An economic model of one-level lumbar arthroplasty versus fusion

The Spine Journal 7 (2007) 558–562

An economic model of one-level lumbar arthroplasty versus fusion Richard D. Guyer, MDa,*, Scott G. Tromanhauser, MDb, John J. Regan, MDc a

Texas Back Institute, 6020 W Parker Rd, Ste 200, Plano, TX 75093, USA b Boston Spine Group, 125 Parker Hill Ave, Boston, MA 02120, USA c West Coast Spine Institute, 1205 Spaulding Dr., Ste. 400, Beverly Hills, CA 90212, USA Received 4 May 2006; accepted 20 September 2006

Abstract

BACKGROUND CONTEXT: Degenerative disc disease (DDD) is a cause of low back pain commonly requiring surgical intervention. The option of lumbar total disc replacement (TDR) represents an advance in the surgical treatment of DDD. However, new treatments, particularly those that include the use of new implants, may lead to increased costs to both hospitals and payers. Therefore, it is both necessary and appropriate to examine the potential costs associated with a new procedure such as total disc replacement compared with traditional treatments for a specific pathology. PURPOSE: To perform an economic analysis of lumbar TDR versus three different techniques for lumbar fusion. STUDY DESIGN/SETTING: A cost-minimization model. METHODS: An economic model examining hospital and payer cost perspectives was developed to compare costs of TDR with the CHARITE´Ò Artificial Disc to three spinal fusion procedures: anterior lumbar interbody fusion (ALIF) with iliac crest bone graft (ICBG); ALIF with INFUSEÒ Bone Graft and LT-Cages, and instrumented posterior lumbar interbody fusion (IPLIF) with ICBG. The hospital perspective compares direct medical costs during the index hospitalization. The payer perspective considers direct medical costs of the index hospitalization and those incurred in the following two-year period. The model contains a Diagnostic Related Group (DRG) arm based strictly on DRG coding and payment, and a per-diem arm that includes a device carve-out cost and payment. RESULTS: In the DRG and per-diem arms of the model, compared with TDR, hospital costs are 12.0% higher for ALIF with ICBG, 36.5% higher for ALIF with INFUSE, and 36.5% higher for IPLIF. For payers, in the per-diem arm compared with TDR, ALIF with ICBG has 4.4% lower cost, whereas ALIF with INFUSE and IPLIF have costs of 16.1% and 27.1% higher, respectively. In the DRG arm compared with TDR, payer cost is 87.1% higher for ALIF with ICBG, 82.8% higher for ALIF with INFUSE, and 99.0% higher for IPLIF. CONCLUSIONS: The model shows that the overall economic effect of one-level TDR procedures on hospitals and payers is likely to be less than or at worse equivalent to one-level lumbar fusion procedures. Ó 2007 Elsevier Inc. All rights reserved.

Keywords:

Lumbar spine; Arthroplasty; Disc replacement; Fusion; Arthrodesis; Economic model

Introduction Advancements in medicine almost always result in an increase in cost to the patient, the hospital (if involved), the payer (insurer), or a combination thereof. The

FDA device/drug status: approved for this indication. Authors acknowledge a financial relationship (DePuy Spine), which may indirectly relate to the subject of this research. * Corresponding author. Texas Back Institute, 6020 W Parker Rd, Ste 200, Plano, TX 75093. Tel.: (972) 608-5114; fax: (972) 608-5020. E-mail address: [email protected] (R.D. Guyer) 1529-9430/07/$ – see front matter Ó 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2006.09.006

increasing cost of health care has outpaced inflation on an annual basis since 1998 [1]. Hospitals and payers are interested in keeping their costs as low as possible while maintaining profit potential. Payers are fully informed of the average cost of treating patients with a particular medical disease or condition, specifically by treatment. This is the reason why many payers do not cover organ transplants because of the high cost and low net health benefit (other than avoiding death) of those procedures. When new treatments with either similar or higher costs are introduced to the market by industry, payers likely have little or no actuarial data on how much the treatment

