In the United States, procedures for the treatment of pediatric spinal deformity are currently performed by surgeons with varying residency and fellowship training experience. Many spinal deformity procedures are performed by pediatric orthopaedic surgeons, who receive orthopaedic residency training followed by pediatric orthopaedic fellowship training. Some spine surgeons, who receive training in spine fellowship programs, also perform pediatric scoliosis surgery. These surgeons usually complete either orthopaedic or neurosurgery residency programs prior to fellowship. Because of these variations in training, controversy exists with regard to the appropriate type of fellowship training for optimal experience with the operative correction of scoliosis as well as with regard to which types of subspecialty surgeons should be performing these operations.
As a result, some surgeons and administrators have proposed that a subspecialty certificate—previously known as a certificate of special qualifications (CSQ) or certificate of added qualifications (CAQ)—should be required for doctors who perform spinal deformity procedures. For example, the Scoliosis Research Society has drafted an application for subspecialty certification in the field of spinal deformity surgery, and members of the newly formed American Board of Spine Surgery (ABSS) have taken the position that “founding members of the ABSS believe that there is general agreement that spine surgery is, and should be, a separate and independent surgical specialty”1. These issues are being discussed and debated by members of many of the world's prominent leadership bodies in orthopaedic and neurological surgery, including the American Academy of Orthopaedic Surgeons1, the North American Spine Society2, the American Board of Spine Surgery1, the Scoliosis Research Society3, the American Academy of Neurologic Surgeons1, and the Pediatric Orthopaedic Society of North America4.
Surgical volume is considered to be an important measure of the quality of surgical training by Residency Review Committees and has been used as a potential predictor of surgical outcomes, both in orthopaedic surgery and in other specialty fields5-13. For example, Birkmeyer and colleagues found that both hospital and surgeon volume were inversely related to the operative mortality associated with various cardiovascular and cancer-resection procedures12,13. However, there is a paucity of documentation regarding the relative exposure to pediatric spinal deformity surgery that fellows in pediatric orthopaedic and spine surgery fellowships receive in each type of program. Moreover, the volume of scoliosis surgery actually performed by surgeons with either type of fellowship training remains unreported.
The purpose of the present study was to investigate the association between fellowship training and the volume of spinal deformity procedures on two levels: the surgeon level and the training-institution level. First, an analysis of all spinal fusion procedures that had been performed for the treatment of scoliosis in patients with an age of eighteen years or less in New York State from 1992 to 2001 was conducted to identify the fellowship training experiences of the surgeons who collectively performed the majority of such operations in the state and to compare the surgical volumes of pediatric orthopaedic fellowship-trained surgeons with that of spine surgery fellowship-trained surgeons. Second, an analysis of all spinal fusion procedures that had been performed for the treatment of scoliosis in patients with an age of eighteen years or less in California from 1995 to 1999 was conducted to identify the fellowship programs of hospitals at which such operations in that state were performed and to compare the surgical volume of hospitals with pediatric orthopaedic fellowship programs with that of hospitals with spine surgery fellowship programs.
Materials and Methods
The Statewide Planning and Research Cooperative System (SPARCS) is a publicly available comprehensive data system established in 1979 by the New York State Department of Health in cooperation with the healthcare industry. The SPARCS inpatient database contains automated discharge records for each discharge from nonfederal, state-licensed hospitals in New York, submitted according to Universal Data Set specifications by trained medical records personnel in each hospital and verified for accuracy by the Department of Health. The records include information relating to the patient's disposition, age, gender, race, admission status, physician identifiers, principal diagnosis, secondary diagnoses, principal procedure, and secondary procedures. SPARCS data have been used effectively for a large number of studies investigating treatment outcomes, provider volume and mortality relationships, and epidemiological trends in both medical and surgical subspecialties and in both pediatric and adult patient populations11,14-18.
