The Spine Journal 12 (2012) 676690

Review Article

Cost-utility analysis in spine care: a systematic review

Christopher K. Kepler, MD, MBAa, Sean M. Wilkinson, BAb, Kristen E. Radcliff, MDa, Alexander R. Vaccaro, MD, PhDa, David G. Anderson, MDa, Alan S. Hilibrand, MDa, Todd J. Albert, MDa, Jeffrey A. Rihn, MDa,*
a

Department of Orthopaedic Surgery, The Rothman Institute, Thomas Jefferson University Hospital, 925 Chestnut St, 5th oor, Philadelphia, PA 19107, USA b Philadelphia College of Osteopathic Medicine, 4170 City Ave., Philadelphia, PA 19131, USA Received 10 June 2011; revised 12 January 2012; accepted 17 May 2012

Abstract

BACKGROUND CONTEXT: Despite the importance of the information provided by cost-utility analyses (CUAs), there has been a lack of these types of studies performed in the area of spinal care. PURPOSE: To systematically review cost-utility studies published on spinal care between 1976 and 2010. STUDY DESIGN: Systematic review. METHODS: All CUAs pertaining to spinal care published between 1976 and 2010 were identied using the cost-effectiveness analysis (CEA) registry database (Tufts Medical Center, Institute for Clinical Research and Health Policy) and National Health Service Economic Evaluation Database (NHS EED). The keywords used to search both the registry databases were the following: spine, spinal, neck, back, cervical, lumbar, thoracic, and scoliosis. Search of the CEA registry provided

FDA device/drug status: Not applicable. Author disclosures: CKK: Research Support: OREF (D), NASS (D). SMW: Nothing to disclose. KER: Nothing to disclose. ARV: Royalties: DePuy Spine (E, Paid directly to institution/employer), Biomet Spine (E, Paid directly to institution/employer), Globus (F, Paid directly to institution/employer), Medtronic (H, Paid directly to institution/employer), Aesculap (B, Paid directly to institution/employer), Book royalties (C, Paid directly to institution/employer), Stryker Spine (G, Paid directly to institution/employer), Alphatec (B, Paid directly to institution/employer); Stock Ownership: K2M (Dissolved, Paid directly to institution/employer), Rothman Institute and related holdings (Partner in Practice, Paid directly to institution/employer), Cytonics (Unknown, Paid directly to institution/ employer), Disc Motion Technology (Unknown, Paid directly to institution/employer), Location Based Intelligence (20%, Paid directly to institution/employer), Progressive Spinal Technology (Unknown, Paid directly to institution/employer), Computational Biodynamics (Unknown, Paid directly to institution/employer), Stout Medical (B, Paid directly to institution/employer), Bonovo Orthopaedics (100,000 shares, Paid directly to institution/employer), Electrocure (D, Paid directly to institution/employer), Vertiex (30,000 shares, Paid directly to institution/employer), Flagship Surgical (Unknown, Paid directly to institution/employer), In Vivo (Unknown, Paid directly to institution/employer), Small Bone Innovations (30,000 shares, Paid directly to institution/employer), Neucore (22,000 shares, Paid directly to institution/employer), Crosscurrent (125,000 shares, Paid directly to institution/employer), Onset Medical (1,535 shares, Paid directly to institution/employer), Orthovita (D, Paid directly to institution/employer), Syndicom (2,750 shares, Paid directly to institution/employer), Biomerix (25,000 shares, Paid directly to institution/ employer), Paradigm Spine (146,875 shares, Paid directly to institution/ employer), Spineology (28,750 shares, Paid directly to institution/employer), Gamma Spine (15%, Paid directly to institution/employer), Replication Medica (15,250 shares, Paid directly to institution/employer), Breakaway Imaging (100,000 shares, Paid directly to institution/ 1529-9430/$ - see front matter 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.spinee.2012.05.011

employer), Crosstree Medical (41,667 shares, Paid directly to institution/ employer), Globus (816,123 shares, Paid directly to institution/employer), FlowPharma (Unknown, Paid directly to institution/employer), Advanced Spinal Intellectual Properties (30%, F, Paid directly to institution/employer), SpineMedica (Unknown, Paid directly to institution/employer), RIS (!5,000 shares, Paid directly to institution/employer); Consulting: Gerson Lehrman Group (B, Paid directly to institution/employer), Benvenue Medical (A, Paid directly to institution/employer), Guidepoint Global (B), Medacorp (B); Trips/Travel: Stryker (B); Board of Directors: AO Spine (North America Board of Directors, B, Paid directly to institution/ employer), ASIA (Past President Board Member, Paid directly to institution/employer), NASS (Past Program Committee Co-Chairman); Research Support (Staff/Materials): AO Spine (E), Cerapeutics (C). DGA: Consulting: DePuy (C), Medtronic (A), Globus (B), Synthes (G); Royalties: Depuy (G), Medtronic (F). ASH: Royalties: Biomet (F), Alphatec (F), Aesculap (A), Zimmer (B); Private Investments: Amedica (20,000 shares, 0%), Nexgen (B), Vertiex (B), Benvenue Medical (B), Pioneer Surgical (C), Lifespine (20,000 shares, 0%, less than 1% of entity), Paradigm Spine (B), PSD (B), Syndicom (20,000 shares, 0%); Scientic Advisory Board: Amedica (20,000 shares of options). TJA: Board of Directors: CSRS (none), United SAB (B), Gentis SAB (none); Consulting: DePuy (C), Biomet (A); Royalties: Depuy (H); Stock: Biomerix (!1,000 shares), Breakaway Imaging (!1,000 options), Crosstree (!1,000 options), Gentis (15,000 options), International Orthopaedic Alliance (1,000 shares), Invuity (!1,000 shares), Paradigm Spine (250,000 shares), PIONEER (100,000 shares), Vertech (20,000 shares). JAR: Research Support: Depuy (D). The disclosure key can be found on the Table of Contents and at www. TheSpineJournalOnline.com. * Corresponding author. The Rothman Institute, Thomas Jefferson University Hospital, 925 Chestnut St, 5th oor, Philadelphia, PA 19107, USA. Tel.: (267) 339-3500; fax: (215) 503-0580. E-mail address: jrihno16@yahoo.com (J.A. Rihn)

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

677

a total of 28 articles, and the NHS EED yielded an additional 5, all of which were included in this review. Each article was reviewed for the study subject, methodology, and results. Data contained within the databases for each of the 33 articles were recorded, and the manuscripts were reviewed to provide insight into the funding source, analysis perspective, discount rate, and cost-utility ratios. RESULTS: There was wide variation among the 33 studies in methodology. There were 17 operative, 13 nonoperative, and 3 imaging studies. Study subjects included lumbar spine (n527), cervical spine (n54), scoliosis (n51), and lumbar and cervical spine (n51). Twenty-three of the studies were based on the clinical data from prospective randomized studies, 7 on decision models, 2 on prospective observational data, and 1 on a retrospective case series. Sixty cost-utility ratios were reported in the 33 articles. Of the ratios, 19 of 60 (31.6%) were cost saving, 27 of 60 (45%) were less than $100,000/quality-adjusted life year (QALY) gain, and 14 of 60 (23.3%) were greater than $100,000/QALY gain. Only four of 33 (12%) studies contained the four key criteria of costeffectiveness research recommended by the US Panel on Cost-Effectiveness in Health and Medicine. CONCLUSIONS: Thirty-three CUA studies and 60 cost-utility ratios have been published on various aspects of spinal care over the last 30 years. Certain aspects of spinal care have been shown to be cost effective. Further efforts, however, are needed to better dene the value of many aspects of spinal care. Future CUA studies should consider societal cost perspective and carefully consider the durability of clinical benet in determining a study time horizon. 2012 Elsevier Inc. All rights reserved.
Keywords: Cost-utility analysis; Cost-effectiveness; Cervical; Lumbar; Value

