Trends and Variations in the Use of Spine Surgery : Clinical Orthopaedics and Related Research®

Secondary Logo

Journal Logo

SECTION I: SYMPOSIUM II: Surgical versus Nonsurgical Management of Spinal Disorders

Trends and Variations in the Use of Spine Surgery

Deyo, Richard A MD, MPH*†‡§; Mirza, Sohail K MD, MPH‡§

Editor(s): Moskovich, Ronald MD, Guest Editor; Nordin, Margareta DrSci, Guest Editor

Author Information
Clinical Orthopaedics and Related Research: February 2006 - Volume 443 - Issue - p 139-146
doi: 10.1097/01.blo.0000198726.62514.75
  • Free

Abstract

Examining patterns of spine surgery with time and across geographic regions can lend insight into variations and changes in clinical decision making. These patterns and trends may also identify responses to the introduction of new surgical technologies. Such studies may have important implications for the quality and cost of care because they may suggest overuse or underuse of certain treatments, areas of limited professional consensus, and opportunities for creating more consistent approaches to clinical care. Population-based data can help to identify and quantify trends in the use of spine surgery, examine a limited number of outcomes, and identify important questions for future research.

United States surgery rates have been increasing in recent years.12,26 Epidemiologic data do not suggest an increasing rate of spinal disorders, higher rates of spinal disease in the U.S. than elsewhere, or dramatic differences in prevalence among regions within the U.S.1,15,24,29 In Britain, substantial variations in physician visit rates for back pain have been documented among regions with similar prevalence of back pain.31 Therefore, critical observers may reasonably ask if rising rates and wide geographic variations suggest a poor professional consensus on treatment approaches and whether technologic innovations and marketing affect these decisions. Those who pay for care-in government, business, and the insurance industry-may reasonably ask whether more expensive styles of practice deliver better results or represent cost without benefit.

Wide geographic variations in the use of medical services have been recognized for many years. Although with most medical services there is substantial variability from one location to the next, back surgery was identified early on as one of the most highly variable types of surgical intervention.33 This seems to be true for small (cities) and large (countries) geographic regions.5,6,30,32

There is a growing body of literature on rates and geographic variations in the use of spine surgery, but its implications for clinicians and researchers rarely has been summarized. We sought to examine population-based data on (1) geographic variations, to examine the magnitude of regional and international variations in surgical rates; (2) the magnitude of increases in spinal surgery; (3) whether any increases were concentrated among certain procedures or segments of the population; (4) the magnitude of shifts to ambulatory surgery, and (5) complications and reoperation rates, which might be useful for surveillance of surgical innovations. We selected literature based upon our own review and provide interpretations where we deem appropriate.

Geographic Variations in Back Surgery

Rates of spine surgery vary approximately fivefold among industrialized countries. The spine surgery rate in the U.S. is the highest in the world, and is approximately five times greater than the rate in England and Scotland. The U.S. rate is double that of most countries, including Australia, Canada, and Scandinavian counties (Fig 1). International rates are associated with the total numbers of neurologic and orthopaedic surgeons per 1000 population in most countries. These international variations are strikingly stable; data collected in 1980 and in 1990 were highly correlated.6

F1-23
Fig 1:
The ratio of back surgery rates in selected countries to the U.S. rate of back surgery is shown. Adapted with permission from Cherkin DC, Deyo RA, Loeser JD, Bush T, Waddell G. An international comparison of back surgery rates. Spine. 1994;19:1201-1206.