In the New York component of the current study, the SPARCS database was searched for all inpatient records from the years 1992 to 2001 in which the operative procedure of spinal arthrodesis (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] codes 81.00 through 81.09) was performed for patients with an age of eighteen years or less who had a primary diagnosis of any type of scoliosis (ICD-9-CM codes 737.30 through 737.39, 737.43, or 754.2). Appropriate procedure and diagnosis codes were identified according to descriptions found in the ICD-9-CM, Sixth Edition, and the ICD-9 Procedure Code Directory19. Physician identifiers for the “Operating Physician” record contained in each spinal fusion case within the SPARCS database were correlated to physician license numbers and physician names with use of the New York State Education Department physician license number Online Verification Searches web site20. The volume of spinal fusion procedures performed for the treatment of scoliosis in patients with an age of eighteen years or less was thereby established for each surgeon for the time-period of 1992 to 2001. Surgeons were further grouped according to categories of “more than fifty” spinal fusion procedures in the ten-year study period or “fifty or fewer” spinal fusion procedures in this ten-year period in order to restrict the analysis to surgeons who performed these procedures with relative frequency. The fellowship training of all surgeons in the “more than fifty” category was identified with use of American Board of Medical Examiners physician directories, online curricula vitae, and telephone calls to surgeons. This volume of more than fifty spinal fusion procedures was examined over a ten-year period, which translates to an annual volume of more than five spinal fusion procedures per year. Fellowship categories included “pediatric orthopaedics” fellowships, “spine surgery” fellowships, or “neither.” Surgeons who were reported to have completed “scoliosis” fellowships in the American Medical Association directories were included in the “spine surgery” group because these fellowships had since become accredited or nonaccredited spine fellowship programs, according to the National Association of Spine Surgeons 2003 online fellowship directory. Physicians in the “neither” category were determined to have completed no fellowship, a general orthopaedic fellowship (in which training occurred under the guidance of a mentor who did not have a particular orthopaedic subspecialty, according to personal telephone conversations), or had a fellowship training experience that was not applicable to either of the other two categories.
Similar to the SPARCS database, the Office of Statewide Health Planning and Development (OSHPD) inpatient hospital database is a publicly available healthcare data system that includes hospital and patient-identified discharge data from all state-licensed hospitals in the state of California. Beginning in 1980, the state of California mandated the semiannual collection of various data elements from all patient discharges from all state-licensed hospitals through the California Health Facilities Commission; today, the OSHPD database has been expanded to include eighteen data elements and is edited for errors and consistency among data elements within each discharge data record. The OSHPD previously validated the accuracy and quality of the data in a reabstracting analysis21, and over the past several years this database has formed the basis of numerous large-scale investigations into the outcomes of medical, surgical, and obstetric procedures in the state of California22-24.
In the California component of the current study, five consecutive annual OSHPD inpatient discharge databases, from 1995 to 1999, were merged into one. From this merged database, spinal fusion procedures in patients with scoliosis were selected in a manner identical to that performed for New York procedures contained in the SPARCS database.
Active fellowship programs in California and their associated hospitals were identified from the North American Spine Society and Pediatric Orthopaedic Society of North America web sites and through telephone calls to program directors. These hospitals were identified in the OSHPD database through the use of a hospital facility code called a record linkage number. Hospitals at which fellows actively rotated were categorized as either “pediatric orthopaedic surgery,” or “spine surgery” sites. Fellowship programs were excluded if there were no fellows who trained at the program during the period of the study (two programs) or if no spinal fusions had been performed for the treatment of scoliosis in children eighteen years of age or less during that period (one program). Furthermore, some hospitals had both spine and pediatric orthopaedic fellowship programs; in order to analyze such institutions, hospitals at which at least 90% of procedures for the treatment of scoliosis were performed by surgeons in one type of fellowship were classified according to that type of fellowship, at the exclusion of the other (two hospitals). No hospital had both spine and pediatric orthopaedic fellowship programs in which surgeons in one of the fellowship programs performed <90% of the procedures for the treatment of scoliosis. This yielded a total of three pediatric fellowships and nine spine surgery fellowships. The mean annual number of spinal fusions performed for the treatment of scoliosis over the five-year span was identified for each center. Fellowship directors were contacted to determine the number of fellows in each program during the five-year period. During years in which programs had fellows, the mean number of spinal fusion procedures per fellow was calculated for all fellowships; scoliosis discharges during years in which no fellows were present at the institution were excluded.