R.D. Guyer et al. / The Spine Journal 7 (2007) 558–562

will cost. Payers may refuse coverage until more data become available so they, in part, can calculate the average cost for a patient they cover undergoing the new treatment. In the authors’ experience, in patients requiring surgical intervention, noncoverage of a new medical device or treatment they are indicated for adds significant frustration to coping with their disease or condition; and physicians experience frustration when noncoverage of Food and Drug Administration (FDA)-approved treatments makes those treatment options unavailable in many cases. Lumbar arthroplasty for the treatment of degenerative disc disease (DDD) has been performed outside the United States for over 2 decades [2,3]. In October 2004, the FDA approved the use of the CHARITE´ Artificial Disc (DePuy Spine, Raynham, MA) for the treatment of one-level lumbar DDD at either L4–L5 or L5–S1. Since the FDA approval, reimbursement for total disc replacement (TDR) in indicated patients has been challenging [4–6] despite published, peer-reviewed level I clinical data showing that TDR is as least as good as fusion in terms of clinical success 2 years after surgery [7]. Economic modeling is used in a number of medical disciplines [8–12], although there has been little use of economic modeling relating to spinal procedures [13–20]. A recent example in the spine is the article by Polly et al. [20] showing the cost-effectiveness of anterior lumbar interbody fusion (ALIF) with rh-BMP2 versus autograft. Another type of economic model is one comparing direct costs. We report on a direct cost economic model comparing one-level TDR with the CHARITE´ Artificial Disc with three different one-level fusion procedures. Does the introduction of lumbar TDR increase costs for hospitals and payers?

Materials and methods The model was created with the assistance of Covance Market Access Services, Inc. (San Diego, CA), a consultant to DePuy Spine. The model contains four comparators, all one-level procedures: (1) TDR with the CHARITE´ Artificial Disc, (2) ALIF with iliac crest bone graft (ICBG), 3) ALIF with LT-Cages and INFUSE (Medtronic Sofamor Danek, Memphis, TN), and (4) posterior lumbar interbody fusion with an adjunct posterolateral fusion and transpedicular fixation (IPLIF). The model assesses direct costs (not charges) from both the hospital and payer perspectives. Data sources used for the model are listed in Table 1. A Medline search of applicable TDR and lumber fusion peer-reviewed published papers and abstracts was performed for the years 1998 to 2005, generating 31 results. FDA-related documents such as protocols, data summaries, and so on were provided by DePuy Spine. Commercial payer claims data for 214 patients receiving the CHARITE´ Artificial Disc at one-level, post-FDA-approval of the device, were obtained from 71

559

Table 1 Sources of data Data type

N

Source

Peer-reviewed medical literature Pre-marketing approval applications, protocols, summaries, etc Commercial payer claims data (fusion comparators) Commercial payer claims data (CHARITE´ comparator) Clinical expert opinion

31 articles/ abstracts 8

Medline Search 1998–2005

1,145

Milliman Database 2002–2003

214

4

DePuy Spine

Post FDA-approval from 71 hospitals Nonauthor spine surgeons

hospitals and used to calculate average charges and subsequent costs for the CHARITE´ comparator. Average charges for the three fusion comparators were calculated by using 1,145 claims from the Milliman Database and included claims from 2002 to 2003. The clinical expert panel included four nonauthor spine surgeons who performed

Table 2 Direct costs Hospital perspective (Time5index hospitalization)

Payer perspective (time5index hospitalizationþ2-year follow-up)

Index hospitalization Facility costs Operating room time Recovery room time Accommodation Therapy costs

Index hospitalization Hospital facility payments Per-diem or DRG-based payment Inpatient physician services CPT code (T-code for CHARITE´) Complications (not all apply to all groups) Vascular injury anterior approach

Physical/occupational therapy Medical devices/supplies/ medication costs Device(s) Surgical supplies Anesthesia Pharmacy Diagnostic tests Radiology (X-ray, MRI, CT, etc) Laboratory Other Blood Respiratory services Cardiology services

Deep wound infection anterior or posterior Donor graft site deep wound infection Donor graft site wound dehiscence Successful surgery follow-up care Office visits Radiographs Medications Therapy Physical therapy sessions Lumbosacral orthosis Unsuccessful surgery follow-up care Office visits Radiographs CT Scan(s) Medications Therapy Physical therapy sessions Lumbosacral orthosis Revision surgery Revision or removal

560

R.D. Guyer et al. / The Spine Journal 7 (2007) 558–562

Table 3 Hospital cost perspective: per diem and DRG payment arms ALIF ALIF w/ Instrumented CHARITE´ w/ICBG INFUSE PLIF Costs ($US, 2006) Facility 4,632 Therapy 177 Medical devices/ 10,914 supplies/medications Diagnostic tests 750 Other 127 Total cost per patient 16,601 Compared with d CHARTIE´ (%)