Even within countries, small geographic areas show marked variations. For example, we found in 1989 that rates of spine surgery among Medicare beneficiaries ranged from 30 per 100,000 (in Rhode Island) to 132 per 100,000 (in Utah).7 Data from 20015 show much higher rates overall, but demonstrate similarly striking variations. In some states the overall surgery rates among Medicare enrollees were unusually low and in some states the rates were unusually high (Table 1). These rates ranged from1.8 operations per 1000 Medicare enrollees in Hawaii up to 9.2 operations per 1,000 Medicare enrollees in Wyoming.5

T1-23
TABLE 1:
Geographical Variations in Spine Surgery within the 2001 U.S. Medicare Population

In many cases, the smaller the geographic areas examined, the more striking variations become. Investigators at Dartmouth University have identified hospital referral regions throughout the U.S. that correspond in large measure to U.S. cities (Table 1). Surgery rates particularly were low in many east coast cities such as New York, NY, with a rate of 2.1 surgeries per 1000 Medicare enrollees. The lowest rate was observed in Terre Haute, IN, with 1.6 operations per 1000 enrollees. In contrast, some of the highest rates were observed in the western states. The highest rate identified was in Bend, OR, with 10.2 operations per 1000 enrollees. Sometimes the rates varied widely within a state. For example, the rate was 2.1 operations per 1000 in McAllen, TX compared with 8.3 operations per 1000 in Amarillo, TX. Similarly, the rate was4.4 in Colorado Springs, CO, compared with 8.6 in Greeley, CO.5

In comparison with other orthopaedic procedures, back surgery shows greater geographic variation than surgery rates for other conditions (Fig 2). Rates of hospitalization for hip fracture vary only modestly among geographic areas, whereas spine surgery varies substantially. The rates of spinal fusion surgery are even more highly variable than rates for spine surgery in general. Overall rates of spine surgery varied sixfold among geographic regions whereas fusion rates varied 10-fold.21 Similarly, when we examined counties within Washington State, we found back surgery rates varied approximately twofold among counties whereas rates of spinal fusion surgery varied threefold.28

F2-23
Fig 2:
This graph shows the ratios of rates of treatment for hip fractures, all spinal surgery, and spinal fusions among hospital referral regions in the U.S. A long scale, centered about the national average(1.0), was used for clarity. Reprinted with permission from Lurie JD, Weinstein JN. Shared decision making and the orthopaedic work force. Clin Orthop Relat Res. 2001;385:68-75.

Total joint arthroplasty falls between hip fracture and spine surgery in its variability. For example, in 1996 and 1997, Medicare data showed 2.2-fold differences in hip fracture repair rates among U.S. cities, 4.4-fold differences in joint arthroplasty, and 5.9-fold differences in spine surgery.32 Low variation conditions such as hip fracture generally require nondiscretionary forms of care and the diagnosis is usually unequivocal. In contrast, high variation conditions generally are those in which there is considerable physician discretion, and in which diagnostic criteria may be less standardized.32

The explanation for these geographic variations in surgery rates remains unclear. Surgical rates might be influenced by disease prevalence; patient factors (eg, expectations); physician factors (eg, differences in training or clinical philosophy); and system factors (eg, numbers of surgeons, availability of imaging, financing methods). Authors of epidemiologic studies suggest rates of back pain and disc herniation are similar in most countries of the world and in most geographic areas, varying only by a few percentage points.15,31 Differences in disease prevalence probably account for only a small portion of the variation in rates.

Lurie and Weinstein21 observed areas with very high surgery rates often are immediately adjacent to areas with low rates. This pattern suggested the variability stemmed from physician practice styles rather than characteristics of the population. Some patient and system factors were examined in an analysis of variability in the state of Washington. Surgery rates were higher in counties with a large percent of workers with heavy lifting or transportation occupations, counties with higher levels of poverty, and counties with lower rates of hospital bed occupancy. However, these statistics were strongly associated with each other, and only explained about 12% of the variability in surgery rates.30

Lurie et al20 found rates of spine surgery are correlated with rates of magnetic resonance imaging (MRI) and computed tomography (CT) in different geographic areas. Rates of advanced spinal imaging accounted for 22% of the variability in overall spine surgery rates among hospital referral regions of the U.S. A simulation model showed the imaging studies obtained in the specific patients who had surgery accounted for only a small part of the correlation between imaging rates and surgery rates, suggesting the broader pattern of imaging practice was critical.19 These findings raised questions about the appropriate indications for spinal imaging and concerns that imaging studies may be overused in some areas. Lurie et al20 suggested an improved consensus on the use and interpretation of advanced spinal imaging studies could have an important effect on variations in spine surgery rates.

