J Thorac Cardiovasc Surg 1995;110:570-585
© 1995 Mosby, Inc.
REPORTS OF INTEREST TO THORACIC SURGEONS |
THORACIC SURGERY WORKFORCE REPORTThe fourth report of the ThoracicSurgery Workforce Committee of The American Association for Thoracic Surgery andThe Society of Thoracic Surgeons
Lawrence H. Cohn, MD*,
Richard P. Anderson, MD**,
Floyd D. Loop, MD**,
Richard G. Fosburg, MD***,
Joseph N. Cunningham, MD***,
Hillel Laks, MD**
Received for publication March 10, 1995. Accepted for publication March 15, 1995.
Address for reprints: Lawrence H. Cohn, MD, Department of Surgery,Division of Cardiac Surgery, Harvard Medical School, Brigham and Women'sHospital, Boston, MA 02115.
Abstract
To determine demographics, practicepatterns, and work volume of North American thoracic surgeons, we sent adetailed survey to all members of The American Association for Thoracic Surgeryand The Society of Thoracic Surgeons between January and May 1993 to determinedata for 1992; 3049 of 3487 (87%) thoracic surgeons responded and 2677 (88%)were in active practice. Ninety-seven percent were male and 3% female, with amean age of 52 years. Sixty-five percent considered fee-for-service as theirprimary compensation mode. Only 24% do isolated subspecialty work: 2% pediatriccardiac surgery, 10% general thoracic surgery, and 12% adult cardiac surgery.Seventy-six percent of respondents do both cardiac and thoracic operations.Workload data for adult cardiac, pediatric cardiac, general thoracic, peripheralvascular, and pacemaker operations were requested. Volume data werecross-correlated with age, 10 geographic regions including Canada, type ofpractice, and type of compensation and were cross-checked by hospital dischargedata for 1992. These data were compared with data from similar surveys performedin 1976, 1980, and 1985, under the auspices of the same two societies; theselatter surveys used diplomates of the American Board of Thoracic Surgery astheir database. Workloads have increased over previous surveys. Most surgeons doa wide variety of thoracic operations, and exclusive designations are in theminority. (J THORAC CARDIOVASCSURG 1995;110:570-85)
Surveys of the Thoracic Surgery Workforce and the work they do areimportant (1) to ascertain volume of work in particular areas, especially thosethat overlap other specialties; (2) to document that, in fact, The AmericanAssociation for Thoracic Surgery (AATS) and The Society of Thoracic Surgeons(STS) do represent the group of surgeons performing the vast majority ofthoracic surgical procedures in the United States and Canada; and (3) to helpdetermine if the numbers of thoracic surgeons are adequate for projected demandin the future.
Accordingly, in 1992 the fourth Ad Hoc Committee of the AATS/STS wasformed to survey the Thoracic Surgery Workforce to determine demographics,caseloads by subspecialty area of thoracic surgery, and types of practice andpractice patterns according to age, region of North America, and type ofcompensation and practice.
METHODS
In December 1992 the first mailing of a two-page questionnaire of twelvedetailed questions with a covering letter was sent to the entire membership ofthe AATS and STS. Two subsequent mailings in March and May 1993 were also doneto maximize response. The anonymity of the respondent was preserved by thetwo-envelope system used in previous surveys. We mailed 3487 questionnaires andreceived 3049 responses (87%). Of the 3049 responses, 2677 were from practicingsurgeons (88%). Twelve percent of respondents had retired--a slightly higherpercentage than in the 1985 report ( [approximately] 8% of respondents).
1
The questionnaire included the following: thoracic organizationmembership; age and sex; whether respondent was practicing; principal type ofcompensation and how obtained; primary mode of practice; boardcertification/eligibility; regional area; type and location of practicehospital; residency programs at hospital; number of procedures per year foradult cardiac surgery, pediatric cardiac surgery, general thoracic surgery,peripheral vascular surgery, and pacemaker procedures; and self-designation asto primary type of practice.
All responses were tabulated, collated, and analyzed at the headquartersof the STS by means of a data entry program called Knockout. Once entered, thedata were then read into a tabulating program (A-Cross) for analysis, graphics,and statistics.
