Background

Why is an AI lab mucking with PROs? Machine learning needs good labels in order to work its wonders. Focusing on label quality is critical. In our lab we’ve both looked at existing outcome measures as well as investigate new ones.

PROs vs other measures

Comparing PROs against other measures isn’t new. Most of those studies show varying correlations. By leveraging the OsteoArthritis (OAI) data set, we could do this analysis at a whole new scale.

So, how do PROs reflect patient performance on standardized functional tests? After all, the goal of total knee arthroplasty is to return functionality. Turns out the correlations are very weak.

Do Patient-Reported Outcomes Reflect Objective Measures of Function? Implications for Total Knee Arthroplasty

Brandon G. Hill, Shivesh Shah, Wayne E. Moschetti, Peter L. Schilling

The Journal of Arthroplasty  ·  01 Jul 2023  ·  doi:10.1016/j.arth.2023.04.049

Patient-reported outcomes (PROs) are used in research, clinical practice, and by federal reimbursement models to assess outcomes for patients who have knee osteoarthritis (OA) and total knee arthroplasty (TKA). But do they reflect observed evaluation we see standardized functional tests?

pro tka osteoarthritis oai

What about reflecting structural degradation of the knee? We looked at that too (waiting for publishing). Now, clinicians already know this by experience, but you can have a terrible knee but say much on your PROs.

Other Measures

This all implies that measuring successful outcomes probably needs more than PROs. We are currently researching other possibilities.

Pain

What if there was an objective way to measure how often a patient has pain? This would give researchers something more independent of the psycho-social model.

Gait

What if we could measure patient gait improvement without a full gait lab?