Prediction of patient-reported outcomes after proximal humerus fractures in elderly patients does not appear to be a credible option

Listen to Brian and Lisa discuss the paper 'Prediction of patient-reported outcomes after proximal humerus fractures in elderly patients does not appear to be a credible option' published in the October 2025 issue of Bone & Joint Research.

Click here to read the paper.

Be the first to know when the next episode is live! Follow our social media accounts, @BoneJointPortal and @BoneJointRes on X, and @bonejointres on Instagram for updates!

Transcript

[00:00:00] Welcome back to another episode of AI Talks with Bone & Joint from the publishers of Bone & Joint Research. Today we're discussing the paper 'Prediction of patient reported outcomes after proximal humerus fractures in elderly patients does not appear to be a credible option' published in October 2025 by A Reito and colleagues. I am Brian and I'm joined by my co-host Lisa.

Hello Brian and greetings to all our listeners. This paper explores an important topic, predicting outcomes for elderly patients with proximal humerus fractures. These fractures are quite prevalent, particularly among older adults, making it essential to understand the implications of different treatments.

So Brian, what were the main motivations behind this study?

A good question, Lisa. The researchers aim to identify which patient- and fracture-related factors could predict functional outcomes after two-years. They also explored whether certain types of patients would benefit more from operative treatments as [00:01:00] opposed to nonoperative ones.

That's fascinating and how did they conduct this research? What methodologies were employed?

The researchers performed a secondary analysis of two prior randomized controlled trials, involving 248 patients aged 60 or older. These patients were divided based on the type of fracture, either a two-part or a multipart fracture.

They assessed functional outcomes using the Disabilities of Arm, Shoulder and Hand questionnaire, the Oxford Shoulder Score, and the Constant-Murley score at the two-year mark. Ordinary least squares regression and methodologies from the Predictive Approaches to Treatment effect Heterogeneity or PATH were used to analyze the data.

It seems like a thorough approach. Now, let's move on to their findings. What did they discover about predictors and outcomes?

Interestingly, the frequency of dizziness was the most significant predictor for the DASH and Oxford Shoulder Scores. While the dominance of the injured arm was the key predictor for the Constant score, they found that the variability and [00:02:00] outcomes could be moderately explained by these baseline factors.

However crucially, the interactions between predicted outcomes and treatments were small and uncertain suggesting a lack of Treatment effect Heterogeneity.

So in simpler terms, they couldn't identify a specific subgroup of patients who would benefit more from surgery compared to nonoperative treatment.

Precisely, Lisa. The study indicated that on average operative treatment did not confer additional benefits over nonoperative treatment. Given the absence of credible predictors for better surgical outcomes, they concluded that nonoperative treatment should remain the standard for older patients with proximal humerus fractures.

That's a significant finding. It definitely simplifies clinical decision-making. If surgery doesn't offer a clear benefit, avoiding it seems logical considering the risks and recovery time involved. Indeed, it shifts the focus to managing the overall health and wellbeing of the patient. For instance, chronic dizziness was [00:03:00] highlighted as a major predictor of poorer outcomes and managing such factors could indirectly improve recovery from fractures.

This research seems to advocate for a more holistic approach, concentrating on the patient's overall health rather than just the fracture treatment. Anything else noteworthy from the study?

Yes, Lisa. Another interesting aspect is the historical evolution of treatments. The study acknowledged the increasing popularity of reverse shoulder arthroplasty over the years, which wasn't part of their original trials but is now becoming more common. This provides context to the evolving treatment paradigms in orthopaedics. 

Absolutely. It's always intriguing to see how treatment approaches develop over time based on new evidence. So to summarize, what are the key takeaways for our listeners?

The key takeaways are first, predicting functional outcomes in elderly patients with proximal humerus fractures can be done with moderate accuracy. Essential predictors like body mass index and dizziness are [00:04:00] largely unmodifiable. Second, there is no significant evidence suggesting that certain subgroups benefit more from surgical treatment. Lastly, nonoperative treatment should remain the preferred option given the current evidence.

Thank you Brian, and thank you to our listeners for tuning in. Until next time on AI talks with Bone & Joint. Stay curious and keep exploring new horizons in orthopaedic research.