Introduction The ultimate goal of medicine is to improve the overall health and quality of life of patients and society as a whole. However, the cost of care to both the payer and society must also be considered. For this reason, there has been a recent shift in health care toward a value-based system, that is, trying to maximize the quality and minimize the cost of health care. The concept of value-based health care takes into consideration both the quality and cost of care measured over time. There are three main components of the value equation as follows: the quality of care must be measured using some type of health outcome metric, the cost of care must be measured, and these measurements must be made over an appropriate time period to capture future events that will affect the overall value of an intervention. Cost-effectiveness analysis (CEA) is a specic type of economic analysis that estimates the value of an intervention. The purpose of this analysis was to calculate the ratio between the cost of a specic health care intervention (numerator) and the benet provided by that intervention as determined by a health outcomes measure (denominator). A cost-utility analysis (CUA) is a specic type of CEA in which the benet is expressed as a utility measure, that is, a preference-based measure of health-related quality scored as a number ranging from 0 (death) to 1 (perfect health). The utility measure can be used to calculate quality-adjusted life years (QALYs). From the policy makers and payers perspective, the value of specic types of interventions should be compared not only with each other within the eld of spinal care but also with that for other disease states, for example, the treatment of diabetes or coronary artery disease. A CUA, based on the preference-based utility measures, provides data that allow for this comparison across various

disciplines of medicine. It is this type of economic analysis that is likely to be used to guide payers and policy makers when making coverage decisions. In 1996, the US Panel on Cost-Effectiveness in Health and Medicine published consensus-based recommendations for performing a CEA [1]. According to these recommendations, the four key components of a CEA are the following: the use of the societal perspective, appropriate incremental comparisons between treatments, appropriate discounting of both the cost and health effect of the treatment, and the use of a community preference-based utility measure [1,2]. Additionally, the panel recommended the use of sensitivity analysis to address uncertainties within the study and recommended that studies have a sufciently long-term time horizon for follow-up, as not to miss incidents that may occur much later in time and affect the cost and/or outcome of a particular intervention [1]. This concept of durability of results is particularly important for interventions with a high initial cost such as spinal surgeryan extended time horizon allows spreading out of the cost across more years assuming that the intervention continues to provide clinical benet. Despite the importance of the information provided by CUAs, there has been a relative paucity of these types of studies performed in the area of spinal care. A systematic review of CUA in orthopedic surgery published in 2005 reviewed all relevant studies published between 1976 and 2001 [2]. The comprehensive review identied only 37 CUAs published within this timeframe. Only ve (14%) of these 37 studies pertained to areas involving spinal care [2]. A study by Dagenais et al. [3] in 2009 systematically reviewed CUAs that address interventions for low back pain dating back to 1998. These authors identied 15 CUAs, the majority of which were published in the last 3 years.

678

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

The purpose of this study was to systematically review all CUAs published between 1976 and 2010 that relate to the management of spinal disorders. This information will be helpful in understanding which spinal interventions have cost-utility data; understanding what these cost-utility data show for these specic interventions; and identifying which areas of spinal care are lacking cost-utility data. Methods All CUAs pertaining to spinal care published between 1976 and 2010 were identied using the CEA registry database (Tufts Medical Center, Institute for Clinical Research and Health Policy) [4]. This is a comprehensive database of all health-related CUAs published from 1976 to 2010. The methods of the development of this database are described in previous publications [2,57]. The data collection process includes a MEDLINE search using the keywords QALYs, quality, and cost-utility analysis [4,6]. The articles are then screened by CEA registry team members to determine whether they contain original costutility estimates. Articles were excluded if they were reviews, editorials, or methodologies; did not measure health benet in terms of QALYs; and were not written in the English language [4,6]. A standard set of data is then collected on each article. This database has been used in numerous studies on cost-effectiveness in the past [2,6,810]. At the time of this analysis, costs in the CEA registry database were expressed in 2010 US dollars. The keywords used to search the CEA registry database were the following: spine, spinal, neck, back, cervical, lumbar, thoracic, and scoliosis. Articles pertaining to the diagnosis and/or treatment of osteoporosis were excluded. Two of the authors (JAR and SMW) performed this search independently and obtained identical results. Each article was reviewed in detail for the subject of the study, methodology, results of the analysis, and length of the follow-up. Studies were also categorized according to the area of spine (eg, lumbar and cervical) and type of intervention (eg, operative, nonoperative, and diagnostic) and comparison type (eg, nonoperative vs. nonoperative, operative vs. nonoperative, and operative vs. operative). Data contained within the CEA registry database for each article were reviewed and recorded, including the source of funding, perspective of the analysis, discount rate used in the analysis, cost-utility ratios, and follow-up period. The database also contains a subjectively assigned rating score for each article based on a seven-point Likert scale, which is based on the recommendations set forth by the US Panel of Cost-Effectiveness in Health and Medicine [1]. Next, we used a similar database, the National Health Service Economic Evaluation Database (NHS EED), as a second potential source of relevant articles. The NHS EED is a database compiled based on searches of MEDLINE, CINAHL, EMBASE, and PsycINFO, which also incorporates hand searching of journals [11]. After articles

are identied, they are analyzed and incorporated into the database by health care economists. This database was searched using the same methodology described previously for the CEA registry to identify additional studies. Results Twenty-eight cost-utility studies on spinal care were identied by the CEA database, and another ve unique studies were identied by the NHS EED database giving 33 total studies for inclusion that included 60 total cost-effectiveness ratios. Seventeen (51%) of the studies involved government funding, 8 (24%) had foundation funding, and 5 (15%) involved industry funding. There were 17 operative, 13 nonoperative, and 3 imaging studies. Twentyseven studies involved the lumbar spine, 4 involved the cervical spine, 1 involved adolescent idiopathic scoliosis, and 1 involved both the lumbar and cervical spines. There was wide variation among the studies in methodological practices. Twenty-three of the studies were based on the clinical data from prospective randomized studies, 7 were based on a decision model analysis, 2 were based on the prospective observational data, and 1 was based on a retrospective case series. Twenty-ve (76%) of the studies and 47 (78%) of the reported cost-utility ratios were published in between 2005 and 2010 (Fig. 1). The average grade of the 33 papers on the seven-point Likert scale was 4.4, with a range of 2 to 6.5. Only four of 33 (12%) articles met the four key recommendations for a CEA described by the US Panel of CostEffectiveness in Health and Medicine, all four of which addressed interventions for lumbar spinal disease [1215]. Overall, 60 cost-utility ratios were reported in the 33 articles, eight of which pertained to the cervical spine. Of the ratios, 19 of 60 (31.6%) were cost saving, 27 of 60 (45%) were less than $100,000/QALY gain, and 14 of 60 (23.3%) were greater than $100,000/QALY gain.