Although some of the variability in surgery rates may be related to patient behavior and care-seeking patterns, the majority remains unexplained. Many observers believe a majority of the variability reflects professional uncertainty about indications for imaging and various types of surgery. This uncertainty may in turn stem from variation in training practices and local habits, and a paucity of data on the outcomes of different approaches to care.

We cannot easily determine which of these practice patterns may be “correct.” Are surgery rates excessive in some areas, and do they reflect underuse in others? Which rate is right? Some outcome data from the state of Maine (summarized below) suggest surgical outcomes are not better in high rate areas, and may even be worse.18 This lends credence to the idea that in some areas, surgery may be overused.

Trends in Spine Surgery Rates in the United States

Authors of studies conducted a decade ago indicated rates of lumbar and cervical spine surgery rose steadily during the 1980s.9,26 For degenerative conditions of the lumbar spine specifically, the rate of spine surgery increased by 55% from 1979 to 1990.4 When subdivided according to the type of surgery, it seemed laminectomy or discectomy without fusion increased by about 47% during this time whereas surgery involving a lumbar fusion increased 100% during this decade.26

More recent estimates of surgical rates in the U.S. are complicated by the fact that a substantial proportion of discectomies are now done on an ambulatory basis and no longer are reflected in hospital discharge databases. A trend toward the use of ambulatory surgery began around 1994,4,19 and national data on ambulatory surgery rates are unavailable. However, a handful of states collect data on ambulatory surgery, and we have estimated the growth of ambulatory lumbar spine surgery from four states representing different regions of the U.S. (New York, Florida, Maryland, and Colorado). These four states had a steady increase in the proportion of all discectomies done on an ambulatory basis from 1994 through 2001. By 2001, approximately 17% of all lumbar spine surgeries for degenerative conditions were being done on an ambulatory basis and approximately 21% of nonfusion surgeries were being done on an ambulatory basis.17 It may be hazardous to extrapolate from these four states to national rates. However, if the proportion of surgery being done on an ambulatory basis in these states is representative, it seems overall surgery rates in the U.S. have continued to rise gradually since 1994.

The most rapid increase in surgery rates has been observed among the elderly. For example, between 1979 and 1992, there was a 40% increase in age-adjusted and sex-adjusted back surgery rates among adults younger than 65 years. However, for those older than 65 years, there was almost a fourfold increase in surgery rates, mostly attributable to an eightfold increase for spinal stenosis.26 The trend for the largest increases in surgical rates to occur among the elderly continued during the 1990s.12

Trends in Spinal Fusion

Because spinal fusion still is done almost uniformly on an inpatient basis, the inpatient national databases still provide an adequate picture of changes in spinal fusion surgery. Data from the National Hospital Discharge Survey show spine fusion surgery rates continued to rise in the 1990s even more rapidly than in the 1980s (Fig 3).12 Overall, the rate of lumbar spinal fusion surgeries for degenerative conditions tripled during the 1990s, after already doubling during the 1980s. This made spinal fusion surgery one of the most rapidly increasing forms of inpatient surgery in the U.S. Fusion rates rose especially quickly after 1996, when intervertebral fusion cages were approved by the Food and Drug Administration (FDA) (Fig 3). All types of spinal fusion (cervical and lumbar, including all diagnoses) increased 77% between 1996 and 2001 (Fig 4). In contrast, other major orthopaedic procedures, such as hip replacement and knee arthroplasty increased by only 13% to 15% during this time.13

F3-23
Fig 3:
The trend in the number of lumbar spinal fusions for degenerative conditions done per 100,000 adults in the U.S. is shown. Data are from the National Hospital Discharge Survey.
F4-23
Fig 4:
The annual numbers of knee arthroplasties, hip arthroplasties, and spinal fusion operations (all levels, all diagnoses) done in the U.S. are shown. Data are from the National Inpatient Sample from the Agency for Healthcare Research and Quality. Reprinted with permission from Deyo RA, Nachemson A, Mirza SK. Spinal fusion surgery: The case for restraint. N Engl J Med. 2004;350:722-726.

