Age and Baseline Disability Inform Tradeoffs in Cost Utility of Adult Spinal Deformity Surgery.

Abstract

BACKGROUND AND OBJECTIVES: Surgery can be an effective intervention for adult spinal deformity (ASD), but it is expensive and associated with a high complication rate. The aim of this study was to analyze the cost-effectiveness of ASD surgery, with age and baseline (BL) disability as key variables. METHODS: Decision-analytic models were constructed to assess 3 treatment strategies: operative, nonoperative, and nonoperative with delayed surgery. Model inputs were derived from prospective registry data and published literature, and modeled stochastically. Equivalent decision trees but with different parameter values were constructed for 6 patient subgroups categorized by age ("middle-aged" group: mean age 50 years vs "elderly" group: mean age 70 years), and disability level (low: Oswestry Disability Index (ODI) <20, moderate ODI 20-40, high ODI >40). 1000 Monte Carlo simulations of a hypothetical population of 10 000 patients were generated and used to determine cost-effectiveness metrics and their uncertainty. RESULTS: ASD surgery was cost-effective at a $150 000/quality-adjusted life year (QALY) willingness-to-pay threshold for middle-aged groups with moderate disability (incremental cost-effectiveness ratio (ICER) = $91 340/QALY) favored in 60.1% of patients, and high disability (ICER = $66 090/QALY) favored in 69.9% of patients. For elderly patients with high disability (ICER = $154 300/QALY), surgery was favored in 49.7% of patients. For all other groups, the ICER was above $194 000 and surgery was favored in less than 46.3% of patients. Middle-aged patients gained higher incremental QALYs across all disability levels, and operative strategy demonstrated higher cost-utility in middle-aged patients at the same BL disability. One-way deterministic sensitivity analysis revealed probability of failed nonoperative treatment favored surgery, especially in patients with low disability, whereas probability of complicated operative course favored nonoperative treatment, especially in elderly patients. CONCLUSION: Our analysis revealed that middle-aged patients and those with higher BL disability are more likely to achieve cost-effective surgical intervention. These findings set a basis for further investigation that could better inform clinical decision making for elderly patients experiencing ASD.