Fig. 1. Number of cost-utility analysis studies and ratios published by year. CEA, cost-effectiveness analysis.

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

679

Of the studies involving the lumbar spine, 17 addressed low back pain, 5 specically addressed disc herniation, 2 addressed lumbar stenosis/degenerative spondylolisthesis, 2 addressed imaging, 1 addressed low back pain after failed surgery, and 1 addressed the treatment of lumbar radicular pain in general. Of the 17 studies that addressed low back pain, 12 compared two types of nonoperative treatment, 3 compared operative to operative treatments, and 2 compared nonoperative to operative treatment. There were 51 cost-utility ratios reported in the 28 studies involving the lumbar spine. These ratios are listed in Table 1. Table 1 also includes three cost-utility ratios reported in the only CUA published on the topic of adolescent idiopathic scoliosis. Of the ve studies that specically addressed lumbar disc herniation, four of them compared operative to nonoperative treatment, with a total of ve cost-utility ratios [15 18]. The reported cost utility of lumbar discectomy ranges from cost saving to $79,000/QALY gain compared with nonoperative treatment (Table 2). A total of 12 costutility ratios were reported in the two studies that address lumbar stenosis and degenerative spondylolisthesis [12,14]. These ratios address decompression versus nonoperative treatment, decompression versus fusion, and instrumented versus noninstrumented fusion in the treatment of stenosis alone and stenosis combined with degenerative spondylolisthesis (Table 3). The study by Tosteson et al. [14] was the only study to report cost-utility ratios for the comparison of decompression versus nonoperative treatment for the treatment of stenosis and degenerative scoliosis. Overall, when looking at all surgically treated patients with spinal stenosis without degenerative spondylolisthesis (total n5394; decompression alone, n5320; decompression and fusion, n543), these authors reported a cost/QALY gain of $90,000 (2010 US dollars) for the surgical treatment of spinal stenosis compared with nonsurgical treatment. When looking only at those patients who had decompression surgery for spinal stenosis without fusion (n5320), the cost/QALY gain was $54,000 (2010 US dollars) [14]. The studies by both Kuntz et al. [12] and Tosteson et al. [14] assess the cost utility of fusion surgery for the treatment of degenerative spondylolisthesis. The study by Kuntz et al. [12] was a decision model analysis that compared laminectomy and noninstrumented fusion with laminectomy alone and laminectomy and instrumented fusion with laminectomy and noninstrumented fusion in the treatment of lumbar stenosis and spondylolisthesis. The resulting comparisons demonstrated that the cost/QALY gain by performing laminectomy and noninstrumented fusion in comparison with performing laminectomy alone was $67,000/ QALY (2010 US dollars), whereas the cost/QALY of performing instrumented fusion compared with noninstrumented fusion was $4,200,000/QALY (2010 US dollars). The study by Tosteson et al. [14] compared various forms of surgical treatment for degenerative spondylolisthesis and stenosis to nonoperative treatment. Overall,

when looking at all surgically treated patients with spinal stenosis with degenerative spondylolisthesis (total, n5368; decompression alone, n519; decompression and fusion, n5344), these authors reported a cost/QALY gain of $130,000 (2010 US dollars) for the surgical treatment of spinal stenosis with degenerative spondylolisthesis compared with nonsurgical treatment. These authors also divided the patients by type of fusion (ie, instrumented, n5269 and noninstrumented, n575) and type of instrumentation (ie, posterolateral pedicle screws, n5209 and circumferential, n560), for which the analyses yielded cost/QALY gain of greater than $100,000 [14]. Of the studies involving the cervical spine, two addressed the surgical treatment of cervical degenerative disease, two addressed nonoperative treatment of neck pain, and one addressed imaging. There were a total of eight cost-utility ratios reported from these cervical studies (Table 4). The costutility data on cervical spine interventions are very limited. All cost-utility ratios pertaining to cervical spine care had a cost/QALY of less than $100,000, including surgery for the treatment of cervical radiculopathy [19], use of a plate when performing anterior cervical discectomy and fusion [20], the use of allograft compared with autograft in anterior cervical discectomy and fusion [20], the use of acupuncture to treat chronic neck pain [21], and the use of screening cervical computed tomography scan compared with radiography in patients with trauma with a moderate-to-high risk of cervical spine fracture [22]. The one exception to this was the use of standard physiotherapy compared with the use of a brief period of physiotherapy alone, which had a cost/QALYof $120,000. The use of screening cervical computed tomography scan compared with radiography in patients with trauma with a low risk of cervical spine fracture [22] also exceeded the $100,000 threshold, with a cost/ QALY of $120,000 (2010 US dollars).

Discussion There is a current emphasis among the stakeholders in health care delivery on value-based health care, that is, an emphasis on providing those interventions/treatments that offer a high degree of a health benet at a reasonable cost. The concept of value is most commonly expressed as cost/ QALY. It is unclear in the United States what is currently considered an acceptable cost/QALY. In Great Britain, the National Institute for Health and Clinical Excellence uses cost/QALY to make coverage and payment decisions. The cutoff for what is considered cost-effective (ie, the willingness-to-pay threshold) ranges between 20,000 and 30,000 per QALY gain for a particular intervention [23]. A willingness-to-pay threshold does not yet exist in the United States, although arbitrary thresholds have ranged from $50,000 to $100,000/QALY gain. Regardless, both private payers and the US government have taken measures to limit the utilization of treatments or diagnostic

680

Table 1 Cost-utility studies and ratios published on the lumbar spine* Perspective (payers vs. society) Payer Discount rate (%) Not used Utility measure EQ5D Study design RCT Likert rating 5 Follow-up period (mo) 12 Cost/ QALY $3,400

Study Lamb et al. [26]

Comparison and population Cognitive behavioral intervention-based therapy versus standard advice for low back pain UK patients with low back pain Recombinant human bone morphogenetic protein 2 versus ICBG Patients aged O60 years undergoing decompression and posterolateral fusion Active physical treatment plus graded activity plus problem solving training versus graded activity plus problem solving training Patients aged 1865 years with low back pain O3 mo with disability Active physical treatment plus graded activity plus problem solving training versus active physical treatment Patients aged 1865 years with low back pain O3 mo with disability Standard open discectomy versus nonoperative care US Medicare patients with lumbar disc herniation Standard open discectomy versus nonoperative care US general patients with lumbar disc herniation Circumferential (360  ) fusion surgery versus nonoperative care US patients with spinal stenosis and spondylolisthesis Posterolateral fusion surgery versus nonoperative care US patients spinal stenosis with degenerative spondylolisthesis Instrumented fusion surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Noninstrumented fusion surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Fusion surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Decompression surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Decompression surgery versus nonoperative care US patients with spinal stenosis without spondylolisthesis

Carreon et al. [27]

Payer

SF6D

RCT

24

Cost saving

Smeets et al. [28]

Society

Not used

EQ5D

RCT

15

Dominated C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

Society

Not used

EQ5D

RCT

15

$39,000

Tosteson et al. [15]

Society Society

3 3 3

EQ5D EQ5D EQ5D

RCT RCT RCT

5 5 6.5

24 24 24

$40,000 $80,000 $120,000

Tosteson et al. [14]