The SPARCS database was utilized because it provides information regarding surgeon volume (but not hospital volume), whereas the OSHPD database was utilized because it provides information regarding hospital volume (but not surgeon volume). Both databases were thought to provide complementary yet unique information regarding surgical volume, as both surgeon volume and hospital volume have been shown to predict the outcomes of surgical procedures12,13. All data were analyzed with use of the Statistical Package for Social Sciences (version 11.0.1; SPSS, Chicago, Illinois). For the New York analysis, independent-samples t tests were conducted to compare mean annual surgical volumes between surgeons training at spinal surgery and pediatric orthopaedic fellowship programs. For the California analysis, independent-samples t tests were used to compare mean annual surgical volumes between hospitals with spinal surgery and pediatric orthopaedic fellowship programs and to compare mean annual surgical volumes per fellow between hospitals with spinal surgery and pediatric orthopaedic fellowship programs. Time-trend analyses were conducted with use of chi-square analysis to compare the change in surgical volume over time among pediatric orthopaedic and spine surgery programs.
From 1992 through 2001 in New York State, 6910 spinal fusions were performed for the treatment of scoliosis in patients of all ages, as identified in the SPARCS database. Of these, 5136 procedures (74%) were performed for patients eighteen years or less. Among the 228 surgeons who performed one or more spinal fusion procedures in this age-group, only thirty (13%) had performed more than five spinal fusion procedures per year (more than fifty spinal fusion procedures in this ten-year span). However, these thirty surgeons accounted for 75% (3858) of all 5136 procedures in this age-group. The thirteen surgeons who had completed a pediatric orthopaedic fellowship performed 49% (1882) of the 3858 procedures, whereas the twelve surgeons who had completed a spine surgery fellowship performed 33% (1256) of these 3858 procedures (Fig. 1). The pediatric orthopaedic fellowship-trained surgeons had performed a mean of 14.5 procedures per surgeon per year, whereas the spine surgery fellowship-trained surgeons had performed a mean of 10.5 procedures per surgeon per year; this difference was not significant (p = 0.30) (Fig. 2). The five surgeons who had not completed either type of fellowship performed 720 (19%) of the 3858 procedures, with a mean of 14.4 procedures per surgeon per year.
When this mean volume was compared with that of the pediatric orthopaedic fellowship-trained surgeons (p = 0.79) and spine surgery fellowship-trained surgeons (p = 0.49), no significant differences were found.
From 1995 through 1999 in California, 4085 spinal procedures were performed for the treatment of scoliosis in patients of all ages, as identified in the OSHPD database. Of these, 3357 procedures (82%) were performed for patients eighteen years of age or less. Three active pediatric orthopaedic fellowship programs and nine active spine surgery fellowship programs performing scoliosis surgery during this time-period were identified (Table I). From 1995 to 1999, 1590 spinal fusion procedures were performed for the treatment of scoliosis at hospitals affiliated with either type of fellowship. Of these, 56% (885) were performed at hospitals with a pediatric fellowship and 44% (705) were performed at hospitals with a spine fellowship. The mean annual volume (and standard deviation) of spinal fusion procedures performed for the treatment of scoliosis was 59.0 ± 46.6 procedures per year for the three pediatric fellowship programs and 15.7 ± 29.7 procedures per year for the nine spine surgery fellowship programs (p = 0.232).
The mean number of fellows at the three pediatric orthopaedic fellowship programs was 1.9 ± 1.0 per year, and the mean number of fellows at the nine spine surgery fellowship programs was 1.2 ± 0.5 per year (p = 0.093). An analysis of the number of spinal fusion procedures per fellow revealed an annual volume of 31.6 ± 43.6 procedures per fellow per year for the three pediatric orthopaedic fellowship programs and of 12.7 ± 30.6 procedures per fellow per year for the nine spine surgery fellowship programs (p = 0.292) (Fig. 3). Time-trend analyses revealed that, over the study period, there was a significant increase in the mean number of spinal fusion procedures per year both at hospitals with spine programs (14.8 per year in 1995 compared with 16.8 per year in 1999; Χ2 = 20.62, p < 0.001) and at hospitals with pediatric programs (48.7 per year in 1995 compared with 70.7 per year in 1999; Χ2 = 15.23, p = 0.018), although the percentage increase was greater for hospitals with pediatric orthopaedic fellowship programs than for hospitals with spine surgery fellowship programs (45.2% compared with 13.5%).
Since the seminal investigation by Luft et al.5 in 1979, it has been well established that high surgical volume is associated with improved outcomes in a variety of orthopaedic and nonorthopaedic procedures6-13. The present study was conducted to investigate the association between fellowship training and the volume of procedures performed for the treatment of spinal deformity by comparing surgical volumes between groups of surgeons who had previously received or were receiving either pediatric orthopaedic fellowship training or spine surgery fellowship training.
