W. Ryan Neuhofel, DO, MPH via KevinMD / April 5, 2019
As a family physician in the trenches, I routinely see blatantly poor medical care in the history of my new patients. Far too many people get unwarranted medications and tests, while important things go unrecognized or unaddressed. This paradox is maddening.
We must do better.
To this end, policymakers have developed programs to promote “quality” and “performance” among clinicians. The main approach has been altering payments to serve as financial incentives and penalties with the intention of leading clinicians to make better, “value-based” decisions. This seems logical on the surface; pay more for good care, less for bad.
Here are the rubs:
Objectively measuring quality on a massive scale from afar is really, really hard (particularly without creating loads of administrative work).
More importantly, it’s a fundamentally flawed premise that misses what motivates us primary care physicians in our work.
Most medical professionals — me included — can agree on a specific example of medical care that is not in line with the standard of care. This is the basis for morbidity and mortality conferences. I routinely discuss cases of bad patient care with medical students I mentor. However, even an individual patient case can often be difficult to analyze without context or complete medical records.
Now, consider judging medical decisions en masse via millions of physician-generated codes. This is a wholly different, complex process. To date, most performance measures have focused on structure (e.g., percentage of clinicians using electronic health records) and process (e.g., number of cancer screenings completed). Despite a heavy investment of time and money, we now realize that most of these metrics have been misguided. A recent review of 86 Medicare “performance measures” in the New England Journal of Medicine found that just 37 percent were proved valid.
Thirty-seven percent? (Gulp.)
This is staggering. Nearly two-thirds of the boxes physicians have checked to demonstrate quality have not been validated with any degree of certainty. I shudder to think of how many hours physicians have wasted collecting this data throughout the past decade. Every minute took away from crucial time that could’ve been spent face-to-face with a patient.
Not to worry, though. To reduce the ever-expanding administrative burden, there are calls and concessions to simplify the gathering of quality metrics. CMS has even started using the mantra of “Patients Over Paperwork.” Although it’s great to see people recognizing this problem, slightly reducing unhelpful paperwork isn’t going to be a game-changer for primary care physicians (PCPs) or patients.
Also, in an attempt to take a broader view, measuring patient outcomes as a proxy of quality care has become fashionable. We see several examples of this in value-based payment programs already: incentive-based payments (the Physician Quality Reporting System and the Merit-based Incentive Payment System), capitation programs (the Comprehensive Primary Care program), and risk-bearing arrangements (accountable care organizations).
Intuitively, turning to these programs may seem wiser than counting smaller beans. A move away from fee-for-service payments might make sense at some levels. However, in many respects, basing payments on health outcomes is more complicated and fraught with potential hazards.
For starters, we know that 70 percent to 90 percent of health outcomes are determined by socioeconomic and lifestyle factors. Appropriate health care, particularly primary care, can tilt the odds toward better outcomes through medical intervention, lifestyle guidance, and advocacy. Primary care can improve individual lives and help budgets by reducing the likelihood of more expensive downstream care.
But remember that medical care accounts for that remaining 10 percent to 30 percent of health outcomes. The consequences of a patient’s genetics and environment — and how those factors figure into the development of chronic diseases — are decades in the making. In many circumstances, no amount of exceptional medical care can help a patient avoid bad outcomes.
Should we penalize physicians or systems that decide to take care of these complicated patients when things inevitably go south? Of course not. These are precisely the type of people who need great care the most.
To mitigate the potential for cherry-picking patients, we could make outcome-based payments dependent on patients’ risk factors, such as existing diseases and social factors, and pay higher rates for riskier patients and vice versa. You can imagine how complicated it would be to provide such benchmarks and scoring. If our goal is to reduce administrative burden, should we be introducing a new bunch of messy risk metrics for every patient? Also, the gaming of such risk scores by process (favoring intake over actual care) and paperwork (lots of it) is inevitable. Predictably, systems and clinicians will be geared toward maximizing codes to justify higher payments.
Regardless of how we attempt to measure quality, I ask: By what underlying principle do these incentive programs work? It would seem that the premise in all of them is to nudge health care professionals to personally do better through financial pressures.
