120. An enhanced risk-adjustment method for defining bundled payments in complex spine surgery: an analysis of Medicare beneficiaries

Abstract

BACKGROUND CONTEXT Despite an increasing interest in adoption of bundled-payments as a means of controlling costs in the elective spine surgical cases, providers remain wary of extending these value-based reimbursement models in complex spine surgeries. Due to the variability seen in the difficulty of these complex cases–bundled-payment models will ultimately create dis-incentives among surgeons and providers if adequate risk-adjustment is not carried out in order to define appropriate target episode prices. The current BCPI model for complex spine surgery relies on the use of three DRG systems to stratify prices: (1) DRG-456: Spinal fusion except cervical, with spinal curvature/malignancy/infections or 9+ fusions with major comorbidities, (2) DRG-457: Spinal fusion except cervical, with spinal curvature/malignancy/infections or 9+ fusions with clinical complications and (3) DRG-459: Spinal fusion except cervical, with spinal curvature/malignancy/infections or 9+ fusions without clinical complications or major co-morbidities. While these DRG systems may be able to risk-adjust prices to some degree, based on the complexity of the patient at presentation, they do not take into account other important clinical factors. For instance, anterior fusions fall under one broad DRG umbrella, despite them having a different postoperative outcome as compared to posterior-only fusions. Similarly, these DRG systems also do not account for differences in principal diagnosis (infection vs tumor vs deformity). PURPOSE We propose an enhanced risk-adjustment model for defining 90-day bundled payments in complex spine surgeries. STUDY DESIGN/SETTING Retrospective analysis of 2008-2014 100% Medicare Standard Analytical Files (SAF100). PATIENT SAMPLE The 2008-2014 100% Medicare Standard Analytical Files (SAF5) were queried using DRG codes (456, 457 and 458) to identify patients undergoing complex spine surgery. Records were filtered to include those patients who had complete follow-up data starting from day 0 of surgery up to 90-days postoperatively. Patients who died during the 90-day follow-up period and/or had incomplete follow-up were excluded. Patients who had missing data with regard to gender, age and state were also excluded. OUTCOME MEASURES Factors associated with significant cost-variation in a DRG-based 90-day bundle for complex spine surgery. METHODS Multivariate linear regression models were built to assess the marginal cost-increase of patient-level (age and gender), geography-level (state), co-morbidities and principal diagnosis on 90-day costs, while controlling for other co-variates. RESULTS A total of 28,444 patients with complete cost-data were included in the study. The average 90-day cost of the entire study sample was $55,978 ± $33,741, with individual 90-day bundled reimbursements of DRG-456 = $87,281 ± $42,135, DRG-457 = $56,252 ± $29,359 and DRG-458 = $38,108 ± $19,356. Procedure-level factors associated with significant increases in 90-day costs were anterior fusions (+ $4,242), undergoing a 4-8 level fusion (+ $4,421), >8 level fusion (+ $16,416), use of interbody (+ $919), concurrent discectomy (+ $1,043), concurrent laminectomy (+ $2,202) and use of intraoperative neuromonitoring (+ $3,082). Undergoing surgery for infection/osteomyelitis was associated with higher costs (+ $6,217), as compared to patients undergoing surgical correction for primary deformity. The top five co-morbidities associated with the highest marginal cost increases were malnutrition (+ $12,511), CVA/stroke (+ $11,397), coagulopathy (+ $8,597), hyper-coagulopathy (+ $8,157) and obesity (+ $7,124). Significant state-level variation was seen, with Alaska (+ $29,208), Maryland (+ $20,126), Massachusetts (+ $18,834), Washington (+ $14,865), Vermont (+ $13,764) having the highest costs. CONCLUSIONS Due to the significant heterogeneity present in the DRG definitions of complex spine surgery, the current BPCI DRG-based bundled payment model is not entirely adequate for defining payments associated with these patients. For instance, infection cases should not be part of the same bundled payment episode as primary deformity cases. Physicians should promote the utilization of these additional, yet crucial factors, for risk-adjusting target 90-day episode prices and prevent the creation of a financial disincentive in a value-based care approach in complex spine cases. FDA DEVICE/DRUG STATUS This abstract does not discuss or include any applicable devices or drugs.