It’s tempting to believe that physicians are logical, meticulous thinkers who perfectly weigh the pros and cons of treatment options, acting as unbiased surrogates for their patients.
In reality, this is often far from the case. Bias, which takes many forms, affects how doctors think and the treatment decisions they make.
Racial biases in treatment decisions by physicians are well documented. One study found that black patients were significantly less likely than white patients to receive pain medication in the emergency department, despite reporting similar levels of pain. Other research suggests that longstanding racial biases among providers might have contributed to racial differences in patient trust in the health system.
But a growing body of scientific research on physician decision-making shows that doctors exhibit other biases as well — cognitive ones — that influence the way they think and treat patients. These biases lead doctors to make the same mistakes as the rest of us, but usually at a greater cost.
Cognitive biases refer to a range of systematic errors in human decision-making stemming from the tendency to use mental shortcuts.
Prominent examples include confirmation bias, the tendency to interpret new information in a way favorable to one’s preconceptions; and anchoring, the tendency to overly weight an initial piece of information, even when order does not matter. Anchoring helps explain why if you see a car priced at $20,000 and a second car priced at $8,000, you might conclude the second car is cheap, whereas if the first car cost $3,000 you might conclude that the second car is expensive.
In health care, such unconscious biases can lead to disparate treatment of patients and can affect whether similar patients live or die.
Sometimes these cognitive biases are simple overreactions to recent events, what psychologists term availability bias. One study found that when patients experienced an unlikely adverse side effect of a drug, their doctor was less likely to order that same drug for the next patient whose condition might call for it, even though the efficacy and appropriateness of the drug had not changed.
A similar study found that when mothers giving birth experienced an adverse event, their obstetrician was more likely to switch delivery modes for the next patient (C-section vs. vaginal delivery), regardless of the appropriateness for that next patient. This cognitive bias resulted in both higher spending and worse outcomes.
Doctor biases don’t affect treatment decisions alone; they can shape the profession as a whole. A recent study analyzed gender bias in surgeon referrals and found that when the patient of a female surgeon dies, the physician who made the referral to that surgeon sends fewer patients to all female surgeons in the future. The study found no such decline in referrals for male surgeons after a patient death.
This list of biases is far from exhaustive, and though they may be disconcerting, uncovering new systematic mistakes is critical for improving clinical practice.
In a new study of physician treatment decisions, published on Thursday in The New England Journal of Medicine, we document signs of left-digit bias. This is the bias that explains why many goods are priced at $4.99 instead of $5, as consumers’ minds round down to the left-most digit of $4.
We hypothesized that doctors may be overly sensitive to the left-most digit of a patient’s age when recommending treatment, and indeed, in cardiac surgery they appear to be. When comparing patients who had a heart attack in the weeks leading up to their 80th birthdays with those who’d recently had an 80th birthday, we found that physicians were significantly less likely to perform a coronary artery bypass surgery for the “older” patients. The doctors might have perceived them to be “in their 80s” rather than “in their 70s.” This behavior seems to have translated into meaningful differences for patients. The slightly younger patients, more likely to undergo surgery, were less likely to die within 30 days.
Our study confirms previous work that found doctors are overly responsive to patient age when diagnosing illness, and that showed how seemingly irrelevant factors‚ such as the difference of a few weeks of age, could govern physicians’ decisions about treatment, with potentially life-altering consequences for patients.
Left-digit bias could affect many clinical decisions. For example, patients with hemoglobin levels of 9.9 milligrams per deciliter may be perceived as being substantially more anemic than patients with hemoglobin levels of 10.0 milligrams per deciliter (the difference in the two values has no clinical significance).
Awareness of these cognitive biases has prompted efforts to reduce them in clinical decision-making. One trial studied the effect on general practitioner diagnostic accuracy of a computer program that provided “nudges” in the right direction and highlighted particularly relevant information. The trial of these so-called decision support systems, while small, found that they improved diagnosis at a relatively low cost. Another, larger study found that changing an electronic medical record’s default options for opioid prescribing — an example of so-called choice architecture — nudged physicians to prescribe fewer of the drugs.
Tools such as these have the potential to overcome some, but not all, of the cognitive biases that commonly plague clinical decision-making. Given our growing understanding of the errors that doctors can make, these biases are too costly to ignore.
Anupam B. Jena is a professor of health care policy and medicine at Harvard. Follow him on Twitter at @AnupamBJena. Andrew R. Olenski is a graduate student in economics at Columbia. Follow him on Twitter at @andrewolenski.