How should value be defined in spine surgery?

Introduction

In a health care economy with limited resources, providers and consumers of health care services need to be accountable for the end result and the cost of care. The value proposition in health care is an analysis of the benefits of care relative to the direct cost and risk of providing the care. Measurement of benefits and costs is challenging, and a consensus on the measures that encompass the relevant components of the value equation has not been reached. Traditional outcome measures in orthopedics including survival, radiographic outcomes, and disease-specific outcome tools do not adequately reflect the patient’s health care experience, or the impact of an intervention on health-related quality of life. Similarly, measuring cost of care is complex, and may encompass both direct costs of treatment and alternative treatments, and indirect costs including time from work or family role, loss of productivity, and cost of caretakers.

The value equation may vary depending on the perspective of the stakeholder in the health care economy. Hospitals and facilities providing care may measure outcome and costs by factors that affect their short-term, single admission interaction, including length of stay, implant utilization, and complications. Third-party payors may focus on a timeframe that is longer that a single admission, and may include factors in the value equation such as readmission within 90 days, or cost of outpatient care. Hospital- and payor-based quality measures may be misinterpreted as measures of outcome or value. Length of stay, surgical times, compliance with antibiotic or thromboembolic prophylaxis, and perioperative complications are process measures that may be useful to compare hospital and provider performance when appropriately matched and stratified. However, they are not useful in measuring a patient’s health care experience, or the impact of an intervention on long-term health-related quality of life. In fact, a focus on quality and process measures alone may be misleading in the pursuit of value in health care, and may provide incentive for counterproductive care strategies that serve the measurement system rather than the patient.

The health care provider and the patient measure outcome by the impact of an orthopedic intervention on health-related quality of life (HRQoL). The timeframe for the patient and provider is a lifetime, rather than a single admission. As Porter and Teisberg wrote in 2006, the right goal for health care delivery is superior patient value, which is measured at the level of specific medical conditions. Measurement of outcomes of care needs to reflect the patient’s long-term health care experience, and the impact of one intervention compared with alternatives on the patients self-assessment of HRQoL.

In orthopedic surgery, patient self-assessment of health status and quality of life may include disease-specific and general health status measures. Disease-specific measures are useful to optimize responsiveness to change, and psychometric properties. General health status outcome tools are useful in providing a measure of outcome that may be translatable across a spectrum of medical and surgical conditions. The priority of the patient’s self assessment of the impact of care on his or her long-term health status requires emphasis, and efforts to substitute process measures for health status measures need to be avoided in the pursuit of a value-based health care system. The purpose of this Round Table discussion is to address the measures of value that are currently used in orthopedics and spine surgery, and to provide direction for the orthopedic community to measure outcomes that are useful in leading to an evidence-based and cost-effective approach to care.

Sigurd Berven, MD
Moderator

Round Table Participants Moderator Sigurd Berven, MD Director of Spine Fellowship and Resident Education Department of Orthopaedic Surgery University of California San Francisco San Francisco, Calif. Todd J. Albert, MD Thomas Jefferson University Richard H. Rothman Professor Chairman, Department of Orthopaedic Surgery Thomas Jefferson University and Hospitals Philadelphia, Pa. Gunnar Andersson, MD, PhD Professor and Chairman Emeritus The Ronald L. DeWald, MD, Chair in Spinal Deformities Rush University Medical Center Chicago, Ill. Steven D. Glassman, MD Professor of Orthopedic Surgery University of Louisville School of Medicine Attending Surgeon Norton Leatherman Spine Center Co-Chair, Professional Societies Coalition Task Force on Lumbar Fusion Louisville, Ky. Frank M. Phillips, MD Professor, Orthopedic Surgery Rush University Medical Center Chicago, Ill. James N. Weinstein, DO, MS President, Dartmouth-Hitchcock Clinic Director, the Dartmouth Institute for Health Policy and Clinical Practice Dartmouth-Hitchcock Medical Center and Dartmouth Medical School Lebanon, N.H.

