P82. Multi-level procedures and interbody fusions require target price adjustment to build a sustainable lumbar fusion bundled payment model

BACKGROUND CONTEXT Bundled payment models, designed to incentivize high-value care, require risk adjustment to ensure appropriate targets are set based on patient comorbidity profiles. While this may be standardized for many services, spine fusions demonstrate significant variability in approach, invasiveness, and use of implants, that may require further risk adjustment. PURPOSE To evaluate the variability in costs of spinal fusion episodes in a private insurer bundle payment program and identify whether current procedural terminology (CPT) code modifications or exclusions are necessary for the successful and sustainable implementation of these models. STUDY DESIGN/SETTING Retrospective single-institution cohort study. PATIENT SAMPLE A total of 542 total lumbar fusion episodes in a private insurer bundled payment program from October 2018–December 2020. OUTCOME MEASURES There were 120-day episode of care net surplus or deficit, presence of profit/loss, 90-day readmissions, discharge disposition, and length of hospital stay. METHODS A retrospective review was conducted of all lumbar fusions in a single institution's payer database. Surgical characteristics (fusion approach [posterior lumbar decompression and fusion (PLDF), transforaminal lumbar interbody fusion (TLIF), and circumferential fusion], number of levels fused, primary vs revision procedure, and number of interbody cages) were collected from manual chart review. Episode of care cost data were collected and reported as net profit or loss with respect to set target prices. A multivariate linear regression model was constructed to measure the independent effect of primary vs revision, levels fused, and approach on the net cost savings or loss. RESULTS Most procedures were PLDFs (N = 312, 57.6%), single-level (N = 416, 76.8%) and primary fusions (N = 477, 88.0%). Overall, 197 (36.3%) resulted in a net loss and 345 (65.7%) resulted in cost savings. Fusions resulting in a loss were more likely to be three levels (7.11% vs 2.03%, p=0.005), revision fusions (18.8% vs 8.12%, p<0.001), and TLIF (47.7% vs 35.1%, p<0.001) or circumferential fusions (p<0.001). One-level PLDFs resulted in the greatest cost savings per episode ($6,883), as only 28.0% of episodes resulted in a loss. TLIFs demonstrated cost savings of +$165 per episode for single level fusions but losses of -$9,219 for two-level TLIFs. Across both PLDFs and TLIFs, three-level procedures resulted in significant losses of -$23,040 and -$18,887, respectively. For circumferential fusions, one-level fusions resulted in losses of -$17,169 per case which rose to -$64,485 and -$49,222 for two- and three-level fusions. All two- and three-level circumferential spinal fusions resulted in a loss. On multivariable regression, TLIF and circumferential fusion losses were independently associated with a loss of -$7,378 (p=0.004) and -$42,185 (p<0.001), respectively, compared to PLDF. Three-level fusions were independently associated with a -$26,003 loss compared to single-level fusions (p<0.001). CONCLUSIONS Interbody fusions, especially circumferential fusions, and multi-level procedures are not adequately risk adjusted by current bundled payment models. Health systems may not be able to financially support these alternative payment models without improved procedure-specific price risk adjustment. FDA Device/Drug Status This abstract does not discuss or include any applicable devices or drugs. Bundled payment models, designed to incentivize high-value care, require risk adjustment to ensure appropriate targets are set based on patient comorbidity profiles. While this may be standardized for many services, spine fusions demonstrate significant variability in approach, invasiveness, and use of implants, that may require further risk adjustment. To evaluate the variability in costs of spinal fusion episodes in a private insurer bundle payment program and identify whether current procedural terminology (CPT) code modifications or exclusions are necessary for the successful and sustainable implementation of these models. Retrospective single-institution cohort study. A total of 542 total lumbar fusion episodes in a private insurer bundled payment program from October 2018–December 2020. There were 120-day episode of care net surplus or deficit, presence of profit/loss, 90-day readmissions, discharge disposition, and length of hospital stay. A retrospective review was conducted of all lumbar fusions in a single institution's payer database. Surgical characteristics (fusion approach [posterior lumbar decompression and fusion (PLDF), transforaminal lumbar interbody fusion (TLIF), and circumferential fusion], number of levels fused, primary vs revision procedure, and number of interbody cages) were collected from manual chart review. Episode of care cost data were collected and reported as net profit or loss with respect to set target prices. A multivariate linear regression model was constructed to measure the independent effect of primary vs revision, levels fused, and approach on the net cost savings or loss. Most procedures were PLDFs (N = 312, 57.6%), single-level (N = 416, 76.8%) and primary fusions (N = 477, 88.0%). Overall, 197 (36.3%) resulted in a net loss and 345 (65.7%) resulted in cost savings. Fusions resulting in a loss were more likely to be three levels (7.11% vs 2.03%, p=0.005), revision fusions (18.8% vs 8.12%, p<0.001), and TLIF (47.7% vs 35.1%, p<0.001) or circumferential fusions (p<0.001). One-level PLDFs resulted in the greatest cost savings per episode ($6,883), as only 28.0% of episodes resulted in a loss. TLIFs demonstrated cost savings of +$165 per episode for single level fusions but losses of -$9,219 for two-level TLIFs. Across both PLDFs and TLIFs, three-level procedures resulted in significant losses of -$23,040 and -$18,887, respectively. For circumferential fusions, one-level fusions resulted in losses of -$17,169 per case which rose to -$64,485 and -$49,222 for two- and three-level fusions. All two- and three-level circumferential spinal fusions resulted in a loss. On multivariable regression, TLIF and circumferential fusion losses were independently associated with a loss of -$7,378 (p=0.004) and -$42,185 (p<0.001), respectively, compared to PLDF. Three-level fusions were independently associated with a -$26,003 loss compared to single-level fusions (p<0.001). Interbody fusions, especially circumferential fusions, and multi-level procedures are not adequately risk adjusted by current bundled payment models. Health systems may not be able to financially support these alternative payment models without improved procedure-specific price risk adjustment.