Most people have had the sense of being in a grocery aisle, frozen in front of 40 nearly identical bottles of olive oil. That moment of indecision, known as choice paralysis, influenced everything from product placement to app design. But its power extends well beyond regular buying. For nearly three decades, public health officials and medical software developers have believed that doctors feel the same paralysis when presented with more than one appropriate therapeutic option. However, a recent large-scale clinical trial calls this long-held idea into question, and its findings have the potential to transform global health policy and digital health systems. 

At the centre of this discussion lies an ongoing issue: unjustified healthcare variation. When two patients with the same ailment receive drastically different care from two doctors—not because their needs differ but because their habits and local conventions differ— the outcome can be disastrous. Overuse of painkillers, premature surgical referrals, and uneven follow-up treatment are all signs of this variation. To reduce it, health systems rely largely on digital nudges in electronic health records (EHRs): warnings that encourage physicians to make better, safer decisions. 

For nearly thirty years, these nudges have been deliberately limiting. Designers often show clinicians only one recommended alternative. This single-option method stemmed from a seminal 1995 experiment by Riddlemeyer and Shafir, who discovered that providing two equally excellent options actually reduced the likelihood of a doctor switching from a poor plan. The argument suggests that having more choices created cognitive traffic congestion, which led doctors to revert to the status quo. 

New trial discovered the exact reverse 

Published in the JAMA Network. Altinger and colleagues' project, which will open in November 2025, is one of the most thorough attempts to investigate decision fatigue in clinical settings. It took place in 2024 and involved more than 400 practicing primary care physicians in the United States, each of whom was presented with two high-risk clinical scenarios. Their objective was to select whether to continue with a low-quality plan or upgrade to a safer, guideline-recommended option. 

Half of the physicians received the standard single-option nudge. The other half were given two, three, or four equally appropriate alternatives, which in this case were various NSAIDs like ibuprofen and naproxen. 

The findings contradict decades of governmental guidance

In the one-option group, doctors selected the better strategy 44% of the time. However, when doctors were given two or more good options, the rate increased to 62.1%. The impact was even greater in the opioid scenario. When faced with a normal refill request for oxycodone, just 30.5% of single-option physicians picked the safer NSAID. When given many equivalent options, that figure increased to 56.4%. 

The study also discovered that two options were sufficient to trigger this shift; adding a third or fourth did not significantly improve the effect. In other words, two choices overcome the status quo bias, while more than two result in diminishing rewards. 

Why is this transformation happening? The researchers believe that expert decision-making is fundamentally different from regular consumer choices. Doctors do not see four different pain medications as independent decisions. Instead, they automatically organise things into categories. In this case, all NSAIDs belong to the same mental category: "safe, effective options". Providing many options consequently increases the prominence of that category. A single-option notice can be disregarded, whereas two options indicate a deliberate, evidence-based approach. 

For policymakers, this discovery is significant. Much of modern EHR design, from medicine warnings to best-practice reminders, is based on the idea that too many options lead to poor decisions. However, current evidence suggests that constraining choices may potentially impede safer care, particularly in high-risk areas such as opioid prescribing. 

The ramifications go beyond software. Health ministries, regulatory organisations, and hospital systems today face a critical question: should guideline-recommended alternatives in public health programmes be broadened to include at least two possibilities rather than just one? If broadly adopted, such a change might reduce unnecessary opioid prescriptions, prevent premature surgical referrals, and reduce variation in primary care. 

Future systems may potentially organise possibilities into pre-sorted clinical categories, replicating the mental shortcuts that physicians currently employ. Such "smart grouping" could improve the clarity, speed, and alignment of digital nudges with real-world decision-making. 

This influential trial does not pretend to rewrite every element of clinical behaviour; it definitely disproves the notion that doctors freeze when given more than one good alternative. For public health policy, the conclusion is clear: when properly structured, choice may strengthen, not diminish, safer clinical judgements.