Sigurd Berven, MD:How do we measure quality and value in orthopedic interventions?

Todd J. Albert, MD: There are many ways to measure the value. Presently most use process measures such as blood loss, time of operation, rate of pulmonary embolism, readmission to hospital and revision surgery. Some people are appropriately collecting disease-specific and generic health outcome measures from which utility scores can be calculated.

Gunnar Andersson, MD, PhD: The classic clinical measures of quality are clinical outcome and avoidance of complications. Value however, means different things to different players in the health care arena. To the patient, process quality is very important. Easy access, rapid communication, a clean and efficient office, short wait times, pleasant staff, etc. Sometime process is even more important than quality outcome because it is difficult for the patient to evaluate differences in outcomes. Cost is often not an important factor in the eyes of the patient unless there is a direct out-of-pocket payment. To the third-party payor, value includes other aspects. While they want to increase enrollment which means that their customers have to be happy with both the process and outcome, cost is critical. This means that if the cost is higher there has to be benefit. In many cases the buyer of the insurance services is an employer. To the employer the satisfaction of the employee is critical and may justify a higher cost.

Steven D. Glassman, MD: In reality, quality is a multi-factorial parameter which might be characterized by objective measures such as improved function or low complication rate and by subjective measures such as the patient’s perception of pain. Despite the desire for a completely objective measure of quality, patient-based health status measures such as the SF-36 and Oswestry Disabilities Index (ODI) have gained increasing acceptance as the defacto measure of quality for orthopedic interventions. Regardless of the measure or group of measures selected, the concept of quality is one with which orthopedic surgeons are inherently familiar and comfortable.

On the other hand, value implies benefit or improvement in quality for a given cost. In general, the concept of value has not been something that orthopedic surgeons have effectively measured or documented in the past. However, the reality of today’s health care economics is changing that mindset.

Frank M. Phillips, MD: The value of an orthopedic intervention is defined as the quality of that intervention divided by the cost of the intervention over a defined period of time. Both quality and cost are extremely challenging to measure. In addition, if interventions have a lasting effect that reduces the need for subsequent utilization of health care resources, the value of the intervention will improve over time. The various stakeholders may utilize differing measures of quality including: patient-centered outcome measures (disease specific or general health measures); process measures (such as Physicians Quality Reporting Initiative) which are easy to measure but are a poor surrogate for quality; and safety measures. The cost component of the value equation should ideally take into account both direct costs, which relate to the treatment and any complications as well as indirect costs such as productivity loss attributable to the disease state and intervention.

James N. Weinstein, DO, MS: Quality and value are the most important measures we can apply. Simply put, it is vital, not only given the future direction of health care and reimbursement, but to the patients we treat and our research base. In Spine Patient Outcomes Research Trial (SPORT) and in the Spine Center, we tracked (and continue to track) patient self-reported outcomes and/or an EQ5D tracked longitudinally. This is an essential ingredient that is missing in current analysis and reporting. In order to have the data to know what we are doing as surgeons has real value, ie quality over cost over time that is safe, we must include measures like these. Moreover, they need to be included in the electronic health record and any health information technology (HIT) standards that are implemented.

Berven: Quality and value are both important goals for health care provision, but the measures for each are distinct. Quality of care is measured by standardized processes of care, and a rate of compliance with those processes. Quality measures may include the presence of an electronic medical record, nurse-to-patient ratios, and rates of adherence to established perioperative protocols. Value of care is measured by an analysis of the patient’s self-assessment of the benefit of care over time relative to the cost of that benefit. An accurate assessment of the value of a specific intervention should account for the incremental benefit and cost of care compared to the alternative.

Berven: What should the spine community measure to demonstrate the effectiveness and value of spine surgery?