Society

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$34,000

Society

EQ5D

RCT

6.5

24

$55,000

Society

EQ5D

RCT

6.5

24

Society

EQ5D

RCT

6.5

24

Society

EQ5D

RCT

6.5

24

Nielsen et al. [29]

Society

Not used

I5D

RCT

Society

Not used

I5D

RCT

Society

Not used

I5D

RCT

Society

Not used

I5D

RCT

van der Roer et al. [30] van den Hout et al. [16] Hollinghurst et al. [31]

Society

Not used

EQ5D

RCT

4.5

Society Payer

Not used Not used

EQ5D, SF6D, VAS EQ5D

RCT RCT

5 5.5

12 12

Payer

Not used

EQ5D

RCT

5.5

12

Payer

Not used

EQ5D

RCT

5.5

12

Critchley et al. [32]

Payer

3.5

EQ5D

RCT

3.5

18

Payer

3.5

EQ5D

RCT

3.5

18

Fusion surgery versus nonoperative care US patients with spinal stenosis without spondylolisthesis Surgery for stenosis with degenerative spondylolisthesis versus nonoperative care US patients with spinal stenosis Spine surgery for spinal stenosis versus nonoperative care US patients with spinal stenosis Extensive postoperative rehabilitation including alcohol and tobacco cessation therapy versus 1-day postoperative standard rehabilitation Danish patients undergoing fusion for degenerative lumbar disease Extensive postoperative rehabilitation including alcohol and tobacco cessation therapy versus 1-month postoperative standard rehabilitation Danish patients undergoing fusion for degenerative lumbar disease Extensive postoperative rehabilitation including alcohol and tobacco cessation therapy versus 3-month postoperative standard rehabilitation Danish patients undergoing fusion for degenerative lumbar disease Extensive postoperative rehabilitation including alcohol and tobacco cessation therapy versus 6-month postoperative standard rehabilitation Danish patients undergoing fusion for degenerative lumbar disease Intensive training protocol (therapy, back school) versus standard physiotherapy Dutch patients with nonspecic low back pain Early surgery versus prolonged conservative care Dutch patients with disc herniation, radicular symptoms Exercise prescription with nurse-based follow-up versus conservative care UK patients with chronic or recurrent low back pain Alexander technique (six lessons) versus conservative care UK patients with chronic or recurrent low back pain Therapeutic massage (six sessions, 1/wk) versus conservative care UK patients with chronic or recurrent low back pain Physiotherapist-led pain management versus individual therapy British chronic low back pain patients Spinal stabilization versus individual physiotherapy British chronic low back pain patients

$190,000

$130,000

$90,000

Cost saving

Cost saving

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

Cost saving

Cost saving

$11,000

Cost saving $5,100

$12,000

Dominated

Cost saving $2,200 (Continued)

681

682

Table 1 continued Perspective (payers vs. society) Payer Discount rate (%) Not used Utility measure EQ5D Study design RCT Likert rating 4 Follow-up period (mo) 12 Cost/ QALY $11,000

Study Johnson et al. [33] Soegaard et al. [34]

Comparison and population Exercise, education, and cognitive therapy versus conservative care British patients with disabling low back pain Circumferential lumbar fusion versus posterolateral fusion Danish patients with chronic low back pain and/or leg pain from degenerative spondylolisthesis or spondylosis Titanium cages versus femoral allograft for circumferential fusion British patients with chronic low back pain Surgical treatment versus nonsurgical treatment Swedish patients with lumbar disc herniation Preoperative rHuEPO transfusion versus PAD 15-year-old female undergoing surgery for adolescent scoliosis PAD versus control group (no preoperative blood donation) 15-year-old female undergoing surgery for adolescent scoliosis rHuEPO versus control group (no preoperative rHuEPO) 15-year-old female undergoing surgery for adolescent scoliosis SCS versus reoperation US patients diagnosed with FBSS Standardized advice book (The Back Book) and physiotherapy versus The Back Book alone British patients with subacute and chronic low back pain Routine neurosurgical lumber spinal surgery versus no surgery Finnish patients with lumbar radicular pain Up to 10 acupuncture treatments versus usual care only English patients with persistent nonspecic low back pain Conventional 3-week inpatient rehabilitation plus cognitive behavioral pain management program versus rehabilitation program alone German patients with nonspecic LBP of at least 6 months Spinal fusion versus rehabilitation (exercise and education) British patients with chronic low back pain

Society

EQ5D

RCT

24

Cost saving

Freeman et al. [35] Hansson and Hansson [17] Vitale et al. [36]

Payer

3.5

SF6D

RCT

24

Dominated C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

Society Payer

5 Not used

EQ5D Based on literature Based on literature

Prospective case-control Decision tree model Decision tree model

3.5 4

24 Decision tree model Decision tree model

Cost saving $9,900,000

Payer

Not used

$1,500,000

Payer

Not used

Based on literature

Decision tree model

Decision tree model

$1,700,000

North et al. [37] Rivero-Arias et al. [38]

Payer Society

Not used Not used

Not reported EQ5D

RCT RCT

4 4.5

37 12

Cost saving $2,500

Rsnen et al. aa [19] Ratcliffe et al. [39] Schweikert et al. [40]

Payer

HRQOL questionnaire SF6D

Prospective cohort study RCT

$2,500

Society

3.5

24

$8,200

Society

Not used

EQ5D

RCT

Cost saving

Rivero-Arias et al. [41]

Society

3.5

EQ5D

RCT

4.5

24

$88,000

Taylor and Taylor [42] UK BEAM Trial Team [43]

Payer Payer

6 Not used

Based on literature EQ5D

Decision tree model RCT

4 6

Decision tree model 12

Payer

Not used

EQ5D

RCT

12

Payer

Not used

EQ5D

RCT

12

Gilbert et al. [44] Williams et al. [45]

Payer

EQ5D

RCT

4.5

24

Payer

Not used

EQ5D

RCT

Hollingworth et al. [46] Kuntz et al. [12]

Payer

Based on literature Based on literature Based on literature Based on literature Survey plus literature

Decision tree model Decision tree model Decision tree model Decision tree model Decision tree model

Decision tree model Decision tree model Decision tree model Decision tree model Decision tree model

Payer

Not used

Payer

Not used

Malter et al. [18] Launois et al. [47]

Payer Payer

5 5

4.5 3

SCS versus conservative treatment UK patients with FBSS Manipulation alone plus conservative care versus manipulation and exercise plus conservative care UK patients with low back pain Manipulation alone plus conservative care versus conservative care UK patients with low back pain Exercise alone plus conservative care versus manipulation and exercise plus conservative care UK patients with low back pain Early cross-sectional imaging versus delayed imaging British patients with low back pain presenting to spine surgeon Usual care plus treatment with osteopathy versus usual care alone British patients with mechanical neck pain and upper or low back pain MRI versus lumbar radiographs to exclude cancer as cause of low back pain UK patients with low back pain referred for imaging Laminectomy with instrumented fusion versus laminectomy with noninstrumented fusion US patients with spondylolisthesis and spinal stenosis Laminectomy with noninstrumented fusion versus laminectomy without fusion US patients with spondylolisthesis and spinal stenosis Surgical discectomy versus conservative therapy in patients with herniated intervertebral disc Chemonucleolysis versus surgical discectomy adults with diagnosis of sciatica with clinical signs of lumbar disc herniation

Cost saving $15,000

$6,700

Dominated

$2,600

$6,800 C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

$360,000

$4,200,000

$67,000

$51,000 Cost saving

QALY, quality-adjusted life year; RCT, randomized controlled trial; ICBG, iliac crest bone graft; VAS, visual analog scale; rHuEPO, recombinant human erythropoietin; PAD, preoperative autologous donation; SCS, spinal cord stimulation; FBSS, failed back surgery syndrome; HRQOL, health-related quality of life; LBP, low back pain; MRI, magnetic resonance imaging. * Cost-utility ratio data as reported by the cost-effectiveness analysis registry, in 2010 US dollars [4].

