Loss aversion was introduced by Daniel Kahneman and Amos Tversky. While rooted in economics, the idea captures a fundamental feature of human decision-making across domains. The image is CC from Jim Choate

In orthopaedic trauma, translating evidence into practice is hardest when the evidence tells us to stop operating. Stig Brorson’s excellent review of proximal humerus fractures (the book is open access) illustrates just how difficult surgical de-implementation can be. Multiple randomized trials, most prominently the ProFHER trial, have shown that surgery offers no long-term advantage over non-surgical treatment for most patients. Yet, despite this, surgical rates have remained high, with a growing tendency to replace rather than preserve the joint. This persistent gap between evidence and practice is not primarily explained by ignorance or lack of data. It is better understood through the lens of human decision-making.

Data from Sweden since 1998 based on Socialstyrelsen’s open data. There is little evidence that overall practice has changed, and the recent spike in reverse shoulder arthroplasty raises the question of whether forthcoming RCTs will ultimately justify this trend.

One powerful driver behind this pattern is loss aversion. Loss aversion describes our tendency to experience potential losses as more severe than equivalent gains, an effect that has been estimated to be roughly twice as strong as gains. In a surgical context, the perceived “loss” of missing a narrow window for surgery feels far more threatening than the uncertain benefit of avoiding an operation. If a fracture later displaces or healing is poor, the surgeon may feel responsible for having “done nothing.” Early surgery, by contrast, offers a sense of control and closure, even when the evidence suggests that waiting is usually safe. Acting feels safer than observing, not because it is objectively better, but because it reduces anticipated regret.

This bias is reinforced by several sources of uncertainty:

• Fracture classification is imperfect
• Interobserver agreement is modest
• Real patients rarely look exactly like those included in trials

When we are unsure whether an individual case truly fits the evidence base, surgery becomes a risk-mitigation strategy. Add to this the observation that delayed surgery is often associated with worse outcomes, largely due to selection effects, and the preference for early intervention becomes psychologically coherent, even if scientifically weak.

Understanding cognitive biases such as loss aversion does not mean dismissing surgical judgment. It means recognizing that good clinicians are still human decision-makers. Evidence alone rarely changes practice. Awareness of how we weigh risks, losses, and uncertainty is a necessary step toward more consistent, evidence-aligned care.