Presenting more than one appropriate treatment alternative with a simulated electronic decision-making support tool increased the odds primary care physicians (PCPs) would choose an alternate to standard care, a randomized clinical trial found.

“This challenges earlier suggestions that physicians experience status-quo bias when offered more than one alternative,” according to Gemma Altinger, MBeEcon, an applied behavioral economist at the University of Sydney in Sydney, Australia, and colleagues.

The study, conducted with 402 US PCPs, appeared in JAMA Network Open.

Electronic health alerts are widely used in clinical settings, and their design can significantly influence care decisions — sometimes for better, sometimes for worse, Altinger told Medscape Medical News. “We wanted to understand whether the number of preferred care options presented in these alerts affects the decisions physicians make.”

Her multidisciplinary team had mixed views on whether the number of options would matter. “But that diversity strengthened the trial design and ensured we could reliably test the impact of presenting multiple alternatives on clinical decisions.”

A Two-Scenario Questionnaire

Practicing PCPs from 46 US states and registered with the Qualtrics research network were invited to participate in a survey, then randomized 1:1 to a control or an intervention scenario. Respondents were presented with two common clinical scenarios — one about a surgery referral for hip osteoarthritis and the other about opioid prescribing for lower-back pain. Both involved a decision whether to remain with an existing management plan or to select an alternate plan.

In the control condition, physicians received one appropriate treatment alternative; in the intervention condition, they could receive two, three, or four alternatives. Participating PCPs, equally divided among men and women and 57.5% having fewer than 10 years’ clinical experience, completed the survey in early May 2024.

Physicians had significantly higher odds of choosing a treatment alternative when presented with two or more appropriate alternatives — 44.0% of treatment decisions in the control group vs 62.1% of decisions in the intervention group (unadjusted odds ratio, 2.09; 95% CI, 1.58-2.77).

The multiple-option intervention had a larger effect in the opioid scenario than the surgical scenario. Interestingly, in a secondary study outcome, increasing appropriate alternate options beyond two did not increase the likelihood of a PCP choosing an alternative.

“Our trial challenges a decades-old study suggesting that doctors make poorer decisions when offered more than two new treatment options,” Altinger said. “We found the exact opposite of what this famous study was suggesting,” Altinger said.

She added that the findings have important implications for patient safety. “Simple design changes in clinical decision supports, like presenting two or three good options instead of just one, can help physicians provide better care. But if these systems rely on outdated evidence, they risk doing more harm than good.”

Her group’s aim is to make it easier for physicians to choose wisely. “How choices are framed and presented can sometimes be as important as the options themselves. It’s not just about having good care options; it’s about creating an environment, or a ‘choice architecture,’ that makes it easier for doctors to provide the best care.”

Future randomized trials should evaluate the effectiveness of offering two or more appropriate alternatives, such as in a best-practice alert or a suggested alternative “nudge” on decision-making in a clinical setting, the authors wrote. And introducing more treatment subgroups, such as pharmacological and nonpharmacological treatments should also be examined.

Commenting on the study but not involved in it, Matthew B. Mackwood, MD, MPH, an associate professor of community and family medicine and health policy at Dartmouth Health in Lebanon, New Hampshire, called the main finding novel. “Offering multiple options relatively similar but, to my view superior, to the status quo, gets more engagement in changing the plan than offering a single alternative,” he told Medscape Medical News. “This should help inform how these decision-support tools are designed.”

That said, Mackwood cautioned, “The study was done in a vacuum of sorts compared to real clinical practice, with its time pressures and decision fatigue. I often in practice find these tools helpful when they are novel and I take the time to peruse them, but if I’m busy or at the end of a long day, there’s a good chance I might have more of a status-quo bias.” Evaluating this sort of model in real-world, clinical conditions would be important before assuming it is the ideal option for broad use.

Mackwood also noted that the trial did not factor in patient preferences in decision-making. He supports providing similar guiding information directly to patients. In the real world, he said, deciding between status quo and alternatives entails trade-offs for patients, so “it’s important to not let a single study of clinical decision support tool structures become the whole story for improving patient care.”

This trial was sponsored by the University of Sydney. Co-author Jeffrey A. Linder reported receiving grants from the US National Institute on Aging, National Heart, Lung, and Blood Institute; National Institute of Neurological Disorders and Stroke; and the Agency for Healthcare Research and Quality outside of the submitted work. Linder and Katy J. L. Bell reported receiving grants from the National Health and Medical Research Council outside of the submitted work.