683

684 Cost saving Cost saving

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

modalities with high cost. The cost/QALY calculation will likely play a large role in determining whether health care interventions are covered by payers, including the government, in the future. An important consideration in the critical analysis of cost-effectiveness studies is the study perspective. As cited previously, the US Panel on Cost-Effectiveness in Health and Medicine has recommended the use of a societal perspective for cost-effectiveness studies [1], reecting a belief that indirect costs related to work absenteeism of both the patient and unpaid care providers warrant consideration in cost-effectiveness decisions. In contrast, the NHS has traditionally used a payer perspective for making decisions about funding health care interventions, a bias that is reected in the many cost-effectiveness studies in the United Kingdom performed with government funding. Analysis of cost-effectiveness from a payers perspective is easier to perform as it avoids the need to track patient salary and work schedule or make assumptions about indirect costs for nonworking patients, as is often done when calculating indirect costs. CUAs provide information (ie, cost/QALY) that can be used by the stakeholders in health care delivery to compare the value of interventions across the various disciplines of medicine. Despite the potential signicance of cost/QALY data to policy and coverage decision making, there is a relative lack of such data in the existing literature. A previous systematic review of cost-effectiveness studies in orthopedic surgery between 1976 and 2001 identied only ve studies pertaining to spinal care [2]. This study highlighted the wide variation in methodology that is used to perform CEA and the lack of cost-effectiveness data that exist for spinal care. The study did not provide the specic data or cost-utility ratios derived from these ve studies. An updated review of CEA in orthopedic surgery that searched the literature through 2003 was published in 2007 by Brauer et al. [6]. From 1976 to 2003, the authors identied a total of 52 studies that included a cost/QALY analysis, 13 of which pertained to spine (25%). Like the initial publication, however, this more recent study did not provide any specic data regarding the spine CEA studies. Dagenais et al. [3] similarly identied only 15 studies related to low back pain in 2009. Most studies addressed nonoperative intervention for low back pain, such as education and exercise therapy. The cost per QALY for the interventions included in that study ranged from $304 to $579,527, and the average utility improved from a baseline of 0.57 to 0.67 after the intervention. The current analysis shows that there has been a signicant increase in the number of cost-effectiveness studies published on topics of spinal care over the past 5 years and includes a wider range of studies to provide a comprehensive picture of the current body of literature on comparative effectiveness across a variety of interventions for spinal disease. We acknowledge that our study included only English-language articles, a limitation of our study that is

$40,000

Table 2 Cost-utility ratios specic to the treatment of lumbar disc herniation, comparing operative with nonoperative treatment*

Follow-up period (mo)

24

12

Likert rating

24

24

Markov model QALY, quality-adjusted life year; RCT, randomized controlled trial; VAS, visual analog scale. * Cost-utility ratios as reported by the cost-effectiveness analysis registry, in 2010 US dollars [4]. Payer 5 Based on literature 4.5

Utility measure

Not used

Discount rate (%)

Perspective (payers vs. society)

Society

Society

Society

Tosteson et al. [15]

Study

van den Hout et al. [16] Hansson et al. [17] Malter et al. [18]

Society

EQ5D, SF6D, VAS EQ5D

EQ5D

EQ5D

Prospective case-control Decision tree model

Study design

RCT

RCT

RCT

3.5

Standard open discectomy versus nonoperative care US Medicare patients with lumbar disc herniation Standard open discectomy versus nonoperative care US general patients with lumbar disc herniation Early surgery versus prolonged conservative care Dutch patients with disc herniation, radicular symptoms Surgical treatment versus nonsurgical treatment Swedish patients with lumbar disc herniation Surgical discectomy versus conservative therapy in patients with herniated intervertebral disc

Comparison and population

$80,000

Cost/ QALY

$51,000

Table 3 Cost-utility ratios specic to treatment of stenosis and degenerative spondylolisthesis* Perspective (payers vs. society) Society Discount rate (%) 3 Utility measure EQ5D Study design RCT Likert rating 6.5 Follow-up period (mo) 24 Cost/ QALY $120,000

Study Tosteson et al. [14]

Comparison and population Circumferential (360  ) fusion surgery versus nonoperative care US patients with spinal stenosis and spondylolisthesis Posterolateral fusion surgery versus nonoperative care US patients spinal stenosis with degenerative spondylolisthesis Instrumented fusion surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Noninstrumented fusion surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Fusion surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Decompression surgery versus nonoperative care US patients with spinal stenosis with degenerative spondylolisthesis Decompression surgery versus nonoperative care US patients with spinal stenosis without spondylolisthesis Fusion surgery versus nonoperative care US patients with spinal stenosis without spondylolisthesis Surgery for stenosis with degenerative spondylolisthesis versus nonoperative care US patients with spinal stenosis Spine surgery for spinal stenosis versus nonoperative care US patients with spinal stenosis Laminectomy with instrumented fusion versus laminectomy with noninstrumented fusion US patients with spondylolisthesis and spinal stenosis Laminectomy with noninstrumented fusion versus laminectomy without fusion US patients with spondylolisthesis and spinal stenosis

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$140,000 C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$140,000

Society

EQ5D

RCT

6.5

24

$34,000

Society

EQ5D

RCT

6.5

24

$55,000

Society

EQ5D

RCT

6.5

24

$190,000

Society

EQ5D

RCT

6.5

24

$130,000

Society

EQ5D

RCT

6.5

24

$90,000

Kuntz et al. [12]

Payer

Not used

Based on literature Based on literature

Decision tree model Decision tree model

Decision tree model Decision tree model

$4,200,000

Payer

Not used

$67,000

QALY, quality-adjusted life year; RCT, randomized controlled trial. * Cost-utility ratios as reported by the cost-effectiveness analysis registry, in 2010 US dollars [4]. 685

686

Table 4 Cost-utility studies and ratios published on the cervical spine* Perspective (payers vs. society) Payers Society Payer Discount rate (%) 5 Not used in base case Not used Utility measure HRQOL SF6D EQ5D Follow-up period (mo) 3 3 12

Study Rsnen et al. aa [19] Willich et al. [21] Manca et al. [48]

Study design Prospective cohort study RCT RCT

Likert rating 2 5 4.5

Comparison and population Cervical spine decompression versus no intervention Finnish patients with cervical spine radiculopathy Acupuncture versus conservative care German patients with chronic neck pain Usual physiotherapy versus brief physiotherapy intervention UK patients O18 years old with neck pain of more than 2 weeks Single-level ACDFP versus ACDF with allograft US patients with cervical spondylosis Single-level ACDF with allograft versus ACDF with autograft US patients with cervical spondylosis CT versus radiography US trauma patients with low risk of cervical spine fracture CT versus radiography in trauma US trauma patients with moderate risk of cervical spine fracture CT versus radiography in trauma US trauma patients with high risk of cervical spine fracture