7,756 267 9,058

6,589 256 14,444

6,444 201 14,768

1,393 1,240 1,067 121 138 186 18,596 22,668 22,662 þ12.0 þ36.5 þ36.5

lumbar TDR and lumbar fusion procedures as a major part of their practice. The hospital perspective includes costs associated only with the hospitalization for the procedure and is reported in average cost per patient. The payer perspective includes costs associated with the hospitalization and procedure, as well as follow-up care for 2 years after surgery, also reported as average cost per patient. A detailed list of the line items for both the hospital and payer perspectives is included in Table 2. Collected claims data were collected from claims filed in the years 2002 and 2003, and the medical component of the consumer price index was used to covert all currency values to 2006 US dollars. Direct costs were analyzed using two different payment arms, diagnostic-related group (DRG) and per diem. In both arms of the model, International Classification of Diseases, Ninth Revision codes were used to identify the total charges using claims data. Cost/charge ratios were then applied based on claims geographic location to arrive at direct costs for each comparator. The per-diem reimbursement arm was created because, in many instances, private payers negotiate different contracts with providers that usually do not match the DRG system. Successful and unsuccessful follow-up care is reported on an average per patient basis. Revision rates published in the recent peer-reviewed literature were used to calculate the rate of revision and unsuccessful follow-up care for each comparator. The 2-year revision rates as published in the literature are 5.4% for TDR with the CHARITE´

Artificial Disc [7], 9.1% for ALIF w/ICBG [7], and 10.8% for both the ALIF w/INFUSE [21] and IPLIF comparators [22]. The cost of revision and unsuccessful follow-up care were multiplied by the rate of revision for each comparator, yielding an average cost of unsuccessful procedures and revision into the total treatment costs on a per-patient basis. Assumptions The claims data for the CHARITE´ comparator contained only one-level procedures. The claims data for the three fusion comparators contained an unknown number of multilevel procedures. To ameliorate cost bias in the fusion comparators, costs in those groups were multiplied by a factor of 0.78 because it has been estimated previously that 78% of the charges for of all lumbar fusions are attributable to one-level procedures [23]. The result of this assumption is that costs for all three fusion comparators were adjusted downward. An unknown number of claims in the ALIF w/INFUSE comparator contained data that included posterior instrumentation. These claims could not be factored out of the analysis. Therefore, the average cost of medical devices used was put directly into the model in place of the claims data for this line item [24]. The result of this assumption is that costs for all three fusion comparators, particularly the ALIF w/INFUSE comparator, were adjusted downward. The per-diem arm methodology calculates the payer’s costs according to a pre-established, fixed payment for a day of patient care while, in addition, reimbursing 100% of the spine implant cost can vary, so the model includes 100% of the implant costs for consistency across all comparators. Carve-out agreements for implant cost can vary, so the model includes the 100% implant costs both for consistency across all comparators. The differences between the DRG and per-diem arms resulted in costs for all comparators being adjusted upward in the per-diem arm in order to compensate for the full cost of the implants compared with the DRG arm. Results The model calculates the same direct costs for the hospital perspective for both the DRG and per-diem arms

Table 4 Payer cost perspective: DRG payment arm

Costs ($US, 2006) Index procedure Successful surgery follow-up care Unsuccessful surgery follow-up care Revision surgery (rate) Complications Total cost per patient Compared with CHARITE´ (%)

CHARITE´

ALIF w/ICBG

ALIF w/INFUSE

Instrumented PLIF

9,611 6,000 590 1,218 (5.4%) [7] 194 17,614 d

22,338 6,824 1,023 2,053 (9.1%) [7] 721 32,960 þ87.1

22,165 6,010 1,214 2,437 (10.8%) [22] 370 32,196 þ82.8

24,663 6,010 1,214 2,437 (10.8%) [21] 728 35,052 þ99.0

R.D. Guyer et al. / The Spine Journal 7 (2007) 558–562

561

Table 5 Payer cost perspective: Per diem payment arm

Costs ($US, 2006) Index procedure Successful surgery follow-up care Unsuccessful surgery follow-up care Revision surgery Complications Total cost per patient Compared with CHARITE´ (%)

CHARITE´

ALIF w/ICBG

ALIF w/INFUSE

Instrumented PLIF

16,882 6,000 590 1,218 (5.4%) [7] 194 24,885 d

13,156 6,824 1,023 2,053 (9.1%) [7] 721 23,778 4.4

18,861 6,010 6,824 2,437 (10.8%) [22] 370 28,892 þ16.1

21,231 6,010 6,010 2,437 (10.8%) [21] 728 31,620 þ27.1

because per-diem payments affect only the payer, not the hospital. Results of the hospital cost perspective are shown in Table 3. Compared with TDR with CHARITE´, hospital costs are 12.0% higher for ALIF w/ICBG, 36.5% higher for ALIF w/INFUSE, and 36.5% higher for IPLIF. From the payer perspective, direct costs in the DRG arm are shown in Table 4. Compared with TDR with CHARITE´, payer costs are 87.1% higher for ALIF w/ICBG, 82.8% higher for ALIF w/ INFUSE, and 99.0% higher for IPLIF. In the per-diem arm (Table 5), compared with TDR with CHARITE´, ALIF with ICBG has 4.4% lower costs, whereas ALIF w/INFUSE and IPLIF have costs of 16.1% and 27.1% higher, respectively.

