Comprehensive Summary
This randomized clinical trial (RCT) evaluated the effect of a predictive tool on patient willingness to undergo total knee arthroplasty (TKA) in patients with knee osteoarthritis (OA). Conducted between June 2022 and July 2023, the study involved 211 participants with a mean age of 65.8 years (SD: 8.3), of whom 55.9% were female. Participants were randomly allocated into two groups: the predictive tool group (n = 105) and the treatment-as-usual (TAU) group (n = 106). The predictive tool, named SMART Choice (Knee), provided an online interface predicting quality-of-life improvements after TKA based on age, sex, and baseline symptoms. The TAU group did not have access to the tool. The primary outcome was willingness to undergo surgery at 6 months, measured using binomial logistic regression. Secondary outcomes included treatment preference and decision quality, assessed via the Knee Decision Quality Instrument (K-DQI). At 6 months, there was no significant difference in willingness to undergo TKA between the predictive tool and TAU groups (adjusted odds ratio [OR]: 0.85; 95% CI, 0.42-1.71; P = .64). Notably, the proportion of participants willing to undergo TKA decreased from 69.8% to 52.8% in the TAU group and from 61.0% to 55.2% in the predictive tool group. Among secondary outcomes, the predictive tool group showed a higher preference for nonsurgical treatment at 6 months (45.7%) compared to the TAU group (24.5%). This difference was not statistically significant (OR: 2.16; 95% CI, 0.98-4.92; P = .06). The study found that patients using the tool were more likely to consider conservative management rather than opting for immediate surgical intervention. The SMART Choice tool did not significantly improve decision quality, indicating that patients' surgical preferences were influenced more by individual health perceptions than predictive tool outputs.
Outcomes and Implications
The study’s findings indicate that predictive tools may not significantly alter willingness for TKA but could influence patients to consider nonsurgical options more seriously. Despite the lack of statistical significance in changing willingness for surgery, the higher proportion of participants favoring nonsurgical treatment in the predictive tool group (45.7% vs. 24.5%) suggests that patient decision-making might be influenced by the perception of benefit rather than the objective prediction. The adjusted OR of 0.85 (95% CI, 0.42-1.71) indicates no meaningful impact on willingness to undergo surgery. One critical implication is that predictive tools may still serve as valuable decision aids by facilitating shared decision-making between patients and clinicians, particularly when nonoperative treatments remain viable. Tools like SMART Choice may help guide discussions around rehabilitation, physiotherapy, and lifestyle interventions, ultimately promoting more individualized care plans. While the tool did not significantly reduce willingness for surgery, the trend toward increased preference for nonsurgical management highlights the need for enhanced communication and educational interventions. Additionally, incorporating comprehensive decision aids that address patient values and expectations might improve decision quality and reduce postoperative dissatisfaction. Future research should explore integrating predictive tools into multi-component educational programs to evaluate their effectiveness in a broader clinical context. Further studies should also investigate longitudinal outcomes to assess whether initial tool use influences long-term health outcomes and satisfaction.