Albert: The spine community should be measuring both disease-specific and generic health outcome measures which can be converted to utility scores allowing cost effectiveness measures. Obviously important in this algorithm will be specific measurements of cost both direct cost of the intervention but also indirect cost such as time lost from employment or secondary cost necessary because of the specific disease of the patient. The instruments most useful to measure cost effectiveness and value will likely be: for the lumbar spine ODI (disease specific); SF36 (or SF-6D) and/or EQ5D (generic health outcome measures convertible to utility scores); for the cervical spine the Neck Disability Index will likely be the most useful disease specific measure.

Andersson: I would like to divide the answer to this question into patient-based outcome measures and cost-effectiveness measures. To some degree they relate because in order to study cost-effectiveness it is necessary to have patient based outcome measures however, patient-based outcome measures can be used without inclusion of any cost measures. In a Spine Focus published in Spine in December 2000, Claire Bombardier summarized the symposium recommendations as follows: For studies of general health status use the SF36, for studies of back- specific function use Roland-Morris or ODI; for studies of pain use the bodily pain scale of SF36 or the chronic pain grade scale; for studies of work disability use the work status scale, days off work or time to return to work; and for back-specific satisfaction use the satisfaction with care or satisfaction with treatment outcome scales. All of those are described in detail in the December 2000 Spine. The recommendations are still valid. The same issue also discusses health outcomes which are used to determine quality adjusted life years (QALYs), an important part of the cost-effectiveness analysis in which incremental cost is related to a change in health outcome. Scales often used to that purpose are EQ 5D and SF6D.

Glassman: Spine surgeons have made great strides in recent years in terms of demonstrating the effectiveness of surgical treatment for selective and well-defined pathologies. Substantial improvements in quality of life and reduction of disability have been documented using validated outcomes tools including the SF-36 and ODI for patients with disc herniation, spinal stenosis and spondylolisthesis.

Demonstrating cost effectiveness is more difficult, because surgeons don’t typically collect cost data, and rarely collect any data for the relevant alternative nonoperative treatments. health care economists rely on a measure of cost per QALY as the standard measure of clinical value. Nonetheless, both the SF-36 and the ODI can be converted to QALYs, facilitating the analysis of cost per QALY for any given intervention.

Phillips: Any tool for measuring outcome should be patient-centered, valid and reliable, practical to administer and the score and must measure treatment effects that are important to the patient. General health measures, such as SF-36, have the advantage of being comparable across disease states and therefore are appealing when comparative effectiveness is being studied. Disease specific questionnaires, such as ODI are narrower but are generally more responsive to the specific condition or treatment being studied. In today’s environment, both types of measures will likely be required to show the value of spinal surgery and also to allow for spinal treatments to be compared to other disease states and treatments.

Quality-adjusted life years are widely used as an effectiveness measure because it incorporates morbidity and mortality consequences in a single measure. A QALY is based on measures of patients’ valuations of different health states and outcomes, valued relative to one another on a scale from 0 to 1. When utility weights are multiplied by the time spent in the health state and expressed in years of perfect health, the resulting measure is a QALY. The two extremes of the scale are typically optimal health (assigned the value 1) and death (assigned the value 0). The payer and policy community have largely adopted QALYs as the currency of value determinations.

Weinstein: Value, as discussed prior involves far more than traditional clinical outcomes. We must consider – and as physicians and providers, we must fully understand — both comprehensive measures of value and cost-effectiveness. Our health care system cannot continue to be blind to the costs and cost-effectiveness of the care we provide. In SPORT, we used EQ5D for the cost-effectiveness analyses.

Berven: A consensus regarding outcome measures will be useful for the spine community in the establishment of registries and multicenter collaboration. Disease-specific measures of outcome are responsive to change in health status, and have face validity relative to the specific goals of care. General-health status measures provide the ability to translate health status and outcomes across different medical conditions, and provide utility scores which are the currency needed for measuring cost-effectiveness and value.

Transparency in shared patient-based outcome measures and in costs of care is critical in measuring value of care.

Berven:Which spinal disorders have evidence to support surgical approaches as a cost-effective option for care?