Discussion There are two limitations for this model. The first is that it only derives payer costs out to 2 years after surgery. Costeffectiveness of one treatment over another was not considered in this model and should be addressed as a separate topic of study. Reported rates of adjacent-level disease (transition syndrome) after lumbar fusion requiring reoperation are 20% to 36% [25,26]. TDR has the potential to reduce this rate considerably. In a study of 100 patients with CHARITE´ TDR and 10-year follow-up, Lemaire et al. [27] described a rate of 2% of adjacent-level disease requiring reoperation. More data are necessary to confirm Lemaire’s findings. Therefore, a cost-effectiveness economic model using 5- to 10-year follow-up will be appropriate when more data are available so that the cost of transition syndrome requiring reoperation can be factored into all comparators. The second limitation is the use of instrumented PLIF as a comparator instead of 360 (anterior/posterior) fusion. Unfortunately, the available claims data were not specific enough to select definitive 360 fusion cases so the IPLIF group had to be used as a compromise. Given the fact that 360 fusion procedures require two incisions, different DRG codes, and longer operative times, it can be expected that the direct costs to hospitals and payers for 360 fusion would be higher than the costs in the IPLIF group. Therefore, the cost of CHARITE´ TDR for both hospitals and

payers (primarily payers) would be much less than the cost of 360 fusion. A model produces a reasonable estimate based on available data and assumptions made in the face of a lack of data. In this case, the model produced an estimate of the direct costs to both hospitals and payers of four different lumbar procedures for the treatment of one-level DDD. For hospitals, CHARITE´ TDR results in lower direct costs than any of the three fusion procedures. Even if the CHARITE´ comparator hospital costs are estimated to be too low the cost of CHARITE´ TDR to hospitals would be equivalent to the three fusion procedures. For payers, under the DRG coding system, the costs associated with fusion are 84% to 100% more than CHARITE´ TDR. The primary reason for this is that TDR has been assigned DRG 499 and 500 codes, which have lower reimbursement values compared with fusion codes. In the per-diem arm, in which carve-out payments occur, the cost of CHARITE´ TDR is 5% greater than ALIF w/ICBG but less than the other two comparators. At worst, payers using a per-diem payment system will incur comparable costs for CHARITE´ TDR compared with fusion over a 2-year time period. Level I data showing safety and effectiveness of the FDA-approved CHARITE´ Artificial Disc exists [7]. Based on the results of this model, for payers, cost should not be the primary issue for noncoverage of lumbar TDR. Costs associated with one-level CHARITE´ TDR for the surgical procedure and 2 years of follow-up care are lower than, or at worst, equivalent to, costs associated with one-level lumbar fusion. Hospitals and payers do not incur increased costs from CHARITE´ Artificial Disc TDR compared with fusion. The introduction of lumbar TDR and a novel device for the treatment of DDD to the US market does not increase costs associated with the treatment in indicated patients.

References [1] Strunk BC, Ginsburg PB, Cookson JP. Tracking health care costs: declining growth trend pauses in 2004. Health Aff (Millwood) 2005;(Suppl). W5–286–W5–95. [2] Buttner-Janz K, Schellnack K, Zippel H. [An alternative treatment strategy in lumbar intervertebral disk damage using an SB Charite´ modular type intervertebral disk endoprosthesis]. Z Orthop Ihre Grenzgeb 1987;125:1–6.