Cost/QALY $4,000 $18,000 $120,000 C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

Angevine et al. [20]

Payer Payer

3 3

Based on literature Based on literature Quality of life estimates Quality of life estimates Quality of life estimates

Decision tree model Decision tree model Retrospective cohort Retrospective cohort Retrospective cohort

3.5 3.5

Decision tree model Decision tree model Decision tree model Decision tree model Decision tree model

$41,000 $630

Blackmore et al. [22]

Payer

5.5

$120,000

Payer

5.5

$24,000

Payer

5.5

Cost saving

QALY, quality-adjusted life year; HRQOL, health-related quality of life; RCT, randomized controlled trial; ACDFP, anterior cervical discectomy and fusion with plating; ACDF, anterior cervical discectomy and fusion; CT, computed tomography. * Cost-utility ratios as reported by the cost-effectiveness analysis registry, in 2010 US dollars [4].

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

687

Fig. 2. Quality-adjusted life years are calculated as the area under the curve of utility versus time. This graph represents the utility over time of an intervention that has durable outcomes, that is, the utility is maintained over time. With a durable outcome, the QALY (area under the curve) increases with time. Assuming that the costs do not increase substantially (ie, patient does not require additional treatment because there is a lasting good outcome), the cost/QALY will decrease as the length of follow-up increases and utility is maintained. This scenario represents an intervention in which the value will increase over time. QALY, quality-adjusted life year.

a reection of a limitation of the databases we used in our search. The treatment of lumbar degenerative conditions is a specic area for which cost-utility data exist. Several studies comparing operative and nonoperative treatment of lumbar disc herniation report on the cost-effectiveness of this intervention, with reported cost-utility ratios ranging from cost saving (ie, the intervention had a negative cost/ QALY gain) to $80,000/QALY gain (2010 US dollars) (Table 2). Thus, based on the current literature, it can be concluded that there is sufcient evidence that decompressive surgery for lumbar disc herniation is a cost-effective treatment using a cost threshold of $100,000. Cost-utility data on the treatment of lumbar stenosis and degenerative spondylolisthesis support the cost-effectiveness of surgical treatment of lumbar stenosis. The cost-effectiveness of fusion for degenerative spondylolisthesis (instrumented or noninstrumented) remains a matter of debate. Most of the data

Fig. 3. Quality-adjusted life years are calculated as the area under the curve of utility versus time. This graph represents the utility over time of an intervention with outcomes that decline over time. The QALY is not as high as in Fig. 2. As outcomes decline over time, it is likely that costs will increase as the patient needs additional treatment. This scenario represents an intervention in which the value will decrease over time. QALY, quality-adjusted life year.

on these conditions are derived from the Spine Patient Outcomes Research Trial (SPORT). In 2008, Tosteson et al. [14] published a CUA of the treatment of stenosis and degenerative spondylolisthesis based on the SPORT database. These authors calculated an incremental cost-effectiveness ratio for the treatment of both these conditions at 2-year follow-up and used Medicare maximum-allowable payment amounts to represent direct costs of treatment, a costing strategy that can be used if the entire study cohort is Medicare eligible. Comparing surgical with nonsurgical treatment, the incremental cost-effectiveness ratio is dened as (cost surgerycost nonsurgical treatment)/(QALY gain by surgeryQALY gain by nonoperative treatment). These authors found that surgical treatment for lumbar stenosis had a cost/QALY gain of $90,000 compared with nonsurgical treatment (2010 US dollars). Surgical treatment for degenerative spondylolisthesis had a cost/QALY gain of $130,000 compared with nonsurgical treatment (2010 US dollars) [14]. These were the primary costutility ratios reported; however, several other secondary costutility ratios were reported that look at various subgroups, for example, type of fusion (posterolateral, circumferential, instrumented, noninstrumented, and so forth) (Table 3). Contradictions exist in the literature regarding the costeffectiveness of fusion in the setting of degenerative spondylolisthesis. In 2000, using a Markov model, Kuntz et al. [12] reported on the cost-utility of instrumented and noninstrumented fusion in patients with degenerative lumbar spondylolisthesis and stenosis. The authors reported that performing laminectomy and noninstrumented fusion led to a cost/ QALY gain of $67,000 (2010 US dollars) in comparison with performing laminectomy alone. The cost/QALY for instrumented fusion exceeded that for noninstrumented fusion by $4,200,000 (2010 US dollars), suggesting that the addition of instrumentation to a fusion procedure is not cost effective. It should be noted, however, that this type of analysis is based on numerous assumptions. The initial analysis comparing instrumented and noninstrumented fusion was based on of the assumption that these procedures have similar clinical outcomes, although the instrumented fusion group had a higher fusion rate. If, instead, it is assumed that the instrumented fusion group has a 10% greater improvement in clinical outcomes compared with the noninstrumented group, the cost/QALY was reduced from $4,200,000 to $111,180 (2010 US dollars). If a difference in clinical outcome of 15% in favor of the instrumented group is assumed, the cost/QALY of instrumented fusion decreases further to $73,400 (2010 US dollars). This sensitivity analysis demonstrates just how much small changes in the assumptions used in a decision model analysis can affect the outcome. In 2004, Kornblum et al. [24] reported on a series of patients who underwent decompression and posterolateral fusion for single-level lumbar stenosis and degenerative spondylolisthesis with an average follow-up of 7 years (range, 514 years). These authors found that patients who developed a solid fusion had better clinical outcomes than those who developed a pseudarthrosis. These data, along with

688

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

the published data that show higher fusion rates with the use of instrumentation compared with noninstrumented fusion, suggest that long-term outcomes may be improved with instrumented fusion. As described previously, such differences in clinical outcome would likely, according to the sensitivity analysis performed by Kuntz et al. [12], make instrumented fusion a cost-effective treatment. Unlike the decision model analysis by Kuntz et al. [12], the study by Tosteson et al. [14] was based on the prospectively collected data (ie, the SPORT database). These authors found that fusion for the treatment of degenerative spondylolisthesis was not cost effective, with a cost/QALY gain over nonsurgical treatment of $130,000 (2010 US dollars). None of the specic types of fusion studied (ie, instrumented, noninstrumented, posterolateral, and circumferential) were found to be cost effective. It should be noted that these cost-utility data are based on the 2-year follow-up SPORT data. The 4-year follow-up clinical data for degenerative spondylolisthesis have recently been published, demonstrating that the benet of fusion surgery over nonsurgical treatment seen at 2-year follow-up is present at 4 years as well [25]. This durability of the surgical treatment effect at 4-year follow-up will result in a decrease in the cost/QALY gain for fusion surgery, as no additional costs are accrued but the QALY increases as improvements in outcome are maintained over time (Figs. 2 and 3). The 4-year follow-up cost-effectiveness data have not yet been published, but the concept is true for all treatments with a durable result longer follow-up will lead to more-favorable appearing cost/QALY as the constant cost is divided by a larger total QALY gain through continued clinical benet. Many of the lumbar spine cost-utility studies compare nonoperative forms of treatment (Table 1). Although many of the cost-effectiveness ratios for these interventions are well below the arbitrary $100,000 threshold, there are almost as many different types of disparate interventions as studies. Studies advocating acupuncture, nurse-based follow-up, physiotherapist-led pain therapy, inpatient rehabilitation, use of a particular advice book, osteopathy, and tobacco/alcohol cessation counseling are just some of the therapeutic interventions that have been studied. Although it appears that many of these interventions may provide benet to patients with low back pain at favorable costeffectiveness ratios, it is difcult to make recommendations regarding the best nonoperative intervention, given the heterogeneity of treatments studied. Most cost-utility studies focus on the lumbar spine. Cost-utility data on degenerative cervical conditions, scoliosis, and spinal trauma are lacking in the literature. The few studies that do address cervical spinal care address a variety of topics (ie, spondylosis and radiculopathy, fusion techniques, imaging, and nonoperative care). The only study that addresses the cost utility of surgery for cervical radiculopathy is somewhat limited. In 2006, Rsnen et al. aa [19] studied the cost-effectiveness of surgical treatment for both lumbar and cervical degenerative conditions. There