In addition to a rising rate of lumbar fusion surgery, it seems an increasing proportion of all spine operations includes a fusion procedure. During the 1980s, the proportion of patients having spine fusion for a primary diagnosis of herniated disc changed little. However, the proportion of patients with a primary diagnosis of degenerative changes who received a fusion doubled during decade; for the diagnosis of spinal stenosis, the proportion of patients who had a spine fusion quadrupled.26

Because of the shift to ambulatory surgery during the mid 1990s, it is difficult to estimate the proportion of all operations for a herniated disc that now incorporate a spinal fusion procedure. However, for other diagnoses for which surgery still is done almost exclusively in the hospital, the proportion of cases involving spine fusion has increased steadily. Among patients with degenerative changes, 70% had a fusion procedure in 2001. For spinal stenosis, the proportion was 26%. For patients with possible instability, 93% of operations included a fusion procedure in 2001.12 In the cervical spine, the rate of discectomy leveled off in the 1990s, but the proportion of all operations involving a fusion rose substantially, as in the lumbar spine.2

In addition to an increasing proportion of operations that involve fusion, an increasing proportion of all fusions involve instrumentation. In a review of published clinical studies of lumbar fusion, reported rates of internal fixation nearly doubled from the 1980s to the 1990s. Despite this change, reported outcomes (solid fusion rates, “excellent” or “good” clinical results) showed no improvement.3 These literature-based estimates are supported by population-based rates in patients in the Medicare-eligible population. During the last 5 years of the 1990s, Lurie and Weinstein21 noted a 40% increase in spine surgery rates, a 70% increase in spine fusion rates, and a more than doubling of the rates of instrumented fusions.

Although time trends in the use of surgical procedures cannot be used to establish causal associations, they sometimes generate reasonable hypotheses. For example, we noted previously the rapid increase in spinal fusion rates after the 1996 introduction of intervertebral fusion cages. Similarly, in the state of Washington, we were able to document the apparent impact of a guideline for lumbar fusion surgery and reimbursement standards that were implemented by the statewide Workers' Compensation program in the late 1980s. After implementation of that guideline, the rate of lumbar spine fusion in Washington declined 33%, whereas rates for nonfusion operations essentially were unchanged. The sharpest decline in rates corresponded in time to the implementation of the guidelines.16

Reoperations and Other Surgical Outcomes

Opportunities for studying surgical outcomes using population-based databases are limited. However, the occurrence of a second spine operation can be identified reliably, as can mortality within a short time of surgery. Complications are often included in hospital discharge databases, but generally are underestimated and under recorded. We do not advocate using administrative databases for comparing individual hospital or surgeon complications rates. However, it may be valid to compare relative complication rates between surgical procedures when the different operations are compared during the same time interval and from the same hospitals.

In one study of patients on Medicare benefits in 1985, just over 10% of patients had a second back operation after 4 years of followup.8 The likelihood of a reoperation decreased as patients increased in age. In contrast, reoperations were more common among patients with previous back surgery than among those having a first operation. In the state of Washington, considering all adult age groups (not just adults eligible for Medicare), the 5-year reoperation rate was 15%.23 This is substantially greater than the rates of revision surgery for knee or hip replacement.25

In some cases, surgeons choose to do fusions at the time of discectomy or decompressive laminectomy with the rationale that it will reduce the likelihood of repeat surgery. However, patients who have spinal fusion surgery are not less likely to have subsequent operations than those who have a laminectomy or discectomy alone. In fact, for all diagnoses, patients who have a spinal fusion seem slightly more likely to have a second operation than patients who have a laminectomy or discectomy only.23

In comparing patients covered by Workers' Compensation insurance with patients who had other types of insurance, we found those covered by Workers' Compensation had substantially greater rates of reoperation than patients with other insurance. After only 3 years of followup, 18% of patients on Workers' Compensation benefits had had a second operation in contrast to 10% of patients with other sources of insurance.27 The percentages were the same for the patients who had fusion surgery compared with those having nonfusion operations.

