Clinical Peer Review and Resistance to Scrutiny

Tetsuya Tanimoto, M.D.

Medical Governance Research Institute, Tokyo, Japan

Maintaining quality in a diverse medical workforce

I am an internist, thirty years out of medical school, and help run a group of primary care clinics employing several dozen physicians. Their backgrounds and working arrangements vary considerably. Some have only recently graduated; others are serving university professors or senior clinicians who spent many years heading departments at major hospitals. Some work from morning until night, seeing more than a hundred patients a day. Others hold their main appointments at university hospitals and join us for one evening a week, or cover occasional shifts of just a few hours.

These physicians see large numbers of patients every day across several clinics. Most are experienced, and day-to-day operations generally run smoothly. One persistent concern, however, is the variation in how they approach clinical decisions. In a university hospital, morning conferences, ward rounds led by senior faculty, and case discussions routinely expose physicians to their colleagues’ reasoning. Seeing someone formulate a diagnosis or prescribe a treatment differently can prompt a reconsideration of one’s own approach. Busy primary care clinics offer fewer such opportunities. Once consultations begin, patients arrive continuously, and physicians work at different sites and at different times. Finding time for them to meet, discuss cases, and reach a consensus is difficult in practice.

This has led me to consider a more deliberate use of clinical peer review, in which physicians examine one another’s care.

Few of us enjoy having our practice scrutinized and commented on. Some experienced physicians are openly displeased: they do not want colleagues questioning how they treat their patients. I can understand the pride that comes with years of clinical work. But can an organization assure the quality of its care if it relies on each physician’s confidence in their own judgment?

Medicine changes remarkably quickly. Even for conditions we see every day, including diabetes, heart failure, chronic kidney disease, asthma, and dyslipidemia, what was accepted practice ten years ago may no longer hold. New drugs become available, diagnostic criteria are revised, and the place of particular tests in clinical assessment changes. Seniority does not necessarily imply familiarity with current practice. Equally, a younger physician may know the latest guidelines but still have much to learn from an experienced colleague about diagnostic judgment or communication with patients.

As members of a profession, physicians need arrangements that periodically expose their practice to colleagues’ scrutiny and allow them to correct one another. That is the purpose clinical peer review should serve.

Peer review and the risk of punitive use

The idea that physicians should assess one another’s work has a long history. Early accounts of colleagues examining a physician’s competence and conduct can be found in the writings of the ninth-century Islamic medical author al-Ruhawi. In modern medicine, a particularly influential development was the End Result System proposed by the American surgeon Ernest Codman in the early twentieth century.

The premise was straightforward: follow patients after treatment and, when outcomes are poor, investigate why. This may sound self-evident today. At a time when judgments of a physician’s ability often rested on personal authority or reputation, however, examining actual outcomes was a substantial departure. The approach subsequently informed morbidity and mortality conferences, clinical audit, and hospital quality assurance.

Peer review also has a troubled history. In the United States, from the 1970s onward, it became closely associated with assessments of the appropriateness of care and the management of physicians’ credentials. Although intended to protect patients, the process has sometimes been accused of arbitrary use to remove particular physicians, a practice described as “sham peer review.”

This history gives us reason to be cautious about linking routine peer review to performance appraisal or disciplinary action. If reviewers hold considerable power while the physician under review has no opportunity to explain or challenge their conclusions, physicians may conceal errors, avoid difficult patients, or order unnecessary tests. Candid discussion becomes difficult. A process introduced to protect patients can then undermine their safety.

Examining the conditions in which errors occur

Thinking about patient safety has changed substantially since the 1990s. Previously, when something went wrong, the first question was often who had made the mistake. Clear deficiencies in knowledge and inappropriate conduct cannot simply be overlooked. Yet attributing a problem solely to an individual’s inattention leaves the same failure liable to recur with another physician and another patient.

Consider a delay in responding to an abnormal test result. If the treating physician did not check the result, we might reasonably identify an oversight. But the inquiry should continue. How are abnormal results brought to the physician’s attention? Who checks them when that physician is off duty? Who follows up if the patient does not return? Is there a mechanism for flagging results of particular concern separately from routine reports?

The same applies to a missed diagnosis of serious disease. It is easy to conclude that the treating physician’s differential diagnosis was inadequate. We also need to ask whether the electronic medical record supported the task and whether procedures allowed investigations to proceed promptly and warning signs to be recognized. Clinical judgment depends in part on the organizational arrangements within which it is exercised.

A case review should therefore lead to a practical question: what needs to change to prevent the same problem in the next patient? Identifying a physician whose care was deficient is insufficient by itself.

