In the 1971 film adaptation of Roald Dahl’s classic and disturbing book, Charlie and the Chocolate Factory , the spoiled child Veruca Salt is sent down an “Eggdicator” chute, followed shortly thereafter by her indulgent father. The diagnosis for both: they were “bad eggs.”

There are bad eggs in medicine, too: doctors who perpetrate fraud, who exploit patient vulnerabilities, who commit other misdeeds. In June of this year 301 individuals, including doctors, were charged with >$900 million in Medicare fraud. Fortunately, bad eggs like these, who bring such dishonor to our profession, are rare. Far more common–almost universal, in my experience–are good eggs. And although I am unaware of even a single doctor who has never made a significant mistake, most physicians I know try to do the right things for the right reasons.

But it turns out that doing the right things for the right reasons is not as straightforward as it seems. Judson and colleagues have highlighted the tensions generated by the interplay of intrinsic and extrinsic motivations in medicine. Even for a physician with the purest intrinsic motivations, extrinsic pressures can influence medical decision-making in ways that do not always benefit patients. Take a common example from the specialty of obstetrics and gynecology: a patient in labor with no sign of jeopardy to either mother or fetus. Most US obstetricians agree that in this scenario vaginal birth is better than cesarean delivery. Why? Simply put, cesarean delivery is not likely to improve outcomes for the baby but exposes the mother to the risk of complications in both the current and future pregnancies. Some of these complications can be life-threatening.

So it must be pretty simple, right? Obstetricians should advocate for vaginal birth whenever it is safe to do so. But consider the incentives: a cesarean takes about 45 minutes and can be scheduled at my convenience. In contrast, labor and vaginal delivery often require spending many hours at the patient’s bedside, canceling office hours, inconveniencing other patients, sacrificing sleep and personal time. Remarkably, despite this discrepancy in effort, my hospital’s billing office informs me that reimbursement for cesarean is about 15% higher than for vaginal birth. And perhaps the most powerful incentive of all: it is rare for an obstetrician to be sued for the decision to perform a cesarean, even an unjustified cesarean. In contrast, it is not rare for the obstetrician to be sued and found liable for not performing a cesarean delivery, even one from which neither the baby nor the mother would have derived any benefit.

These factors affect physician judgment by artificially lowering the threshold for concerns about fetal well-being or labor progress, the most common reasons for cesarean delivery performed during labor. With the incentives arrayed so dramatically against vaginal birth, is it surprising that the cesarean rate in the United States, currently about one third of deliveries, is double what it ought to be ? In terms of population health we have no benefit to show for incurring this excess risk.

Thus our system fails patients by turning their good-egg doctors a little bit into bad eggs. Fixing the system would force both patients and doctors to make tough choices. Americans could relinquish the notion that private obstetricians should care for low-risk pregnant women, thus joining most of the rest of the world in relying on well-trained and well-supported midwives and shift-workers who are relatively free from competing interests. We could change our priorities for reimbursement. We could decrease physician exposure to nonmeritorious lawsuits to diminish defensive medical practices that are not in patients’ interests.

Such systemic opportunities for improvement are not confined to obstetrics but span all medical specialties. Consider fee-for-service reimbursement, which incentivizes us to do more things, not to get better results. Consider continued pressure from device and drug manufacturers who, even without questionable inducements such as meals, samples, and travel, are biased providers of information upon which many physicians rely. Consider the pressure to maximize utilization of the facilities we have built and the equipment we have purchased.

To compound the problem, there are squandered opportunities to leverage extrinsic motivation to patients’ benefit. It surprises me that in the procedure-oriented specialty of obstetrics and gynecology, annual Maintenance of Certification requires literature review and demonstration of evidence-based management of select common outpatient problems, but there is no attempt at verifying that complex skills presumably acquired during residency have been maintained, or that newer techniques have been mastered. Such due diligence is left to the discretion of local departments of obstetrics and gynecology, for which there is no universal or minimal standard.

These considerations increase in importance when we consider the fact that patients are to a significant extent in the dark about quality, and thus are not in a position to advocate for themselves. The thing about bad eggs is that until you crack them open you can’t tell that they stink. I find that patients’ judgments regarding their doctors are driven by bedside manner, a trait that has only a modest direct correlation with clinical quality. Don’t get me wrong: patient perceptions of compassion and attentiveness are unquestionably important. Trust, open communication, and the patient experience are significant determinants of quality in medicine and have been associated with improved outcomes. But a good bedside manner might also be a shell that hides serious deficiencies. I have seen some patients heap praise upon their doctors for engaging in worthless testing and treatment, and even for jeopardizing their well-being. I have seen others criticize their doctors for failing to avert unavoidable outcomes of nature or for sparing them from serious morbidity. Viewed through the prism of bedside manner, the line separating remedy from harm can get blurred. The public needs more than that to be assured that the system is working in their favor.

We are headed in the right direction. The coming pivot from fee-for-service toward value-based reimbursement strategies will help. The publication of outcomes data, already underway and set for expansion in the near future, is similarly promising, although assuring that such measures truly reflect upon quality in a sufficiently comprehensive manner will be challenging. Thought leaders are seeking ways of applying principles learned from the field of economics to motivate toward favorable behaviors, but such interventions have not yet been well tested in medicine.

In the meantime we need to do our best to continue aligning incentives with best practices in ways that perpetuate the noble traits of benevolence, compassion, advocacy, and caring that we so value in our physicians. This will require effort and willingness to change. Both physicians and patients might have to sacrifice cherished practices for which there is insufficient evidence of benefit. Let us not forget the lesson of Veruca Salt, who belted out these triumphant words just prior to tumbling down the Eggdicator: “I want the world, I want the whole world…I want it now!”