Although complication rates probably are underestimated in administrative databases, they may provide meaningful contrasts in relative complication rates among different demographic groups, diagnoses, or procedures. For example, we found that among patients on Medicare benefits, surgical mortality during hospitalization or within 6 weeks of discharge rose from about 0.5% among those aged 65 to 69 years, to more than 2% among those patients 80 years or older. Similarly, complication rates increased steadily with an increasing burden of comorbid medical illnesses.7 As expected, complication rates were higher for patients having reoperations than for those having first-time surgeries.11 Even after adjusting for age, sex, comorbidity, race, previous back surgery, diagnoses, and number of hospitalizations in the previous year, we found that patients having spine fusion had approximately twice the risk of complications as those having laminectomy or discectomy without fusion.10,11 For older patients having surgery, mortality rates approximately doubled when a fusion was done, as did the probability of being discharged to a nursing home.11 The higher rate of complications among patients having fusion also was apparent even in younger patients when we examined all age ranges of patients within the state of Washington.10

Data on pain relief and functional improvement generally are unavailable in large administrative databases. However, in a large prospective cohort study involving 655 patients in the state of Maine, we were able to compare symptoms and functional improvement among patients who had surgery in regions of the state with low surgical rates, high surgical rates, and intermediate rates. At 2 to 4 years after surgery, patients had the best outcomes in the region of the state with the lowest surgical rate and the worst outcomes in the region of the state with the highest surgery rate. Outcomes were average among patients in the region of the state with intermediate surgical rates.18 These results suggested higher rates of surgery are not necessarily better and higher surgical rates sometimes may be associated with worse average outcomes.

DISCUSSION

There are several major findings from this review of population-based data. Spine surgery rates are highly variable among geographic regions, and fusion rates are more variable than overall rates of spine surgery. Overall surgery rates are higher in the U.S. than any other country, and are still rising. Spine fusion and instrumentation rates are rising especially quickly. Complication rates are higher for fusions than for decompression surgery alone, and there is no evidence fusion is associated with lower rates of reoperation than decompression alone.

The analysis of large survey or insurance claims databases has strengths and limitations.14 The major strength is that the data are highly representative of practice patterns in an entire geographic region. The findings are not limited to a single surgeon's practice, to academic medical centers, or to those who choose to study and publish their results. On the other hand, these databases are limited in the clinical detail they provide. For example, they include a physician's diagnosis, but no clinical examination findings or imaging data to verify the diagnosis. The severity of disease is not indicated. The databases identify surgical procedures, but do not indicate specific spinal level(s). Databases that depend entirely on International Classification of Diseases (ICD)-9 codes do not indicate whether instrumentation is used. Errors in coding diagnoses and procedures occur, though most of these databases are closely related to billing data, which have improved in recent years and are subject to audit.

An important limitation when comparing reoperation rates or complication rates for particular procedures is that the patient populations selected for different procedures (eg, decompression versus fusion) differ. However, such comparisons are generally made for patients with a particular diagnosis, and with adjustment for potential confounders such as age and comorbid conditions. Furthermore, the wide geographic variations in care suggest patients with similar characteristics may receive different procedures, depending on where they are and who they see.

Factors such as the shift to ambulatory surgery now complicate the use of hospital databases for studying spine surgery rates. As more minimally invasive techniques are introduced, we may anticipate even a larger fraction of spine surgery may be done on an ambulatory basis. This means authors of future studies will need to make use of combined hospital and ambulatory data more often and more effort should be given to creating combined databases.

Our finding of acceleration in the rise of spinal fusion rates coincident with the introduction of interbody fusion cages raises the possibility that the introduction of new technology may sometimes influence procedure rates. We are unaware of other major advances or new data on efficacy or surgical indications that would explain the jump we observed in 1996.12 New spinal implants often are marketed aggressively, and we speculate the impact of marketing efforts may be measurable with data such as these. Such findings emphasize the need for more and better data on efficacy as new spine surgical technologies are introduced.