To investigate the volume of procedures performed for the treatment of scoliosis by attending surgeons, we used the SPARCS inpatient database to analyze cases in New York State, one of the most populated states in the United States. From 1992 to 2001 in New York, the vast majority of spinal fusion procedures in patients with scoliosis were being performed by a relatively small number of surgeons. These data suggest that there are an overwhelming number of surgeons who were performing five spinal fusion procedures per year or fewer. This finding merits additional study, given the strong evidence of volume-outcome relationships across healthcare and in orthopaedic surgery5-13, and, while to date there is no established minimum annual surgeon or hospital volume of scoliosis cases necessary to improve outcomes and reduce morbidity, recent data have suggested that volume-outcome relationships do exist in the area of scoliosis surgery25. Interestingly, in New York, a large percentage of spinal fusion procedures in patients with scoliosis were found to have been performed by surgeons who had not completed either a pediatric orthopaedic fellowship or an orthopaedic spine fellowship. This result can be explained by the finding that a very small number of high-volume surgeons had pursued orthopaedic training before the time that formal pediatric orthopaedic fellowships or spine fellowships had been established. In the future, the number of such surgeons is likely to diminish as subspecialty training becomes increasingly common. Importantly, among relatively high-volume surgeons, there was no significant difference between pediatric orthopaedic surgeons and spine surgeons with regard to the annual number of spinal fusion procedures performed for the treatment of scoliosis in patients who were eighteen years of age or less.
Among state-licensed hospitals in California with active fellowships in spine surgery or pediatric orthopaedics from 1995 to 1999, the number of procedures performed annually in young adults and children with scoliosis was nearly four times greater at hospitals with pediatric orthopaedic programs as compared with spine surgery fellowship programs, although there was a nonsignificant trend of pediatric orthopaedic surgeons performing approximately 50% more operations per surgeon annually than spine surgeons. Moreover, an analysis of the number of spinal fusion procedures per fellow demonstrated that pediatric orthopaedic fellows were exposed to more than twice the caseload of scoliosis procedures per year. Finally, there was a greater increase in the number of scoliosis procedures in hospitals with pediatric orthopaedics fellowships as opposed to spine surgery fellowships, indicating that the gap is widening. In each of these cases, significance was not achieved because of the high variability of surgical volume between hospitals with one of the two fellowships examined.
The present study had several limitations, including the dual-state analysis and the inability to classify all surgeons in the SPARCS database into one of the two major fellowship categories. Additionally, while the accuracy of the SPARCS and OSHPD hospital discharge databases have been previously examined and validated18,21, it remains possible that there were small inaccuracies in the coding of procedures or that a small number of procedures were not entered into the databases at all, which could have resulted in a relative underestimation of procedures that were actually performed. Furthermore, while the current analyses reveal differences in the quantity of procedures performed by surgeons with different fellowship training and at hospitals associated with different fellowship training programs, one substantial limitation is that it lacks further comparisons of the quality of the curricula and the education provided by such programs. However, because such an analysis is not readily quantifiable and lends itself to highly subjective interpretation, this aspect was not considered to be a goal of the current study. Future studies investigating such aspects of the educational quality of fellowship training programs are warranted. Despite these limitations, the findings of the present study may be of value to orthopaedic residents with an interest in seeking advanced training in the operative treatment of scoliosis as well as to the orthopaedic community as it examines the issue of a subspecialty certification for spinal deformity surgery. ▪
The authors did not receive grants or outside funding in support of their research or preparation of this manuscript. They did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated.
Investigation performed at the Division of Pediatric Orthopaedic Surgery, Department of Orthopaedic Surgery, College of Physicians and Surgeons, Columbia University, New York, NY
1. , Herkowitz HN, Benzel EC. Controversies in spine: Should there be subspecialty certification in spine surgery? Spine. 2002;27: 1478-83.
2. . Challenges of the spine specialists. North American Spine Society Presidential Address. Minneapolis, Minnesota, October 1994. Spine. 1995;20: 1749-52.
3. . Professional competence and subspecialty certification: Scoliosis Research Society: Harrington Lecture, September 23, 1999, San Diego, California. Spine. 2000;25: 2280-4.
4. . Certificate of Special Qualification. Bulletin of the Pediatric Orthopaedic Society of North America. 2003: 1.
