - check out the field of pyscho-oncology.

“Number of procedures done” is not a quality indicator. Uncomplicated hip replacements are, by definition, free of infection. Sure there are factors, but they are not “quality indicators” which is the crux of this post, and the basis for the incentives/disincentives in the proposed program.

The question remains - what are the “quality indicators” for routine, uncomplicated care, that makes up the bulk of what hospitals do? READ THE PROPOSAL (http://www.cms.hhs.gov/AcuteInpatientPPS/downloads/HospitalVBPPlanRTCFINALSUBMITTED2007.pdf) - it is cleanliness, nurse communication, doctor communication, etc. Actual clinical measures are proportion of patient who received pneumovax, or MI discharges who were prescribed aspirin, and the like. Again, they are a narrow set of measures, which will improve everywhere, while those things that are not measured will deteriorate.

Surely you must realise there are all sorts of factors, not least the number of procedures performed by surgeons and the resultant rate of infections and revisions. To take a really obvious example, what about mammography? There has been a big drive in the US to raise quality standards.

Not at all. Perhaps I am just ill-informed. If you could please educate me, what are the “quality indicators” for uncomplicated hip replacement?

What are the quality indicators for uncomplicated treatments of medical problems, such as pneumonia (otehr than “time-to-first-antibiotic”), diabetic wound care, idiopathic pleural effusion, diverticulitis, etc.?

I think you’re skating round the issue. It’s in fact easier to put in place quality standards for uncomplicated, routine care. While the population treated will of course be variable - age. socioeconomic, other conditions - in different locations, a hip replacement is still a hip replacement and confounding factors can be controlled for.

Most patients do not have operative errors or hospital acquired infections. Is there research that demonstrates the abundance of “extortionate extras” or their impact on the patients’ outcomes? Is the gold standard for care “no errors and no hospital acquired infections?”

The bulk of patients admitted with pneumononia recover and are discharged home. How is the quality of their care measured? Fortunately, many people admitted with ACS do well, and are stabilized and return home after a short stay. How do you measure the “performance” of the hospital in those cases? How is one program distinguished from another in uncomplicated, common cases of hospital care? How is performance measured in the case of the diabetic admitted for a foot amputaion, who has an uncomplicated post-operative course?

It is not easy. We have reduced performance measurement in acute coronary care to the percentages of aspirin and beta-blockers prescribed, ann the rate of treatment complications. Do you really believe that that tells the whole story? Is it so hard to believe that a hospital could have good outcomes, but poor measures, such as these? Or that a hospital might have great measures, but poor outcomes?

Of course they can. It is very hard, and often subjective. What is not, subjective, however, is the number of dollars.

One big problem witgh performance measures is that there are variables outside of the providers control, which will alter the behavior of a provider. High risk patients will be viewed as high risk, and their care will be deferred because there is no possible outcome that will not harm the providers “performance measures.”