Presentation (CTI)

OBJECTIVES: To estimate the budget impact of a clinical decision-support algorithm predicting clinically meaningful PROM improvement difference (MCID) after lumbar spine surgery and supporting clinician-mediated pathway decisions under mandatory episode-based payment. CMS TEAM serves as an illustrative scenario transposing a prospectively validated European predictive model into an HEOR decision-analytic framework.
METHODS: We developed a decision-tree budget-impact model for a 30-day mandatory episode-based payment scenario using a hypothetical cohort of 10,000 lumbar spine surgery candidates. Predictive inputs were derived from published clinical evidence: a retrospective EHR-based development study (n = 60; synthetic training augmentation) and a prospective multicenter European validation study assessing MCID/no-MCID prediction. Base-case parameters: AUC = 0.816, sensitivity = 0.81 (no-MCID), specificity = 0.59, accuracy = 0.70; no-MCID prevalence varied from 30% to 50%. Patients with low predicted improvement probability entered a clinician-mediated review-and-optimize pathway; a proportion of true no-MCID patients were assumed to have management modified (surgery deferral, avoidance, or intensified optimization). Economic assumptions reflected avoidable episode-level cost exposure and downstream cost variance. Outcomes included budget impact, patients identified for review, pathway-modified patients, and PROM-based benefit.
RESULTS: The algorithm classified 5,300-6,100 of 10,000 candidates as low predicted benefit, including 2,430-4,050 true no-MCID patients. Clinician-mediated pathway modification was applied to 22% of true no-MCID patients (535-891 optimized). Using a composite value of $6,400 per optimized patient with bundled payment component ($3,000, from a $6,000 target-price overrun reduced by 50%) plus avoided downstream cost variance ($3,400) expected gross economic value was $3.4M-$5.7M per 10,000 candidates. All estimates reflect physician-mediated optimization, not automatic surgical exclusion.
CONCLUSIONS: Applying a prospectively validated European predictive algorithm within an HEOR framework suggests that preoperative readiness profiling may generate meaningful implementation-dependent budget impact under mandatory episode-based payment, by targeting clinician-mediated pathway optimization toward patients at elevated risk of insufficient PROM improvement. These findings support the transferability of algorithm-guided preoperative stratification to value-based spine care pathways.