5. , Bunker JP, Enthoven AC. Should operations be regionalized? The empirical relation between surgical volume and mortality. N Engl J Med. 1979;301: 1364-9.
6. , Phillips CB, Baron JA, Fossel AH, Mahomed NN, Barrett J, Lingard EA, Harris WH, Poss R, Lew RA, Guadagnoli E, Wright EA, Losina E. Association of hospital and surgeon volume of total hip replacement with functional status and satisfaction three years following surgery. Arthritis Rheum. 2003;48: 560-8.
7. , Deyo RA, Koepsell T, Swiontkowski MF, Kreuter W. Relationship between the volume of total hip replacements performed by providers and the rates of postoperative complications in the state of Washington. J Bone Joint Surg Am. 1997;79: 485-94.
8. , Purves HR, Guller U, Toth AP, Vail TP, Pietrobon R. Provider volume of total knee arthroplasties and patient outcomes in the HCUP-nationwide inpatient sample. J Bone Joint Surg Am. 2003;85: 1775-83.
9. . The association between hospital and surgeon procedure volume and outcomes of total hip replacement in the United States Medicare population: health policy implications. J Bone Joint Surg Am. 2001;83: 1754-5.
10. . Association between hospital and surgeon procedure volume and outcomes of orthopaedic surgery: caveats for interpretation of data and policy applications. J Bone Joint Surg Am. 2002;84: 1715-7.
11. , Popp AJ, Tranmer B, Fuestel P, Waldman J, Shah D. Relationship between provider volume and mortality for carotid endarterectomies in New York state. Stroke. 1998;29: 2292-7.
12. , Stukel TA, Siewers AE, Goodney PP, Wennberg DE, Lucas FL. Surgeon volume and operative mortality in the United States. N Engl J Med. 2003;349: 2117-27.
13. , Siewers AE, Finlayson EV, Stukel TA, Lucas FL, Batista I, Welch HG, Wennberg DE. Hospital volume and surgical mortality in the United States. N Engl J Med. 2002;346: 1128-37.
14. , Fang J, Alderman MH. Differential use of coronary revascularization and hospital mortality following acute myocardial infarction. Arch Intern Med. 2003;163: 461-6.
15. , Lambrinos J. Investigating the impact of age on outcome of mechanical ventilation using a population of 41,848 patients from a statewide database. Chest. 1995;107: 1673-80.
16. , Barlow B, DiScala C, String D, Ray K, Mottley L. Efficacy of pediatric trauma care: results of a population-based study. J Pediatr Surg. 1993;28: 299-305.
17. . Regional variation in child mortality at hospitals lacking a pediatric intensive care unit. Crit Care Med. 2002;30: 94-9.
18. . SPARCS: the New York State health care data system. J Clin Comput. 1980;8: 255-63.
19. International classification of diseases, 9th revision, clinical modification (ICD-9-CM). 6th ed. Washington, DC: US Department of Health and Human Services, Centers for Disease Control and Prevention, Center for Medicare and Medicaid Services; 2004.
20. Office of the Professions, New York State Education Department. Online verification searches. Available at: http://www.op.nysed.gov
21. , Meux EF, Cox G. Report of results from the OSHPD reabstracting project: an evaluation of the reliability of selected patient discharge data, July through December 1990. Patient Discharge Data Section: Office of Statewide Health Planning and Development, State of California; 1993. p 1-43.
22. , Danielsen B, Parker JP, Castles A, Steimle AE. The California report on coronary artery bypass graft surgery 1999 hospital data: Technical report. San Francisco, CA: Pacific Business Group on Health and the California Office of Statewide Health Planning and Development; 2003.
23. , Zach A, Luft HS, Rainwater J, Remy LL, Campa D. The California Hospital Outcomes Project: using administrative data to compare hospital performance. Jt Comm J Qual Improv. 1995;21: 668-82.
24. , Binns GS, Hughes EF, Saunders-Kirkwood K, Zinner MJ, Cates JA. Open cholecystectomy: A contemporary analysis of 42,474 patients. Ann Surg. 1993;218: 129-37.
Copyright 2005 by The Journal of Bone and Joint Surgery, Incorporated
25. , Arons RR, Hyman JE, Skaggs DL, Roye DP, Vitale MG. The contribution of hospital volume, payer status, and other factors on the surgical outcomes of scoliosis patients: a review of 3,606 cases in the State of California. J Pediatr Orthop. 2005;25: 393-9.