The study of reoperation rates may offer a useful strategy for post-market surveillance of new surgical technology. Recent events have highlighted the need for more rigorous, prospective surveillance of new pharmaceuticals, and the same concern applies to new surgical devices and materials. With improvements in coding for new devices, faster release of claims and survey data, and better access to medical records for validation and detail, tracking re-operation and complication rates may become a useful component of such surveillance.22

Because new technology for spinal surgery is introduced at an ever-increasing pace, we may anticipate substantial changes in the surgical patterns noted here. The introduction of spinal fusion cages may serve as an example of how new technology relates to changing patterns of surgery. The introduction of bone morphogenetic proteins, new implants, bone graft products, and now artificial discs likely will have major impacts on surgical rates and patterns in the future. The analysis of population-based data, such as the data presented here, will help in assessing the long-term impact of these new technologies.

Acknowledgments

The authors thank Katherine Henne for her valuable assistance in preparation of the manuscript, Brook Martin, MPH, for administrative assistance for several aspects of the data presented here, and William Kreuter, MPA, for expert programming and database management for some of the analyses. The authors also thank Darryl Gray, MD, MPH, for conducting the analysis and presenting much of the data on ambulatory surgery rates; the data provided here are from his presentation at an international meeting.

References

1. Andersson GBJ. The epidemiology of spinal disorders. In: Fry-moyer JW, ed. The Adult Spine: Principles and Practice. 2nd Ed. Philadelphia, PA: Lippincott-Raven; 1997:93-141
2. Angevine TD, Arons RR, McCormick PC. National and regional rates and variation of cervical discectomy with and without interior fusion, 1990-1999. Spine. 2003;28:931-940.
3. Bono C, Lee C. Critical analysis of trends in fusion for degenerative disc disease over the past twenty years: Influence of technique on fusion rate and clinical outcome. Spine. 2004;29:455-463.
4. Bookwalter JW, Busch MD, Nicely D. Ambulatory surgery is safe and effective in radicular disc disease. Spine. 1994;19:526-530.
5. Center for the Evaluative Clinical Sciences at Dartmouth. The Dartmouth Atlas of Health Care. Available at: www.dartmouthatlas.org/download.php. Accessed August, 2004.
6. Cherkin DC, Deyo RA, Loeser JD, Bush T, Waddell G. An international comparison of back surgery rates. Spine. 1994;19:1201-1206.
7. Ciol MA, Deyo RA, Howell E, Kreif S. An assessment of surgery for spinal stenosis: time trends, geographic variations, complications, and reoperations. J Am Geriatr Soc. 1996;44:285-290.
8. Ciol MA, Deyo RA, Kreuter W, Bigos SJ. Characteristics in Medicare beneficiaries associated with reoperation after lumbar spine surgery. Spine. 1994;19:1329-1334.
9. Davis H. Increase in rates of cervical and lumbar spine surgery in the United States, 1979-1990. Spine. 1994;19:1117-1123.
10. Deyo RA, Cherkin DC, Loeser JD, Bigos SJ, Ciol MA. Morbidity and mortality in association with operations of the lumbar spine. J Bone Joint Surg. 1992;74A:536-543.
11. Deyo RA, Ciol MA, Cherkin DC, Loeser JD, Bigos SJ. Lumbar spine fusion: A cohort study of complications, reoperations, and resource use of the Medicare population. Spine. 1993;18:1463-1470.