Albert: Thankfully because of the efforts from SPORT, and the cost-effectiveness evaluations that stemmed from that, we now know that surgical treatment for herniated nucleus pulposus, spinal stenosis without instability and spinal stenosis associated with a degenerative spondylolisthesis in the lumbar spine are all cost-effective. While the cost-effectiveness of degenerative spondylolisthesis surgery was not demonstrated at 2 years, it has been since shown at 4 years due to the durability of the results.

Andersson: Certainly there is evidence of cost-effectiveness for surgical treatment of disc herniations, spondylolisthesis and spinal stenosis. There are also cost-effectiveness studies that indicate positive results for lumbar and cervical fusion and lumbar and cervical disc replacement. A major problem in determining cost-effectiveness is that cost can be influenced by the surgeon’s use of technology and by the timing of the evaluation. Surgical cost depends on what is done at the time of the procedure and differences in outcome are often most obvious in the early postoperative period. For example, the main benefit of surgery for a herniated disc is the rapid recovery of the patient. A special case is spinal deformity in which an adolescent with scoliosis rarely has pain and loss of function and therefore cost-effectiveness is difficult to measure.

Glassman: Our best evidence is in the area where we most accurately define the diagnostic entities, that being disc herniation, spinal stenosis, and spondylolisthesis. While there is ample evidence that surgical treatment is efficacious for well-selected patients with these diagnoses, it is the SPORT data which provides the most compelling evidence for cost effectiveness. However, the extensive data set available through the SPORT trial, including a comparative nonoperative group, will not realistically be available across the range of interventions which require cost-effectiveness analysis. Therefore, it will probably be a simpler assessment of cost per QALY, generated through large cohorts or registry studies, which will be the workhorse of cost effectiveness analysis.

Phillips: Recent studies in the spinal literature have placed emphasis on assessing cost-effectiveness. Hansson and colleagues recently reported the quality of life (QoL) of 770 patients with various orthopaedic conditions. They reported that total hip replacement (THR) and to a slightly lesser extent total knee replacement (TKR) essentially normalized QoL. The greatest improvements in QoL were however seen after spine surgery. For example, spinal decompression and instrumented fusion for spinal stenosis with spondylolisthesis resulted in an increase in EQ-5D of 0.461 (the highest observed in the study), a difference of 0.398 for THR.

Recently SPORT has reported sustained benefit of discectomy for the treatment of herniated lumbar over nonsurgical treatment with an acceptable cost per QALY. The cost per QALY gained in the surgical group for a non-Medicare payer was $69,403 and was $34,355 for Medicare payment rates. Tosteson and colleagues noted these costs are within the range considered to support an intervention as cost-effective. They also reported an as-treated analysis of 1,235 patients with spinal stenosis with and without degenerative spondylolisthesis, at 2 years decompression for stenosis improved health to a greater extent than non-operative care (QALY gain 0.17) at a cost of $77, 600/ QALY gained. Decompression and fusion for spondylolisthesis improved health to a greater extent than non-operative care (QALY gain 0.23) at a cost of $115 600/ QALY gained. This cost/QALY is considered beyond a cost-effective threshold at 2 years, however if the value of surgery over nonsurgery continued over the longer-term, the cost per QALY gained would improve. Recently, Klazen and colleagues reported that vertebral augmentation was shown to be superior to nonsurgical care in the treatment of osteoporotic vertebral fractures at a cost of 22,685 Euros per QALY gained.

Weinstein: The SPORT study has shown that treatment for herniated disc, spinal stenosis, and spinal stenosis with spondylolisthesis are all cost-effective at 2 years and even more so at 4 years for surgery. This is a valuable contribution to shared decision-making and the ability to allow patients to make a truly informed choice. Decision aids will be an important part of the future in what we call preference-based decisions, eg, elective spine surgery.

Berven: Spinal disorders that involve neural compression and instability of the spinal column have good evidence for the cost-effectiveness of care. The SPORT data have demonstrated the efficacy of operative care over nonoperative care for disc herniation, spinal stenosis and degenerative spondylolisthesis. Surgery for tumors and trauma affecting the spinal cord or neural elements is clearly more effective than nonoperative care.

