We’ve Seen Defensive Medicine, but It Was Never Taught

A 12-year-old boy arrives in the pediatric emergency department wincing with right lower quadrant pain. The team quickly gets his vitals, draws blood, and gets a bedside abdominal ultrasound. He has what is expected: uncomplicated appendicitis. He is admitted and put on the pediatric surgery OR schedule for the next morning.

Yet, this patient’s parents are particularly worried. Understandable, for any parent. They ask the team to confirm the diagnosis with an abdominal CT scan.

The boy’s diagnosis is clear, but the team orders the CT anyways. The family will get the several thousand-dollar bill for the CT scan in a month. As expected, the scan does not change the plan.

Cases like this are common as we rotate on our medical school clerkships. As medical students seeking to understand the whys of clinical reasoning, we ask our residents and attendings the indications for these scans. The answer is often not that the diagnosis was uncertain or that the test would change how we manage the patient; it’s about protection. Even when it’s not stated explicitly, the implication is clear: physicians are protecting themselves from lawsuits.

Reflecting upon these moments as future physicians, we realize that medicine is practiced for two audiences. One is the patient before us. The other is the hypothetical lawyer who will mercilessly depose us. This is defensive medicine.

Hardly a new concept, defensive medicine emerged in a big way after the 1970s malpractice crisis, which saw skyrocketing claims, jury awards, and eventually, insurance premiums. While estimates of the clinical and economic burdens of defensive medicine vary, one 2014 estimate in JAMA Internal Medicine found 28% of orders and 13% of costs across the hospital system at a large medical center were at least partially defensive.

As medical students, we spend our early mornings and evenings studying management algorithms of evidence-based practice published by the premier medical societies. Yet, another syllabus takes shape during daily clinical practice, a facet of medicine’s “hidden curriculum.”

We see defensive medicine practiced routinely as emergency medicine doctors order extra imaging (“positive” defensive medicine) or cardiac surgeons refer patients to palliative care instead of operating because perioperative risk scores are slightly too high (“negative” defensive medicine). Physicians navigate a system where the consequences of missing a diagnosis can be severe, while the consequences of one additional test are usually minimal to none. And these decisions sometimes come in conflict with the algorithms we memorize.

This clash between our two parallel curricula has a tangible impact. We recall a vignette from a recent clinical reasoning session; a patient presented similarly to the patient we opened this essay with. As is characteristic for this stage of our training, the question asked for the “next best step.” We both said, “order CT,” as we had observed. The correct answer was “admit for surgery.” We realized then that we had been studying two different syllabi: one on rotations, one in our books. The discordance between what we see and what we read creates hesitation as we approach patient care.

The emergency medicine physician who ordered the CT was not practicing in bad faith. It is human nature to seek to minimize personal risk, even when that risk is low. If anything, that choice to order the CT showed an inherent understanding of the complex environment that physicians inhabit. Behavioral economists call this concept “anticipated regret” — in this case, the impulse to avoid the emotional burden of wishing you had ordered one more test. Long before a potential claim is filed in court, physicians must contend with anticipated regret. Defensive medicine is therefore more than just a reaction to the legal system — it is our nature.

Many might argue that in cases of “positive” defensive medicine, they would want the additional scan. After all, what is the harm of one more test? But these choices can negatively impact both patients and hospitals. For patients, unnecessary testing can expose them to excess radiation, lead to false-positive results and incidentalomas, more procedures, prolonged emergency department stays, higher bills, and a profound mental toll. Hospitals already struggling with financial viability due to Medicaid cuts and rising costs must embrace slower workflows, lower-value care, and offer patients longer wait times.

Hospitals have continued to recognize the importance of high value care interventions as potential tools to curb this behavior. One of us (Diamond-Pott) is a former leader of a hospital high value care committee. The other (Midha) is a former management consultant who advised health system leaders. We have seen good outcomes from these efforts: better stewardship of IV antibiotics to limit unnecessary exposure and development of multidrug resistant infections; more proactive discharge planning to limit hospital length of stay to what is medically necessary, thereby reducing nosocomial infection and delirium; much lower “left-without-being-seen” rates in emergency departments.

But we’ve also seen their limitations. A hospital can recommend restraint and flag low value orders in the electronic health record, but the organization itself does not bear the same personal risk as the clinician ordering the test. The incentives are asymmetric.

Defensive medicine is not just the way physicians practice. It is a pervasive hidden curriculum whereby students struggle to tell when a choice is evidence or liability-based. We don’t have the solution to this problem that has become a mainstay of medical culture. After all, it may be a consequence of human nature. But we hope that by raising the conversation early in medical education and training — initially through informal discussions, and potentially later through formal revisions to testing formats — students and trainees will be better prepared to know when and how to apply this facet of practice.