were 169 patients who underwent surgery for cervical disc herniation and/or spondylosis. The cost/QALY gain for surgical treatment of cervical disorders was $4,000 (2010 US dollars). This study was limited, however, in that there was not a comparison group with surgery, the results were based on only 3-month follow-up, and CUA was not performed from the societal perspective and only included direct costs. The body of the literature describing cost-effectiveness of spinal surgery remains relatively small and highly heterogeneous with respect to research methodology, a feature that severely limits the ability to draw robust conclusions regarding cost-effectiveness of spinal procedures and precludes basing policy decisions on patient-derived outcome measures. Based on our review of this fragmented literature, we have several recommendations for authors interested in performing future cost-effectiveness research. Cost-effectiveness studies should present patient-derived outcome data at a minimum of 2 years after enrollment. The 2-year threshold has become the standard minimum length of follow-up required for clinical research in reconstructive subspecialties; this should also hold for cost-effectiveness research. Although longer periods of follow-up may be necessary to demonstrate cost-effectiveness for more-expensive interventions or those with frontloaded costs, we feel that publication of investigations with time horizons of less than 2 years should be deferred until at least 2-year follow-up is reached. Treatments with a more durable clinical result will become more and more attractive from a cost-effectiveness standpoint with longer time horizons. Future cost-effectiveness research should analyze costs from a societal perspective meaning that indirect costs must be considered. The issue of how to address cost is the single most complex issue facing researchers seeking to perform high-quality CEA. Charges are much easier to track than costs and are relatively transparent but are often a poor reection of health care resource utilization. In contrast, costs are difcult to calculate because of considerations such as unclear divisions of shared resources, variations in accounting systems, and a lack of transparency for the cost of services and goods outsourced to vendors. This distinction is important in evaluating CUA studies in the literaturethe use of charges instead of costs limits the ability to generalize a studys results and may misrepresent resource utilization. Strategies such as cost-to-charge ratio and microcosting are strategies that attempt to facilitate the use of costs instead of charges. Cost-to-charge ratio combines hospitalaccounting data collected by the Center for Medicare/ Medicaid Services and hospital charges for services to determine a cost-to-charge ratio that is both hospital- and yearspecic. In contrast, microcosting tries to capture costs associated with care of a single patient through carefully tracking each resource the patient consumes and then determining the local cost of that resource. These two types of cost analysis are very different and each has strengths and weaknesses. We endorse establishing a consensus position regarding cost-accounting methodology with a large group of cost-effectiveness experts to standardize practices; balancing

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690

689

ease of use and comprehensive cost accuracy is essential to establishing a cost accounting convention that is widely accepted and used.

Conclusions Only 33 cost-utility studies and 60 cost-utility ratios have been published on various aspects of spinal care over the last 30 years, the majority of which have been published in the last 5 years. Certain aspects of spinal care, including various nonoperative treatments, the operative treatment of lumbar disc herniation, and lumbar stenosis have been shown to be cost effective, with a cost/QALY gain of less than $100,000. There remains a debate as to the cost-effectiveness of fusion for the treatment of lumbar stenosis and degenerative spondylolisthesis. Available data on fusion for degenerative spondylolisthesis suggest that this intervention provides better outcomes than nonoperative treatment with 4-year follow-up. Although fusion has not been shown to be cost-effective at 2 years, recalculation of the cost/ QALY based on the favorable 4-year data may show costeffectiveness. Certainly, further efforts are needed to better dene the value of many aspects of spinal care.

References
[1] Weinstein MC, Siegel JE, Gold MR, et al. Recommendations of the panel on cost-effectiveness in health and medicine. JAMA 1996;276: 12538. [2] Brauer CA, Rosen AB, Olchanski NV, Neumann PJ. Cost-utility analyses in orthopaedic surgery. J Bone Joint Surg Am 2005;87:12539. [3] Dagenais S, Roffey DM, Wai EK, et al. Can cost utility evaluations inform decision making about interventions for low back pain? Spine J 2009;9:94457. [4] The CEA Registry. Available at: https://research.tufts-nemc.org/ cear4/SearchingtheCEARegistry/SearchtheCEARegistry.aspx. Accessed September 29, 2010. [5] Neumann PJ, Greenberg D, Olchanski NV, et al. Growth and quality of the cost-utility literature, 1976-2001. Value Health 2005;8:39. [6] Brauer CA, Neumann PJ, Rosen AB. Trends in cost effectiveness analyses in orthopaedic surgery. Clin Orthop Relat Res 2007;457: 428. [7] Greenberg D, Earle C, Fang CH, et al. When is cancer care costeffective? A systematic overview of cost-utility analyses in oncology. J Natl Cancer Inst 2010;102:828. [8] Stone PW, Schackman BR, Neukermans CP, et al. A synthesis of cost-utility analysis literature in infectious disease. Lancet Infect Dis 2005;5:38391. [9] Talmor D, Shapiro N, Greenberg D, et al. When is critical care medicine cost-effective? A systematic review of the cost-effectiveness literature. Crit Care Med 2006;34:273847. [10] Otero HJ, Rybicki FJ, Greenberg D, Neumann PJ. Twenty years of cost-effectiveness analysis in medical imaging: are we improving? Radiology 2008;249:91725. [11] University of York, Centre for Reviews and Dissemination. http:// www.york.ac.uk/inst/crd/intertasc/nhs_eed_strategies.html. Accessed May 24, 2011. [12] Kuntz KM, Snider RK, Weinstein JN, et al. Cost-effectiveness of fusion with and without instrumentation for patients with degenerative spondylolisthesis and spinal stenosis. Spine 2000;25:11329.

