How perverse incentives are ruining healthcare

For seven years, I practiced at a New Hampshire-based internal medicine practice affiliated with a local hospital. Before that, I was in solo practice in Maine. Today, I am no longer a practicing physician.

Dissatisfaction and burnout among primary care providers-a function of larger and sicker patient panels, inadequate appointment times, more burdensome and tedious administrative tasks, and poor compensation-have been well documented.  The rewards of the doctor patient relationship and the intellectual stimulation of clinical problem-solving were enough to keep some of us in the primary care trenches.  But it was a close call.

Related: Predicting the future of MU 3

With the introduction of CMS's unethical incentive and penalty programs, most notably the Physician Quality Reporting System (PQRS) and Meaningful Use (MU), an unpleasant practice environment was now an unacceptable one for me.

PQRS

PQRS is a deeply flawed program, as the following examples involving primary care benchmarks illustrate.

Under PQRS guidelines, it is assumed that progressive lowering of average blood glucose in Type 2 Diabetes Mellitus (Type II DM) results in progressive improvement in long-term outcomes.  Average blood glucose over a three-month time period is measured by A1C percentage.  People without diabetes have A1C values in the 4.5% to 6% range.  One PQRS benchmark measures the percentage of a physician’s diabetes patients whose A1C values are below 7.0%, thereby incentivizing aggressive use of medication in this population.

Next: Contradictions

However, recent studies have contradicted the assumption that achieving A1C values below 7% is beneficial-and in fact provide evidence that doing so causes harm. Lowering average blood glucose to this range is associated with an increased risk of all-cause mortality, cardiovascular mortality, and episodes of dangerously low blood glucose (hypoglycemic episodes), according to recent studies.  

Also, there appears to be no reduction in risk of stroke, vision loss, nerve damage, kidney damage, or limb amputation. So when treating a patient with diabetes, the physician must decide between attempting to lower A1C below 7.0% as incentivized by PQRS, or doing what is in the patient’s best interest by taking a more conservative approach.  Meanwhile, patients-unaware of these conflicting goals-assume that their health is the only consideration influencing the physician’s treatment recommendations.

Best EHRs: Physician reviewed

Continuing along the path of obsolete guidelines, we encounter PQRS benchmarks regarding “target” low density lipoprotein (LDL) cholesterol in the diabetic population. Here, CMS wants us to show that a sufficient percentage of the physician’s diabetic patients have LDL cholesterol values of 100 milligrams per deciliter (mg/dL) or less.

Based on evidence compiled after CMS introduced this PQRS “quality” measure, we know that attaining LDL values below 100 mg/dL-or any other target value-is unimportant. Rather, the evidence suggests that nearly all patients with diabetes should be prescribed a relatively high dose of medication from the statin class of cholesterol-lowering agents, regardless of their baseline LDL cholesterol. 

Top article: 15 challenges facing physicians this year

Once on a statin medication, no follow-up LDL measurement is warranted, as the heart disease and stroke-lowering benefits of statin medications accrue in part from factors other than their LDL-lowering properties, according to studies by the American College of Cardiology and the American Heart Association. Yet not only does PQRS incentivize attaining the specific goal of 100 mg/dL or less, it penalizes for not ordering lab testing of LDL cholesterol in this population, even though following the evidence-based guidelines obviates the need for such testing.

Next: MU turns physician-patient encounter into scripted experience

Lest one think that misguided PQRS incentives are limited only to the (very large) diabetic patient population, there are plenty of examples involving other conditions.  One example involves breast cancer screening with mammography.  The assumption is that high screening rates are an indicator of quality care.  Yet recent evidence casts serious doubt on the benefits of screening mammography, particularly in light of the emotional toll incurred by high rates of false positive mammographic findings. So, does the physician encourage as many women as possible to have regular mammograms, as dictated by the PQRS incentive?  Or does the physician engage patients in a dialogue regarding the pros and cons of  mammography, thereby risking penalization under PQRS? 

The secrets of financial flexibility

The notion that PQRS promotes quality care rests on the flawed assumption that a bloated federal bureaucracy, CMS, can be nimble enough to produce quality measures that keep up with the rapidly evolving state of scientific evidence.  Reality contradicts that assumption. Published reviews in JAMA have determined that “most guidelines become outdated after 5 years”-a finding made all too clear in the examples cited above.

Meaningful Use (MU)

MU has turned the physician-patient encounter into a mechanistic and scripted experience that takes a one-size-fits-all approach to patient care, leaving the physician with little time or discretion to address patients’ actual health needs and concerns.

Under MU, the nightmare begins at check-in, when patients are handed pages of forms covering a wide spectrum of demographic and other issues, including questions about race, religion, and ethnicity, as well as questions about family history, smoking status, and whether the patient feels safe at home.  In many cases, these same questions are asked at each visit  and add little or nothing to what the physician already knows about that patient.

Next: MU’s one-size-fits-all model embraces a “dumb it down” approach

Frequently, after the physician enters the exam room, the patient is still filling out forms and unable to fully engage. That lack of engagement cuts both ways, as MU dictates that the physician stares at a computer screen, checking off boxes, ensuring that the EHR is being used in a “meaningful” manner.

How millennials are affecting medicine

Another MU rule requires that physicians or physician practices establish that they are implementing "clinical support rules" to show that physicians are not just using EHRs to record data, but that they are using such data to "meaningfully" improve quality of care.  To meet this requirement, one local practice used spirometry testing on cigarette smokers who scored poorly on a vague symptom questionnaire.  Yet there is no evidence in the literature to suggest that spirometry data gleaned from-and shared with-this population translates into higher smoking cessation rates, decreased chronic obstructive pulmonary disease incidence, or any other favorable endpoints.

To make matters worse, patients and/or their insurance companies are charged for the spirometry testing. So patients are being instructed to undergo useless testing-and incur additional charges-by physicians who have a conflict of interest-all in the absence of informed consent.

Improved patient education is another purported benefit of MU.  In practice, this entails printing handouts containing facile summaries of a given condition from MU-approved proprietary software.  Content is usually vague and/or not supported by evidence.

Because of the “patient education” form requirement, there is little opportunity for searching, printing, and discussing more meaningful, up-to-date information from the Internet.  Some patients prefer to be presented with the complexities and nuances of various topics, while others choose to keep it simple.

Patient education should be tailored to the individual patient. Unfortunately, MU’s one-size-fits-all model embraces a “dumb it down” approach.

PQRS and MU incentivize poor quality care, leaving physicians with a conflict of interest of which their patients are unaware.  It is a shameful situation that warrants outrage on the part of physicians and policy makers.

Peter C. Cook, MD, MPH, spent most of his career in solo practice in Bangor, Maine, before moving to a hospital-affiliated internal medicine practice in Rochester, New Hampshire. He attended the University of Medicine and Dentistry of New Jersey and completed his residency at the Maine Medical Center. Cook has a master's degree in public health from Harvard University.