Thursday, November 7, 2013

Quality: Illusions in Health Care


Quality was a key element in the ACA and when it was being bantered about I expressed significant concern. Mainly because it is illusive. No one can define or measure quality. I wrote extensively about it in mid 2009 as the ACA was raising its evil head. I then detailed it in my Health Care book in 2010. Thus when I see articles now regarding the ACA I am not surprised about anything.

One paper in NEJM states[1]:

... the Affordable Care Act (ACA) created the “value-based payment modifier,” or “value modifier,” a pay-for-performance approach for physicians who actively participate in Medicare. By 2017, physicians will be rewarded or penalized on the basis of the relative calculated value of the care they provide to Medicare beneficiaries….



The value modifier is meant to provide differential payment to a physician or physician group under the Medicare Physician Fee Schedule on the basis of the quality of care furnished as compared with the cost; it will result in a reward or penalty amounting to 1 to 2% of payments for groups of 100 physicians or more in 2015 and for all physicians by January 1, 2017. CMS anticipates increasing the percentage of payments at risk as positive experience accrues. To reduce the burden on physicians, CMS has based the value modifier on the Physician Quality Reporting System (PQRS).

As usual the problem is the term quality, or worse the value of care. Quality is the term which drove the main character in Zen and the Art of Motorcycle Maintenance stark raving mad. Value on the other hand should have a quantitative measure, but in the context of the CMS one will wonder. But here we have a confluence of Quality and Value, as in the last sentence, thus the creation of a true schizophrenia.

The author then states:

The challenge of accurately assigning costs to an individual physician is similarly daunting. Current methods for case-mix adjustment do not adequately capture variations in patients' illness severity, complicating coexisting conditions, or relevant socioeconomic differences — differences beyond the physician's control that affect the cost of care. And we currently don't know how to attribute to an individual physician the costs that Medicare beneficiaries generate across the health care system.

The observation is spot on. As the authors state in their opening:

the practical reality is that the Centers for Medicare and Medicaid Services (CMS), despite heroic efforts, cannot accurately measure any physician's overall value, now or in the foreseeable future. Instead of helping to establish a central role for performance measurement in holding providers more accountable for the care they provide and in informing quality- and safety-improvement projects, this policy overreach could undermine the quest for higher-value health care. Yet the medical profession has been remarkably quiet as this flawed approach proceeds.

That is absolutely correct. Not only cannot they measure an individual physician’s value they cannot do so with quality, and even more so.

In contrast there was a second NEJM article praising this program[2]. They praise the ACA and CMS and its efforts as demonstrated by the chart below from the article.

(Note: The above is from this article and demonstrates "Medicare's Quality-Incentive Programs Leading up to Hospital Value-Based Purchasing, as Compared with Those before the Launch of the Physician Value-Based Payment Modifier." )  

They state:

The PVBM reward formula is a simple, relative system in which performance is assessed in two dimensions (quality and cost), with payments accruing to physicians who have above-average performance along both dimensions. Physicians who perform worse than average or choose not to be involved will be paid less; physicians with average performance will experience no change. The maximum bonus is about 2% of Medicare fees, and the maximum penalty is approximately 1%. For CMS, scoring physicians relative to one another achieves budget neutrality. For physicians, it eliminates the effects of common shocks to performance, such as an influenza epidemic or vaccine shortage. The key disadvantage of this incentive structure is the inherent uncertainty for physicians about the amount of improvement that will be necessary to receive a bonus or avoid a penalty.

How does one make such a comparison? For example, an Internist and an upscale area versus a poor community GP, what is there to compare.

They continue:

The lack of experience with physician-level measurement and reporting has important implications for the PVBM. First, far greater numbers of physicians will need to become engaged in reporting of quality and cost performance. This challenge should not be underestimated: there are nearly 150 times as many physicians who bill Medicare as there are hospitals, the physician population includes physicians of all types (primary medical, surgical, and subspecialists), and many of these physicians work in a wide array of smaller practices that are still acquiring the basic infrastructure (e.g., health information technology) or organizational affiliations (e.g., independent practice associations) needed to measure and improve the quality and cost of care.

We have asked physicians to expand record keeping, billing, Electronic Health Care Records, and no we demand they measure quality. Re-read Zen and the Art of Motorcycle Maintenance, quality measurement and definition will drive one insane! What will happen are more costs related to gaming the system.

One area that could improve health care would be balancing patient expectations with outcomes. All too often the patient’s dissatisfaction is based upon an outcome that they were not expecting because they were not informed. Questionnaires which measure patient psychographics as well as other factors and them assist them in matching treatments, holistically, may help. But that is delimited by not including family expectations. For example for a man with prostate cancer there may be several options. How would we best treat that patient? It would be critical to understand his psychographic profile. That all too often is not done. Then on a post basis we verify that the proper choice is made and we can create a good Bayesian methodology to maximize patient satisfaction, namely they had no surprises.

Thus quality and value are just more terms that confuse rather than enlighten. To paraphrase Osler, “If all else fails listen to the patient.”