562

R.D. Guyer et al. / The Spine Journal 7 (2007) 558–562

[3] Bu¨ttner-Janz K. Letter to the Editor concerning ‘‘Charite´ total disc replacementdclinical and radiographical results after an average follow-up of 17 years’’ (M. Putzier et al.). Eur Spine J 2006;15:510–3, 21–2. [4] Cigna HealthCare Coverage Position: intervertebral disc (IVD) prosthesis. Available at: http://www.cigna.com/health/provider/medical/ procedural/coverage_positions/medical/mm_0104_coverageposition criteria_intervebral_disc_prosthesis.pdf. Accessed May 1, 2006. [5] Blue Cross Blue Shield artificial vertebral disc replacement tec assessment. Available at: http://www.bcbs.com/tec/vol20/20_01.pdf. Accessed May 1, 2006. [6] CMS. Decision Memo for Lumbar Artificial Disc Replacement (CAG-00292N), U.S. Department of Health. Available at http://www. cms.hhs.gov/mcd/viewdecisionmemo.asp?id5170. Accessed May 16, 2006. [7] Blumenthal SL, McAfee PC, Guyer RD, et al. A prospective, randomized, multicenter Food and Drug Administration investigational device exemptions study of lumbar total disc replacement with the CHARITE´Ô Artificial Disc versus lumbar fusion part I: evaluation of clinical outcomes. Spine 2005;30:1565–75. [8] Albertsen PC, Pellissier JM, Lowe FC, et al. Economic analysis of finasteride: a model-based approach using data from the Proscar Long-Term Efficacy and Safety Study. Clin Ther 1999;21:1006–24. [9] Baxter K, Bevan G. An economic model to estimate the relative costs over 20 years of different hip prostheses. J Epidemiol Community Health 1999;53:542–7. [10] Evens RG. Economic implications of a new technology installation: a CT model. AJR Am J Roentgenol 1981;136:673–7. [11] Goddard JA, Malek M, Tavakoli M. An economic model of the market for hospital treatment for non-urgent conditions. Health Econ 1995;4:41–55. [12] Haycox A, Butterworth M, Walley T, et al. Development of an economic model for the management of upper gastrointestinal disease in primary care. Preliminary findings. Pharmacoeconomics 1998; 14(Suppl 2):11–23. [13] Angevine PD, Zivin JG, McCormick PC. Cost-effectiveness of single-level anterior cervical discectomy and fusion for cervical spondylosis. Spine 2005;30:1989–97. [14] Blackmore CC, Ramsey SD, Mann FA, et al. Cervical spine screening with CT in trauma patients: a cost-effectiveness analysis. Radiology 1999;212:117–25.

[15] Blackmore CC, Zelman WN, Glick ND. Resource cost analysis of cervical spine trauma radiography. Radiology 2001;220: 581–7. [16] Hollingworth W, Gray DT, Martin BI, et al. Rapid magnetic resonance imaging for diagnosing cancer-related low back pain. J Gen Intern Med 2003;18:303–12. [17] Kaneriya PP, Schweitzer ME, Spettell C, et al. The cost-effectiveness of oblique radiography in the exclusion of C7-T1 injury in trauma patients. AJR Am J Roentgenol 1998;171:959–62. [18] Lorish TR, Tanabe CT, Waller FT, et al. Correlation between health outcome and length of hospital stay in lumbar microdiscectomy. Spine 1998;23:2195–200. [19] Oleinick A, Gluck JV, Guire KE. Diagnostic and management procedures for compensable back injuries without serious associated injuries. Modeling of the 1991 injury cohort from a major Michigan compensation insurer. Spine 1998;23:93–110. [20] Polly DW Jr, Ackerman SJ, Shaffrey CI, et al. A cost analysis of bone morphogenetic protein versus autogenous iliac crest bone graft in single-level anterior lumbar fusion. Orthopedics 2003;26: 1027–37. [21] Brantigan JW, Steffee AD, Lewis ML, et al. Lumbar interbody fusion using the Brantigan I/F cage for posterior lumbar interbody fusion and the variable pedicle screw placement system: two-year results from a Food and Drug Administration investigational device exemption clinical trial. Spine 2000;25:1437–46. [22] Burkus JK, Heim SE, Gornet MF, et al. Is INFUSE bone graft superior to autograft bone? An integrated analysis of clinical trials using the LT-CAGE lumbar tapered fusion device. J Spinal Disord Tech 2003;16:113–22. [23] Federal Register, Vol 68, No. 148, August 1, 2003. [24] Orthopaedic Network News 2005 spinal fusion instrumentation price comparison. Available at: http://www.orthopedicnetworknews.com/ private1/archives/onn164spinematrix.pdf. Accessed October 12, 2005. [25] Ghiselli G, Wang JC, Bhatia NN, et al. Adjacent segment degeneration in the lumbar spine. J Bone Joint Surg Am 2004;86A: 1497–503. [26] Gillet P. The fate of the adjacent motion segments after lumbar fusion. J Spinal Disord Tech 2003;16:338–45. [27] Lemaire JP, Carrier H, Sari Ali E, et al. Clinical and radiological outcomes with the CHARITE´Ô Artificial Disc: a 10-year minimum follow-up. J Spinal Disord 2005;18:353–9.