Berven: Which disorders require further study to demonstrate cost effectiveness?

Albert: These are: spinal fusion and/or disk replacement for painful degenerative disk disease of the lumbar spine; adult and pediatric spinal deformity surgery; and surgery for radiculopathy and myelopathy of the cervical spine – both fusion and disk replacement surgery. However there are early data demonstrating cost effectiveness associated with both spinal fusion and disk replacement surgery in the neck. This data needs to be further enhanced and validated.

Andersson: To me it’s not only an issue of disorders, but also an issue of the use of technology to accomplish a surgical result. In general, there is a paucity of good cost-effectiveness studies of common spinal disorders such as degenerative disc disease.

Glassman: Not surprisingly, those pathologies which are imprecisely defined lack evidence for either QoL improvement or cost effectiveness. In particular, the catch-all description of degenerative disc disease will never be effectively analyzed until it is broken down into more specific diagnostic indications for treatment. Within that broad characterization there are likely specific diagnostic entities for which treatment is cost-effective and others for which it is neither helpful nor cost effective.

Phillips: Inconsistent results from fusion for low back pain secondary to spinal degeneration that is non-responsive to conservative treatment have been reported. Until diagnostic specificity is enhanced allowing for improved patient selection, it remains unlikely that cost effectiveness will be shown for this intervention. With the emergence of motion sparring technologies such as total disc replacement (TDR), long term follow up will be required to determine whether there is a reduction in subsequent care and surgeries after TDR which would offset the additional procedural costs when compared to fusion. If this were true, this might suggest value and cost effectiveness of such an intervention over time.

Weinstein: Certainly, we need to address the longstanding need for analysis of the effectiveness of spinal fusion, particularly in light of its exponential increase as a surgical procedure. Recent literature has also raised serious questions about the effectiveness of vertebroplasty. Additionally, we need to study the effectiveness of kyphoplasty and artificial disc surgery. This information then needs to be incorporated into decision tools, so that patients can make an Informed Choice about their diagnostic and treatment options.

Berven: Pain of spinal origin, without nerve compression or instability required further investigation compared with nonoperative care. The effectiveness of surgery for pain without instability is dependent on the accuracy of identifying a pain generator, and diagnosis of a source of pain can be challenging in many degenerative spinal disorders. The role of instrumentation and circumferential fusion of the spine also requires further study to demonstrated cost effectiveness compared with alternative approaches to care. The timeframe for measuring outcome is important in considering cost effectiveness, because the durability of the effect size is an important variable in calculating cost effectiveness. Instrumentation and larger surgeries for deformity may only be cost effective in cases that involve more than 2 years follow-up- a time frame that is of clear importance to the patient and provider, but does not reach the radar of the hospital or the third-party payor.

Berven:How does the perspective of the stakeholder influence the assessment of value of interventions? Does the perspective of the patient/physician differ from the perspective of the hospital or the payor regarding an assessment of outcomes and value of care?

Albert: In a utopian world, the perspective of the stakeholder should not matter. However in the world we live in, the perspective of the stakeholder is very important. The needs of the disabled single mother with severe radiculopathy who is the sole support for her family is different than the value of a spinal surgery to the grandmother who cannot walk one block and therefore cannot spend time with her grandchildren or enjoy her retirement. Likewise this is the value of spinal surgery to the hospital managing the balance sheet and looking at the cost of spinal implants. In adult deformity surgery with limited resources, it is different from the patient with severe degenerative spinal stenosis and scoliosis who will have improved self-image and ambulatory ability from improving decompression and fusion and improved sagittal balance.