[13] Soegaard R, Christensen FB, Christiansen T, Bunger C. Costs and effects in lumbar spinal fusion. A follow-up study in 136 consecutive patients with chronic low back pain. Eur Spine J 2007;16:65768. [14] Tosteson AN, Lurie JD, Tosteson TD, et al. Surgical treatment of spinal stenosis with and without degenerative spondylolisthesis: costeffectiveness after 2 years. Ann Intern Med 2008;149:84553. [15] Tosteson AN, Skinner JS, Tosteson TD, et al. The cost effectiveness of surgical versus nonoperative treatment for lumbar disc herniation over two years: evidence from the Spine Patient Outcomes Research Trial (SPORT). Spine 2008;33:210815. [16] van den Hout WB, Peul WC, Koes BW, et al. Prolonged conservative care versus early surgery in patients with sciatica from lumbar disc herniation: cost utility analysis alongside a randomised controlled trial. BMJ 2008;336:13514. [17] Hansson E, Hansson T. The cost-utility of lumbar disc herniation surgery. Eur Spine J 2007;16:32937. [18] Malter AD, Larson EB, Urban N, Deyo RA. Cost-effectiveness of lumbar discectomy for the treatment of herniated intervertebral disc. Spine 1996;21:104854; discussion 1055. [19] Rsnen P, Ohman J, Sintonen H, et al. Cost-utility analysis of aa routine neurosurgical spinal surgery. J Neurosurg Spine 2006;5: 2049. [20] Angevine PD, Zivin JG, McCormick PC. Cost-effectiveness of singlelevel anterior cervical discectomy and fusion for cervical spondylosis. Spine 2005;30:198997. [21] Willich SN, Reinhold T, Selim D, et al. Cost-effectiveness of acupuncture treatment in patients with chronic neck pain. Pain 2006;125:10713. [22] Blackmore CC, Ramsey SD, Mann FA, Deyo RA. Cervical spine screening with CT in trauma patients: a cost-effectiveness analysis. Radiology 1999;212:11725. [23] National Institute for Health and Clinical Excellence. Guide to the methods of technology appraisal. London: National Institute for Health and Clinical Excellence, 2008. [24] Kornblum MB, Fischgrund JS, Herkowitz HN, et al. Degenerative lumbar spondylolisthesis with spinal stenosis: a prospective longterm study comparing fusion and pseudarthrosis. Spine 2004;29: 72633; discussion 733724. [25] Weinstein JN, Lurie JD, Tosteson TD, et al. Surgical compared with nonoperative treatment for lumbar degenerative spondylolisthesis. Four-year results in the Spine Patient Outcomes Research Trial (SPORT) randomized and observational cohorts. J Bone Joint Surg Am 2009;91:1295304. [26] Lamb SE, Hansen Z, Lall R, et al. Group cognitive behavioural treatment for low-back pain in primary care: a randomised controlled trial and cost-effectiveness analysis. Lancet 2010;375:91623. [27] Carreon LY, Glassman SD, Djurasovic M, et al. rhBMP-2 versus iliac crest bone graft for lumbar spine fusion in patients over 60 years of age: a cost-utility study. Spine 2009;34:23843. [28] Smeets RJ, Severens JL, Beelen S, et al. More is not always better: costeffectiveness analysis of combined, single behavioral and single physical rehabilitation programs for chronic low back pain. Eur J Pain 2009;13:7181. [29] Nielsen PR, Andreasen J, Asmussen M, Tonnesen H. Costs and quality of life for prehabilitation and early rehabilitation after surgery of the lumbar spine. BMC Health Serv Res 2008;8:209. [30] van der Roer N, van Tulder M, van Mechelen W, de Vet H. Economic evaluation of an intensive group training protocol compared with usual care physiotherapy in patients with chronic low back pain. Spine 2008;33:44551. [31] Hollinghurst S, Sharp D, Ballard K, et al. Randomised controlled trial of Alexander technique lessons, exercise, and massage (ATEAM) for chronic and recurrent back pain: economic evaluation. BMJ 2008;337: a2656. [32] Critchley DJ, Ratcliffe J, Noonan S, et al. Effectiveness and costeffectiveness of three types of physiotherapy used to reduce chronic low back pain disability: a pragmatic randomized trial with economic evaluation. Spine 2007;32:147481.

690

C.K. Kepler et al. / The Spine Journal 12 (2012) 676690 [41] Rivero-Arias O, Campbell H, Gray A, et al. Surgical stabilisation of the spine compared with a programme of intensive rehabilitation for the management of patients with chronic low back pain: cost utility analysis based on a randomised controlled trial. BMJ 2005;330:1239. [42] Taylor RJ, Taylor RS. Spinal cord stimulation for failed back surgery syndrome: a decision-analytic model and cost-effectiveness analysis. Int J Technol Assess Health Care 2005;21:3518. [43] United Kingdom back pain exercise and manipulation (UK BEAM) randomised trial: cost effectiveness of physical treatments for back pain in primary care. BMJ 2004;329:1381. [44] Gilbert FJ, Grant AM, Gillan MG, et al. Low back pain: inuence of early MR imaging or CT on treatment and outcomemulticenter randomized trial. Radiology 2004;231:34351. [45] Williams NH, Edwards RT, Linck P, et al. Cost-utility analysis of osteopathy in primary care: results from a pragmatic randomized controlled trial. Fam Pract 2004;21:64350. [46] Hollingworth W, Gray DT, Martin BI, et al. Rapid magnetic resonance imaging for diagnosing cancer-related low back pain. J Gen Intern Med 2003;18:30312. [47] Launois R, Henry B, Marty JR, et al. Chemonucleolysis versus surgical discectomy for sciatica secondary to lumbar disc herniation. A cost and quality-of-life evaluation. Pharmacoeconomics 1994;6:45363. [48] Manca A, Epstein DM, Torgerson DJ, et al. Randomized trial of a brief physiotherapy intervention compared with usual physiotherapy for neck pain patients: cost-effectiveness analysis. Int J Technol Assess Health Care 2006;22:6775.

[33] Johnson RE, Jones GT, Wiles NJ, et al. Active exercise, education, and cognitive behavioral therapy for persistent disabling low back pain: a randomized controlled trial. Spine 2007;32:157885. [34] Soegaard R, Bunger CE, Christiansen T, et al. Circumferential fusion is dominant over posterolateral fusion in a long-term perspective: cost-utility evaluation of a randomized controlled trial in severe, chronic low back pain. Spine 2007;32:240514. [35] Freeman BJ, Steele NA, Sach TH, et al. ISSLS prize winner: costeffectiveness of two forms of circumferential lumbar fusion: a prospective randomized controlled trial. Spine 2007;32:28917. [36] Vitale MA, Vitale MG, Zivin JG, et al. Rotator cuff repair: an analysis of utility scores and cost-effectiveness. J Shoulder Elbow Surg 2007;16: 1817. [37] North RB, Kidd D, Shipley J, Taylor RS. Spinal cord stimulation versus reoperation for failed back surgery syndrome: a cost effectiveness and cost utility analysis based on a randomized, controlled trial. Neurosurgery 2007;61:3618; discussion 368369. [38] Rivero-Arias O, Gray A, Frost H, et al. Cost-utility analysis of physiotherapy treatment compared with physiotherapy advice in low back pain. Spine 2006;31:13817. [39] Ratcliffe J, Thomas KJ, MacPherson H, Brazier J. A randomised controlled trial of acupuncture care for persistent low back pain: cost effectiveness analysis. BMJ 2006;333:626. [40] Schweikert B, Jacobi E, Seitz R, et al. Effectiveness and costeffectiveness of adding a cognitive behavioral treatment to the rehabilitation of chronic low back pain. J Rheumatol 2006;33:251926.