12. Deyo RA, Gray DT, Kreuter W, Mirza S, Martin BI. United States trends in lumbar fusion surgery for degenerative conditions. Spine. 2005;30:1441-1445.
13. Deyo RA, Nachemson A, Mirza SK. Spinal fusion surgery: The case for restraint. N Engl J Med. 2004;350:722-726.
14. Deyo RA, Taylor VM, Diehr P, Conrad D, Cherkin DC, Ciol M, Kreuter W. Analysis of automated administrative and survey databases to study patterns and outcomes of care. Spine. 1994;19 (18 suppl): 2083S-2091S.
15. Deyo RA, Tsui-Wu J. Descriptive epidemiology of low-back pain and its related medical care in the United States. Spine. 1987;12: 264-268.
16. Elam K, Taylor V, Ciol MA, Franklin GM, Deyo RA. Impact of a Workers Compensation practice guideline of lumbar spine fusion in Washington State. Med Care. 1997;35:417-424.
17. Grady DT, Deyo RA, Kreuter W, Mirza SK, Heagerty PA, Com-stock BA, Chan L. Population-based trends in volumes and rates of ambulatory lumbar spine surgery. Spine. (In press).
18. Keller RB, Atlas SJ, Soule DN, Singer DE, Deyo RA. Relationship between rates and outcomes of operative treatment for lumbar disc herniation and spinal stenosis. J Bone Joint Surg Am. 1999;81:750-762.
19. Kelly A, Griffith H, Jamjoom A. Results of day-case surgery for lumbar disc prolapse. Br J Neurosurg. 1994;8:47-49.
20. Lurie JD, Birkmeyer NJ, Weinstein JN. Rate of advanced spinal imaging and spine surgery. Spine. 2003;28:616-620.
21. Lurie JD, Weinstein JN. Shared decision making and the orthopaedic work force. Clin Orthop Relat Res. 2001;385:68-75.
22. Malenka DJ, Kaplan AV, Sharp SM, Wennberg JE. Postmarketing surveillance of medical devices using Medicare claims. Health Aff. 2005;24:928-937.
23. Malter AD, McNeney B, Loeser JD, Deyo RA. Five-year reoperation rates after different types of lumbar spine surgery. Spine. 1998;23:814-820.
24. Nachemson A. Epidemiology and the economics of low back pain. In: Herkowitz HN, HN, Dvorak J, Bell G, Nordin M, Grob D, eds. The Lumbar Spine. 3rd Ed. Philadelphia, PA: Lippincott Williams and Wilkins; 2004:3-10.
25. NIH Consensus Panel. NIH Consensus Statement on total knee replacement December 8-10, 2003. J Bone Joint Surg. Am. 2004;86-A:1328-1335.
26. Taylor VM, Deyo RA, Cherkin DC, Kreuter W. Low back pain hospitalization: Recent United States trends and regional variations. Spine. 1994;19:1207-1213.
27. Taylor VM, Deyo RA, Ciol M, Kreuter W. Surgical treatment of patients with back problems covered by Workers Compensation versus those with other sources of payment. Spine. 1996;21:2255-2259.
28. Taylor VM, Dey RA, Goldberg H, Ciol M, Kreuter W, Spunt B. Low back pain hospitalization in Washington state: Recent trends and geographic variations. J Spinal Disord. 1995;8:1-7.
29. U.S. Department of Health and Human Services. Summary Health Statistics for U.S. Adults: National Health Interview Survey, 2002. Vital and Health Statistics Series 10, #222. DHHS Publication # (PHS) 2004-1550, Hyattsville, Maryland.
30. Volinn E, Mayer J, Diehr P, Van Koevering D, Connell FA, Loeser JD. Small area analysis of surgery for low back pain. Spine. 1992;17:575-581.
31. Walsh K, Cruddas M, Coggon LD. Low back pain in eight areas of Britain. J Epidemiol Commun Health. 1992;46:227-230.
32. Weinstein JN, Birkmeyer JD, Abdu WA, Birkmeyer NO, Bronner KK, Cooper MM, Lurie JD, Sharp SM, Shawver TA, Siewers AE. The Dartmouth Atlas of Musculoskeletal Health Care. Chicago, IL: AHA Press; 2000:140-142.
33. Wennberg JE, McPherson K, Caper P. Will payment based on diagnosis related groups control hospital costs? N Engl J Med. 1984;311:295-300.
© 2006 Lippincott Williams & Wilkins, Inc.