Glassman: While there are certainly differences in perspective as to the value of any intervention, it seems likely that the necessity for health care reform will inherently narrow those differences. For the patient, the reduction in pain and improvement in QoL may seem valuable at any cost. For the payer, expensive interventions may seem unjustified regardless of the clinical benefit which is obtained. Historically the role of the physician was to defend the patient’s access to care regardless to cost, but changes in that role seem inevitable. The reality of limited resources means that choices will have to be made; the only question is whether surgeons will guide those choices or whether they will be imposed externally. Clearly the presentation of effective treatment over ineffective treatment is in the interest of both the patient and the physician.

Phillips: Increasingly, the various stakeholders are demanding evidence that interventions provide value. As innovative and potentially more expensive technologies are developed, greater emphasis is being placed on showing that any additional benefit achieved is commensurate with the added cost. This value assessment is complicated by the fact that the incentives of physicians, payers and hospitals are often not aligned. For example, re-admissions for additional treatments can be quite expensive to the payers, yet will result in increased payment to the hospital. The value of an intervention may also be interpreted differently between payors. As an example, an intervention that rapidly returns an injured worker to productivity may be perceived as high value to a commercial payor, but in an elderly Medicare recipient the value of such an intervention may be less apparent.

For all stakeholders, determining value is critical because focusing solely on costs is shortsighted, ignoring the ultimate goal of medicine, ie to deliver better health to the patient and society, and usually results in cost shifting rather than cost saving.

Weinstein: The most important perspective is the patient’s. Shared decision making and informed choice matter, but physicians and health systems also need to break through the iron curtain that shields procedures from the cost effectiveness and value of those treatments. Transparency in sharing clinical and cost-effectiveness measures over time must become part of the treatment equation. Patient-reported outcomes are the missing link to measuring value. The patient should decide, with all parties understanding risks and benefits, clinical and cost effectiveness.

Patient-reported outcomes are the missing link to measuring value. The patient should decide, with all parties understanding risks and benefits, clinical and cost effectiveness.

Berven: The principle stakeholders in a health care episode are the patient, the physician, the hospital and the third-party payor. The patient and the physician have a personal relationship that is a long-term bond. Patient value is the right goal in pursuing value-based care. The patient’s experience over a lifetime is the relevant perspective for evaluation of an intervention, and short-term costs will be discounted dramatically by long-term follow-up. Unfortunately, current measures of quality and value are dissociated from patient value. Specifically, the hospital measures compliance with protocols and length of stay as outcome measures, and neither is remotely related to the patient’s experience of the impact of care on the patient over time. In some scenarios, readmission or complications may provide the hospital more revenue and value by hospital-based accounting. Similarly, the third-party payor does not have a personal or lasting relationship with the patient. Therefore, the payor may value short-term cost without regard to long-term effects. Physicians, hospitals and payors need to merge perspectives, and unite in measuring, and being accountable for patient-based, long-term assessments of outcome and value.

Berven: What are the next steps for the community of spine surgeons to pursue in defining value in spine surgery?

Albert: The next step is for the community of spine surgeons is:

• to pursue and define the value of spinal surgery – this will include collecting similar data sets prospectively and defining the most appropriate outcome measures to be collected;
• start to create a registry of data among the caregivers to ultimately evaluate the value of the interventions; and
• validate the ability to create utility scores from our collected health outcome measures and converting the collection of data to the demonstration and study of cost effectiveness of the interventions.

While there is pressure to do this on a shorter timeline, these steps are incredibly important for the ultimate successful analysis and the improvement of cost effectiveness and for the demonstration of cost effectiveness (or lack thereof) of our interventions especially when compared to other interventions outside the spine surgery arena.

Andersson: We need to do high-quality randomized controlled trials which include cost-effectiveness analyses. We are beginning to, but there is still an unmet need.

Glassman: The important steps for spine surgeons are the clear definition of the diagnostic entities we are treating and the collection of a uniform set of basic outcome measures for the evaluation of those treatments. Without those two basic data points we will be unable to generate the evidence required to demonstrate either quality or value for even our most standard interventions.

