PROMs Data Help Personalize Care in Hip and Knee Arthroplasty

Cleveland Clinic orthopaedic surgery researchers are revolutionizing the practice of total hip and knee arthroplasty by developing personalized care approaches based on patient-reported outcome measures (PROMs).

The research, which has spanned the past decade, is led by Nicolas S. Piuzzi, MD, Enterprise Vice Chair of Research for Orthopaedics and Rehabilitation at Cleveland Clinic. For this pioneering work, Dr. Piuzzi recently earned the 2025 Kappa Delta Young Investigator Award from the American Academy of Orthopaedic Surgeons. The award recognizes outstanding clinical musculoskeletal research by investigators under age 40.

“Gathering PROMs in a standardized way helps us understand the individual patient so we can provide a patient-centric, personalized approach to every step of their care,” Dr. Piuzzi says.

This approach is especially relevant now. In 2025, the Centers for Medicare & Medicaid Services (CMS) will begin requiring the collection and reporting of PROMs for hospital inpatients having total hip or knee arthroplasty in the U.S. The change is part of an overall shift towards value-based medicine.

At Cleveland Clinic, baseline and one-year PROMs data have been collected from patients undergoing all elective hip and knee surgeries for the past decade. Those efforts have yielded best practices for data collection and reporting. In addition, under Dr. Piuzzi’s leadership, the Cleveland Clinic Adult Reconstruction Research (CCARR) team has gone much further, applying PROMs to improve clinical care pathways and outcomes for arthroplasty patients.

“It has taken a lot of work and commitment from all the surgical subspecialties, but we now have everything in place to comply with the CMS mandate,” Dr. Piuzzi says. “Collecting data is just the start, however. The most important part will be how we use the data to drive value-based care.”

Collecting PROMs data

In 2015, Dr. Piuzzi and CCARR codeveloped and validated the Orthopaedic Minimal Data Set Episode of Care database, which captures patient demographics, general health PROMs, joint-specific PROMs, and disease severity and treatment details from patients and surgeons at various points throughout total joint arthroplasty care.

That effort involved developing standardized reporting metrics, including minimal clinically important difference (MCID), patient acceptable symptom state (PASS), substantial clinical benefit (SCB) and clinical relevance ratio (CRR), which provides an estimate of the proportion of patients experiencing meaningful improvements over time while accounting for loss to follow-up.

Understanding the role of preoperative factors and PROMs phenotypes

The team’s approach involves use of baseline demographic characteristics along with PROM profiles to identify patients at higher risk for dissatisfaction or complications. This, in turn, informs the use of targeted interventions to address modifiable risk factors, such as obesity, smoking, poor mental health and socioeconomic disadvantage.

For example, a patient with poor reported physical function before surgery can be sent for pre-rehabilitation. Or a patient with low scores on behavioral health screening can be referred to a behavioral health specialist.

“It’s great to have a data-driven approach,” Dr. Piuzzi says. “We’re not just saying, ‘This patient seems sad, so we’ll send them to behavioral health.’ We actually have a process. We measure it, and we are able to learn from the things that work, so hopefully we can deliver a more consistent product based on the needs of each patient.”

In a study of more than 4,000 primary total hip arthroplasty (THA) patients, Dr. Piuzzi and colleagues identified eight distinct patient phenotypes based on above- or below-median scores on three PROMs tools: the Hip Disability and Osteoarthritis Outcome Score (HOOS)-Pain, HOOS-PS (Physical Function Shortform) and Veterans RAND 12-Item Health Survey (VR-12) Mental Component Summary (MCS). Categorizing patients into different combinations of those scores predicted dissatisfaction at one year after surgery.

Using PROMs for surgical decision-making and personalized outcome prediction

Evidence also suggests that PROMs can be leveraged to help determine whether a given patient might benefit from robotically assisted versus manual arthroplasty, a certain manual surgical approach, a specific type of implant or even a particular surgeon.

The CCARR is currently developing a phenotype-based personalized outcome prediction tool to accurately estimate the likelihood of improved pain, function and quality of life after total knee arthroplasty (TKA).

In one hypothetical case, a 68-year-old woman with obesity, a smoking history, hypertension and depression presents with advanced knee osteoarthritis. Her baseline pain, function and MCS scores place her in the phenotype category associated with higher risk of dissatisfaction after TKA.

The tool then estimates her likelihood of freedom from pain, improved knee function and improved knee-related quality of life, as well as her length of stay and risk of postoperative complications. This informs a shared decision-making process in which the surgeon and patient discuss these factors and possible presurgical interventions.

Also underway is the creation of a data science analytics center to continue to expand this PROMs-based approach.

“Our hope is that in the next five to 10 years, we can put together the resources and the teams to bring these tools to all orthopaedic surgical procedures,” Dr. Piuzzi says. “We want to keep patients safe while delivering high-standard, high-quality surgery with predictable outcomes. It’s an ambitious goal, but that’s where we’re going.”