My most recent blog explained why three prominent Dr. Bobs—Bob Wachter, Bob Berenson, and Bob Centor—are raising important questions about the value of physician performance measures, which, ironically, are supposed to bring greater value to the healthcare system. Their concerns include:
· The proliferation of measures of dubious validity and the associated burden of reporting on them;
· The difficulty, even impossibility, of measuring elements of care, like physician compassion, that patients may value the most;
· The risk of unintended consequences, like treating to the measure and disadvantaging physicians who treat the underserved.
All of these are very real and genuine concerns, shared widely by physicians, and not just those fortunate enough to be named Bob.
The question I am left with, though, is 'What’s the solution to physicians’ anxiety about performance measures?' For practical reasons, I don’t see the country saying, never mind, measuring performance is an impossible or ill-advised task, let’s just repeal the new value-based payment system created by the Medicare Access and CHIP Reauthorization Act (MACRA), let’s abandon all efforts to link payments to performance. Nor do I think it would be advisable for the medical profession to declare open warfare against performance measures. For one thing, there are known gaps in quality and effectiveness of care, and measurement can be one way to identify areas for improvement and progress in achieving it. Performance measurement can also help policymakers, the public, and physicians assess the impact of new delivery and payment models. (To be clear, the Dr. Bobs weren’t completely calling for an end to performance measurement, but a better approach to measuring performance).
In my opinion, Dr. Yul Ejnes, a former chair of the ACP Board of Regents, offers a wise and appropriate balance in his commentary in today’s @KevinMD blog:
I have a "like-hate” relationship with clinical metrics, performance measurements, and other such things. By now, almost all physicians live with them in the form of insurer “report cards,” PQRS, and “meaningful” use. Some of us have even more exposure to them by participating in patient-centered medical homes and accountable care organizations.
Why “like”? Because I believe they can help you to know how you’re doing. Happy patients, full schedules, phones ringing off the hook with new patient requests, and the belief that you’re doing a good job delivering care aren’t enough. Few things are more sobering than seeing data on the percentage of your diabetes patients who are not at goal, those with hypertension whose pressures are not under control, or those who haven’t undergone colon cancer screening. I know that many question the relevance of some of the clinical measures, which often look at intermediate and perhaps less meaningful outcomes or report on process, but they can be more informative than the gut sense that we have on how we’re doing our jobs.
What I hate about the measures is that they also get in the way of patient care. I’ve written about this in earlier columns — concern about measures distracting us from our primary jobs, the administrative burdens of “capturing” data so that it can be counted, and the effect of all of this on the quality of clinical documentation are some examples. Add to this the open question of whether the current version of performance measures truly improves quality of care.
Instead of rejecting performance measurement, Dr. Ejnes calls for 'a more thoughtful approach to measurement and for EHR products to make [them] more seamless and less intrusive. The American College of Physicians’ (ACP) 2012 paper on performance measurement had many recommendations that, if followed, would prevent much of our current suffering. For example, minimizing burdens in collecting data, using EHRs to facilitate (not complicate) the process, and most importantly, that “performance measures that have not been shown to improve value to include higher quality, better outcomes, and reduced costs (and higher patient and physician satisfaction) should be removed from performance–based payment programs.”'
did exactly what the good Dr. Ejnes ordered, and what our 2012 paper (which is
still right on the mark!), calls for, which is to call
on CMS to “reimagine” performance measurement—starting
with scrapping the measures currently used for Medicare’
existing Meaningful Use, PQRS, and Value-Modifier programs:
"The College strongly recommends that CMS actively work to improve the measures to be used in the quality performance category of MIPS. Therefore, we believe that CMS should NOT consider the existing quality measure sets within Physician Quality Reporting System (PQRS), Value-Based Payment Modifier (VBM), and Meaningful Use (MU) as the starting point for its measure development plan.
In the short term, ACP recommends that CMS utilize the core set of quality measures identified and recently released by the America’s Health Insurance Plans (AHIP) collaborative . . . Further, the College recommends that CMS consider the recommendations made by ACP’s Performance Measurement Committee with regard to measure selection within MIPS.
Over the longer term, ACP stresses that it will be critically important for CMS to continue to improve the measures and reporting systems to be used in MIPS to ensure that they measure the right things; move toward clinical outcomes, patient- and family centeredness measures, care coordination measures, and measures of population health and prevention; and do not create unintended adverse consequences.”
Further, ACP identified the need to “constantly monitor the evolving measurement system to identify and mitigate any potential unintended consequences, such as increasing clinician burden and burn-out, adversely impacting underserved populations and the clinicians that care for them, and diverting attention disproportionately toward the things being measured to the neglect of other critically important areas that cannot be directly measured (e.g., empathy, humanity).”
In other words, the solution to physician anxiety over performance measures is for the medical profession to work with policymakers to fix performance measurement so that we are measuring the right things without creating unintended consequences and without increasing clinician burden, not to do away with them. As ACP first articulated in its 2012 paper and reaffirmed in yesterday’s letter to CMS, the medical profession must insist that measures be:
- Reliable, valid, and based on sound scientific evidence
- Clearly defined
- Based on up-to date, accurate data
- Adjusted for variations in case mix, severity, and risk
- Based on adequate sample size to be representative
- Selected based on where there has been strong consensus among stakeholders and predictive of overall quality performance
- Reflective of processes of care that physicians and other clinicians can influence or impact
- Constructed to result in minimal or no unintended harmful consequences (e.g., adversely affect access to care)
- As least burdensome as possible
- Related to clinical conditions prioritized to have the greatest impact on improving patient health
Or, as Dr. Ejnes so aptly put it, “It was supposed to be that if we provided high-quality care to our patients, the measurements would reflect that. Instead, the mantra is that if we score well on our measures, then that means that we provided high-quality care. In other words, the cart has become the horse. It’s time to fix that.”
Today’s question: How would you fix performance measurement?