Internal Validation of a Predictive Model for Satisfaction After Primary Total Knee Arthroplasty

Area of Science:

• Orthopedics
• Health Services Research
• Medical Informatics

Background:

• Rising total knee arthroplasty (TKA) rates increase healthcare costs, particularly for dissatisfied patients requiring additional care.
• Bundled payment models necessitate accurate prediction of postoperative outcomes to manage costs effectively.
• Identifying patients at risk for dissatisfaction post-TKA is crucial for proactive intervention and cost containment.

Purpose of the Study:

• To internally validate a novel predictive model for assessing postoperative patient satisfaction following total knee arthroplasty (TKA).
• To determine the efficacy of an 11-component knee survey in predicting patient satisfaction after TKA.
• To establish a reliable tool for identifying patients likely to experience dissatisfaction post-TKA.

Main Methods:

• A cohort of 484 primary TKA patients (January 2014-January 2016) was analyzed.
• Patients were stratified into four risk tiers (low, mild, moderate, high) based on a novel 11-component knee survey.
• Binary logistic regression, multivariate linear regression, and receiver operator characteristic (ROC) curve analysis were employed to validate the survey's predictive capabilities.

Main Results:

• A knee survey score of 96.5 demonstrated high sensitivity (97.5%) and negative predictive value (95.7%) for patient satisfaction.
• Higher knee survey scores were significantly associated with increased odds of postoperative satisfaction (OR 1.03, P=.003).
• Increasing risk tiers, determined by the survey, correlated significantly with decreased satisfaction levels (P=.007).

Conclusions:

• The novel knee survey is a validated tool with high sensitivity and negative predictive value for identifying patients at risk of dissatisfaction after TKA.
• This predictive model can aid healthcare providers in managing patient expectations and resource allocation post-TKA.
• The survey's effectiveness in predicting dissatisfaction is independent of postoperative complications.