I would be naïve to say that money doesn’t matter. On a systems scale, money certainly influences budgets and business decisions. Hospitals build palaces for cardiac intervention, yet relegate endocrinology to the basement. Medical students choose specialties in large part by income potential. Insurance networks are built around financial negotiations.
However, does a bit more — or less — money significantly change the behavior of individual physicians on a day-to-day basis? Particularly at the primary care level?
Being a family physician is a complicated blend of applying science and cultivating relationships. Our judgments are often made in gray areas within difficult circumstances. Even when a clear universal goal exists, for many patients and conditions, the best we can do is slowly, steadily move them to a better place. Atul Gawande brilliantly described this in “The Heroism of Incremental Care” in The New Yorker.
Although knowing what the standards of care are is certainly necessary, the physician-patient relationship is a huge driver in achieving those standards.
The vast majority of family physicians I know are capable and truly want to provide great care. They didn’t choose this specialty for a fat paycheck. Yet, clearly, too many of us fall short of these standards. Why? (I will leave out the fact that many Americans no longer have any sort of actual PCP.)
The list of things that hamstring PCPs from doing their best is long, but let’s start with some simple math. The average full-time PCP is now responsible for a panel of 1,200 to 3,000 patients. (There is some debate on how to calculate this number in a traditional fee-for-service clinic.) Physicians are often rushed through 15 to 30 office visits per day that last 15 to 20 minutes at most. Given the complexities of modern health care and the growing prevalence of chronic illness, this is insane.
A 2012 study in Annals of Family Medicine estimated that a “primary care physician would (need to) spend 21.7 hours per day to provide all recommended acute, chronic and preventive care for a panel of 2,500 patients.” Although not all PCPs are caring for 2,500 patients, by division, even 1,500 patients would require 13-hour workdays.
In addition to sheer patient volume, physicians are increasingly distracted by a mountain of clerical and other nonclinical tasks. We can always strive for efficiency and delegation of tasks (teamwork) when appropriate, but there is no substitute for a doctor’s time. It’s no wonder family physicians are often inaccessible to patients’ acute needs and feeling burned out.
In this context, I’ll ask again: Can financial incentives drive PCPs to do better?
Imagine this in another profession: If you gave an engineer four hours to design a complex suspension bridge, could he or she get a blueprint done? Sure, but would the quality of work be the same as it would if he or she took a more appropriate amount of time? Surely not. Would giving the rushed engineer a bonus for quality and safety improve that work?
In effect, this is what we are attempting to do with PCPs through pay-for-performance schemes. Some would argue that our mishmash of payers and provider organizations is what makes effective implementation of incentive programs so difficult. However, even in countries with tightly regulated, government-owned systems, such as the U.K., pay-for-performance hasn’t proved to be effective at the primary care level.
If you understand the puzzle of personal motivation — what moves us to be better at our tasks — none of this should come as a surprise. Although financial incentives can improve performance on repetitive, mechanical tasks (e.g., assembly line work), they are not what really push professionals of a cognitive nature (e.g., those engaged in complex problem-solving) to do better.
This isn’t just my opinion. The social science on this matter is clear and convincing. There are even a few voices in the health policy world who are coming to this realization.
Rather than apply more external pressure, we must recognize that intrinsic factors are what matter most to a PCP. We need an environment that fosters a sense of autonomy, mastery and purpose to fulfill our potential. The only way to create this culture is to allow PCPs the following opportunities:
- the opportunity to sit and truly listen to each patient to understand his or her story;
- the opportunity to educate patients and allow them to ask questions;
- the opportunity to develop long-term, trusting relationships with patients and staff;
- the opportunity to utilize (not just collect) relevant and coherent patient data;
- the opportunity to learn and grow in our knowledge and skills; and
- the opportunity to stay sane and happy while doing all of the above.
If a reform plan does not increase these opportunities, frankly, it is likely to be a waste of time and money. At worst, misguided pressures are driving my colleagues to anguish and retirement. So, please, policymakers, put down your carrots and sticks. They have not worked, and they will not work.
Give primary care physicians the opportunity to provide better care, and we will.