Keeping review manageable in a busy clinic

An elaborate review process is unlikely to last in a busy practice. Spending thirty minutes or an hour discussing a single case is rarely feasible in a crowded outpatient clinic. I would begin with a small sample of cases for each physician.

For example, the sample might include one randomly selected case, one meeting a predefined criterion such as a return visit within seven days, and one drawn from a common monthly theme. The theme could be fever and antibiotic use one month, cough and asthma the next, then chest pain. Each review should be brief.

The scope should also be limited. Was the history adequate, and was the differential diagnosis reasonable? Were the investigations appropriately selected and interpreted? Did treatment depart substantially from current standards? Were follow-up arrangements and instructions about what to do if symptoms worsened adequate? Where the clinical record does not support a judgment, the reviewer should refrain from making one.

Numerical scores are unnecessary. I doubt that assigning physicians marks of 60, 80, or 90 out of 100 would improve their care. What matters is the action taken after a case has been reviewed.

Four broad categories seem sufficient to me: cases that raise no concerns; cases offering a learning point for the treating physician or the reviewer; cases suggesting that an organizational procedure should change; and cases requiring closer examination because of a patient safety concern.

Cases in the third category are particularly valuable. Variation in the management of cough may prompt revision of shared protocols for asthma and chronic cough. Frequent antibiotic prescribing may indicate a need to clarify the approach to febrile illness. If ultrasound has not been performed, we can consider whether simpler procedures or additional training would make it easier to use. Reviewing one physician’s care of a single patient may improve the subsequent practice of several dozen colleagues. This is where peer review can make a practical contribution.

What experienced and younger physicians can learn from one another

Some physicians may feel that thirty years in practice makes review unnecessary. I would argue that an occasional assessment by a colleague remains especially useful as experience accumulates.

The experience acquired over thirty years is something a younger physician cannot yet possess. An experienced clinician may notice an incongruity in a patient’s account or recognize something troubling in a facial expression. That kind of clinical perception cannot be acquired solely by reading guidelines. At the same time, some of what we learned thirty years ago will inevitably be outdated. Keeping up with every change in the use of new drugs, or with the withdrawal of recommendations for previously accepted treatments, is difficult amid the demands of clinical work.

Younger and more experienced physicians bring different strengths to a review. Both should be able to contribute what they know and learn from the other. A process in which senior doctors simply grade junior colleagues would miss much of this value. Exchanging knowledge and experience helps each participant recognize limitations they might otherwise overlook.

For a group operating several clinics, review across sites is also essential. Clinical conditions and approaches can differ considerably even within the same organization. A physician at another clinic can read a case as a colleague in the same field, with some distance from the working relationships at the originating site. “I would have added this test here.” “We have been using this drug differently recently.” Even a brief exchange of this kind can offer perspectives that are hard to obtain during routine consultations.

Building an organization that learns from its care

Risk cannot be eliminated from medicine. However many accomplished graduates of Japan’s leading medical schools an organization recruits, diagnoses will sometimes be delayed and clinical judgments will differ. This is especially true in frontline outpatient practice, where consultations continue from morning into the evening with little respite. An error deserves attention, but so does the failure to learn from it, which leaves the underlying weakness in place.

A good clinic or hospital cannot depend on having only infallible physicians of the kind portrayed in the Japanese television drama Doctor X. I have greater confidence in an organization that recognizes variation in care, examines cases that went badly, and uses what it learns to make successive improvements in how it works.

For that to happen, the boundary between peer review and personnel appraisal must be explicit. Review findings should not be linked directly to pay or performance evaluations. Serious safety concerns should be referred for a separate assessment by another physician or a responsible clinical leader, outside the routine process intended for learning. Physicians who disagree with a finding must have an opportunity to explain their reasoning. They need to be able to regard review as a way of drawing on their colleagues’ judgment to improve their own care.

The measure of success should change accordingly. We should count improvements in the delivery of care, rather than the number of physicians in whom a problem has been found. A standard consultation template has been introduced. A prescribing regimen has been updated. The procedure for following up test results has been revised. As these small changes accumulate, care across the organization becomes more consistent.

We undertake peer review because all of us, myself included, overlook things and make assumptions we fail to question. Sharing our experience and knowledge gives us a way to address those limitations. Criticism of colleagues is no end in itself.

As medicine continues to change, a culture in which physicians refuse scrutiny of their own practice will limit what an organization can learn. Each reviewed case should help improve the care of the next patient. The challenge is to make that a routine part of clinical work, without allowing review to become preoccupied with finding individual fault.

Clinical peer review is a modest undertaking. For primary care clinics whose physicians work across different locations and shifts, I believe it offers a practical means of maintaining the quality of care.