Predictive tool use has little effect on knee surgery decision-making

Researchers demonstrated that the use of a tool to predict total knee arthroplasty (TKA) in patients with knee osteoarthritis had little impact on patient-reported willingness to undergo the procedure, according to a study published recently in JAMA Network Open.

The research team indicated that roughly 10 to 20 percent of patients report dissatisfaction following TKA, but levels of dissatisfaction vary significantly. Predictive tools to aid decision-making for patients with knee osteoarthritis considering TKA could provide more individualized information about likely surgical outcomes, but the effectiveness of these tools has not been rigorously assessed.

To help close this research gap, the team conducted a randomized clinical trial (RCT) to evaluate the impact of one predictive model – the SMART Choice (Knee) tool – on patient willingness to undergo TKA.

The study compared predictive tool use with treatment as usual in a cohort of 211 patients with unilateral knee osteoarthritis who had previously tried non-surgical interventions to manage the condition and were considering TKA. The RCT was conducted between June 30, 2022, and July 31, 2023, with participant follow-up for six months.

The intervention group, which contained 105 patients, was given access to the SMART Choice tool online. The model is designed to use factors like patient sex, age and baseline symptoms to forecast the likelihood of improvement in a patient’s quality of life if they undergo TKA.

The control group received treatment as usual and did not receive access to the predictive tool.

Participants’ willingness to undergo surgery at 6 months after tool use, treatment preferences and the quality of their decision-making processes were each measured.

Patients’ baseline willingness for surgery was also recorded, and after adjusting for differences in those baselines, the analysis revealed that the predictive tool did not significantly impact willingness for surgery.

These findings indicate that additional research around predictive tools for TKA is needed, the researchers concluded.

“Despite the absence of treatment effect on willingness for TKA, predictive tools might still enhance health outcomes of patients with knee osteoarthritis,” the authors wrote. “Additional research is needed to optimize the design and implementation of predictive tools, address limitations, and fully understand their effect on the decision-making process in TKA.”

Forecasting surgical outcomes and reducing patients’ risk is a promising potential application of predictive analytics in healthcare across specialties.

In October, researchers from the University of Buffalo showed that a tool originally designed to predict a patient’s risk level for various addictions could also accurately predict bariatric surgery outcomes.

Bariatric surgery can be successful in patients with severe obesity, but the research team underscored that some patients experience adverse outcomes – like weight regain or the development of new addictive behaviors – following the procedure. The predictive tool is designed to analyze relevant psychosocial and genetic factors to determine who is at risk for such outcomes.

The results identified a significant portion of the cohort with a genetic risk for addictive behaviors and showed that participants carrying a particular gene variant were more responsive to weight loss treatment.