Phillips: The spine community must embrace the concept of defining value in spine surgery. If we fail to define the value of what we do, others with their own agendas will do so. Because of the complexity of measuring value, we have already seen regulators impose process measure data collection (such as recording the use of DVT prophylaxis as part of the PQRI) that provides little information about the value of spinal surgery. Measuring value will require spine surgeons in both academic and community settings to collect both outcome and cost data for precise diagnoses. A spinal surgery registry could provide impetus to this effort. The gamete of spinal care involves numerous physicians and treatments in addition to those provided by surgeons. Cumulatively, cost of treatments, such as medications, spinal injections, chiropractic care, and physical therapy far exceed those of spinal surgery and these modalities must also be studied to determine their value to patients and society.

Weinstein: We need to do more studies like SPORT that incorporate shared decision-making and longitudinal outcomes measurement in order to know we are doing what is best and most effective — and are consistent with the patient’s values. These studies are arduous, complex, and expensive, as SPORT demonstrated, but vital to our ability to practice evidence-based medicine. NIAMS (NIH) should be recognized for supporting this important work. Spine surgeons can and should continue to lead in this area. Comparative effectiveness should incorporate cost-effectiveness. We also need national data bases and registries.

Berven: Adopting a perspective of patient-based assessment of value is the most important goal for the spine community. Developing a consensus of outcomes tools that are shared across centers will facilitate multicenter studies. Disease-specific measures combined with the EQ-5D may provide the most useful information on outcomes. Shared access to radiographs and information on comorbidities will facilitate an appropriate classification and stratification of disorders. Comparing like with like is critical in developing an evidence-based approach to care. Adding transparency of information on cost is critical to the development of a value-based approach to care.

References:

• Bombardier C. Spine Fusion Issues Introduction: Outcomes assessment in the evaluation of treatment of spinal disorders. Spine. 2000;25(24):3097-3099.
• Hansson T, Hansson E, Malchau H. Utility of spine surgery: a comparison of common elective orthopaedic surgical procedures. Spine 2008;33:2819-2830.
• Klazen CA, Lohle PN, DeVries J, et al. Vertebroplasty versus conservative treatment in vertebral compression fractures (Vertos II):An open randomized trial. Lancet. 2010:25(376):1085-1092.
• Porter ME, Teisberg EO: Redefining Health Care. Creating Value-based competition on results. Harvard Business School Press, 2006.
• Tosteson AN, et al. Spine. 2008;33(19):2108-2115.
• Tosteson AN, Lurie JD, Tosteto JD et al. Surgical Treatment of spinal stenosis with and without degenerative spondylolisthesis: Cost-effectiveness after 2 years. Ann Int Med 2008;149(12):845-853.

• Todd J. Albert, MD, can be reached at The Rothman Institute, 925 Chestnut St., Philadelphia, PA 19107; 267-339-3500; tjsurg@aol.com.
• Gunnar Andersson, MD, PhD, can be reached at Rush Presbyterain St. Lukes Medical Center, 1653 W. Congress Parkway Chicago, IL 60612; 312-942-4867; e-mail: Gunnar_Andersson@ rsh.net.
• Sigurd Berven, MD, can be reached at 500 Parnassus Ave., MU320W, San Francisco, CA 94143-0728; 415-514-2064; e-mail: Bervens@orthosurg.ucsf.edu.
• Steven D. Glassman, MD, can be reached at Department of Orthopaedic Surgery, University of Louisville School of Medicine, Louisville, KY 40202; e-mail: steven.glassman@nortonhealthcare.org.
• Frank M. Phillips, MD, can be reached at Rush University Medical Center, Midwest Orthopaedics, 1725 W. Harrison St., Suite 1063, Chicago IL 60612; 312-432-2339; e-mail: frank.phillips@rushortho.com.
• James N. Weinstein, DO, MS, can be reached at One Medical Center Drive, Lebanon, NH 03756; 603-650-5000; e-mail: james.n.weinstein@dartmouth.edu.

Follow OrthoSuperSite.com on Twitter