When bias affects your medical decision-making

Despite their best intentions, physicians can sometimes fall victim to cognitive bias. Such errors in thinking have the potential to alter not only the course of patient care, but also the dissemination of scientific knowledge when bias affects editors’ selection of which articles get published in medical journals.

To avoid cognitive bias, there are strategies clinicians can utilize such as guided reflection and cognitive forcing.

How cognitive bias works

Systematic errors in thinking give rise to cognitive biases. Cognitive bias happens when intuitive—or “fast”—thinking takes the place of analytical thought, according to research published in the Journal of the Royal College of Physicians of Edinburgh.^[]

Heuristics, or mental shortcuts, rely on only a few indicators to make decisions. Analytic thought is used less in everyday life because it takes too much time.

Humans usually perform daily tasks while engaged in intuitive thinking, a practice that could contribute to cognitive bias. According to the Journal of the Royal College of Physicians of Edinburgh, more than 100 types of bias exist, including:

• Anchoring bias—Implicit reference to the first data point

• Attribution bias—Looking for reasons for what was observed

• Confirmation bias—Showing preference for data that confirms one’s beliefs

• False-consensus bias—Overestimating the number of people who agree with us

• Framing bias—Preferring information based on whether it’s presented in a positive or negative light

• Not-invented-here bias—Bias vs external knowledge

• Search-satisficing bias—Believing that current knowledge is complete and sufficient

Cognitive bias in action

For instance, although serum HLA antibodies were known to impact cardiac transplantation outcomes beginning in the 1970s, it wasn’t until the mid-2000s that HLA antibody monitoring after cardiac or renal transplantation made it into the guidelines. The authors of the Journal of the Royal College of Physicians of Edinburgh article attributed this delay to confirmation bias.

Other examples of cognitive bias in transplant medicine included long-held misconceptions involving inflammation, antibody-mediated rejection (AMR), and endothelial activation. A detailed description of this bias at play was published in JACC: Basic to Translational Sciences.^[]

Dating back to 1989, studies demonstrated the adverse role of endothelial injury and antibody binding in heart transplant patients. Subsequent evidence from series published through 2000 established that multiple episodes of AMR predicted cardiovascular-related death following heart transplantation. Nevertheless, bias interfered with taking action.

“Framing of the information in a skeptical light by experts emphasized controversy rather than the scientific facts in meetings to address this topic,” the JACC authors continued. “As a result, it took 24 years since the first description of AMR in heart transplant for AMR diagnostic criteria to be included in consensus guideline documents. Publication bias and affinity bias delayed the development and adoption of AMR guidelines.”

Impacts of publication and framing biases

In cases of publication bias, editors are more likely to choose to publish articles based on their personal cognitive biases. For example, publishers typically prefer studies with positive results, to boost their journal’s impact factor. However, this type of publication bias can inflate a treatment’s benefits and lead to early adoption by clinicians before the drug’s effects have been established with certainty.

Similarly, publishers often prefer to publish meta-analysis and guideline documents compared with original research, again because of concerns over impact factor, according to the JACC: Basic to Translational Sciences article.

COVID-19 was first framed as a disease of the elderly, and therefore younger people were less concerned. Because the danger to younger individuals was less publicized in media, younger communities experienced false-consensus bias—the belief that everyone young was relatively safe from infection.

Consequently, during the height of the COVID-19 pandemic, younger people still chose to party in big gatherings such as Mardi Gras or on the beaches.

Combating bias in clinical practice

Cognitive error is common in clinical practice, with up to 75% of errors in internal medicine, for example, attributable to it. Cognitive bias can affect all levels of diagnosis, including information gathering, triggers, context interpretation, processing, and verification, according to the research published in the Journal of the Royal College of Physicians of Edinburgh.

Getting physicians to recognize bias when it occurs may be key to combating it. Research published by the AMA Journal of Ethics cited three occurrences that may mitigate errors from bias in clinical decision-making:^[]

1. Full appreciation of the impact of cognitive bias on errors related to medical decision-making

2. Recognition that such errors are avoidable

3. Optimism about the ability to decrease bias in decision-making

Along these lines, guided reflection (as well as cognitive forcing) may help, as discussed in the AMA Journal of Ethics article. Guided reflection intervention entails searching for and remaining open to alternative diagnoses, along with being willing to reason and reflect on one’s own conclusions with feedback or challenge from a mentor. Guided reflection has produced the most consistent improvements in diagnostic reasoning, according to the AMA Journal of Ethics.