RESULTS
Age
The age range of the thoracic surgery respondents ranged from 30 to 70+years, with a mean of 52 and a median of 51.5 years. This continues a trendnoted in previous surveys of increasing mean age: 46 years in 1976, 48 years in1980, and 50 years in 1985. A total of 56 actively practicing thoracic surgeonswere older than 70 years of age, and 68% of respondents older than 60 years werestill practicing. Fig. 1A demonstrates age distributionof respondents and Fig. 1B shows current practice statusby age.
Sex
Of the respondents, 97% were male and 3% were female.
Memberships
Eighty-eight percent of respondents were members of the STS, 28% of theAATS, 20% of the Southern Thoracic Surgical Association, and 9% of The WesternThoracic Surgical Association.
Board eligibility/certification
All respondents were certified or eligible to be certified by theAmerican Board of Thoracic Surgery (ABTS). Seven percent had vascular surgerycertification, 6% had Royal College of Canada certification, and 5% had otherboard certification.
Regions
Table I lists geographic regions analyzed.Fig. 2 shows numbers of respondents per region. The Mid-Atlantic area with 497 and theSoutheast with 395 had the highest number of respondents; the Rocky Mountainregion had the lowest number.
Hospital and residency demographics
Twenty-one percent of respondents practiced in hospitals of fewer than300 beds, and cardiothoracic residencies were found in less than 30% ofrespondents' hospitals. Fig. 3 shows the types ofhospitals in which respondents practice, with private teaching and communityhospitals dominating.
Self-designation
Respondents were asked to list all types of operations performed andtheir self-designation as a practitioner. These data are seen inTable II.Note that 1.9% do only pediatric cardiac surgery, 9.9% do only general thoracic,and 12.4% do only adult cardiac surgery. Thus approximately 75% of thoracicsurgeons do combinations of the five major operative procedures.
Practice profile
demonstrates primary type of practiceand numbers of respondents in each category. The most common practice mode wasthe single-specialty private practice group. This was the mode of practice for1199 or 45% of respondents. The percentage of geographic full-time academicrespondents was 498 or 19%. Solo or individual practice of thoracic surgery wasidentified as the primary practice mode by 609 or 23% of respondents. Fig. 4B demonstrates individual practice versussingle-specialty group by age, suggesting that as the respondents age they aremore likely to become solo practitioners.
The compensation profile by all respondents is shown in Figs. 5A to 5C. Fee-for-service in 1992 was, by far, themost common mode of compensation (Fig. 5A).Fig. 5B shows distribution of fee-for-service versussalary compensation by type of practice; the percentage of salary compensationis highest in multispecialty group practice. Fig. 5Cshows type of compensation for each region in the United States and Canada. Thehighest percentage of salaried thoracic surgeons is in the Northeast and thehighest fee-for-service ratio is in Canada.
Workload in 1992
Table III lists the mean and median number ofoperations done by respondents of this survey in all categories. Fig. 6A Fig. 6b Fig. 6C Fig. 6D Fig. 6E shows performance in 1992 of adultcardiac cases in segments of 50 cases for 2103 surgeons responding that they didadult cardiac surgery in 1992. Of note is that 57% of respondents did 150 orfewer and 36% did 100 procedures or fewer per year. Fig.6B shows the volume of procedures by age group. Most cardiac surgery volume isdone in the age groups 30 to 50 years. Practicing respondents older than 70years still do about 100 cases per year. Fig. 6C comparesmean number of adult cardiac cases performed by the type of practice; thesingle-specialty groups do the highest mean number and the health maintenanceorganizations (HMOs) do the fewest. No significant difference by region wasidentified, although the Far West did the fewest per respondent (131) (Fig. 6D). In Fig. 6E cardiac cases in 50case segments are divided by type of practice. The lower the number of cases peryear, the higher the percentage done by individuals; the higher the caseload,the greater the percentage done by surgeons in single-specialty groups. Fig. 7A Fig. 7B Fig. 7C
Fig. 7A demonstrates the number of thoserespondents performing pediatric cardiac surgery by 25 case segments per year.The percentage doing 25 cases or fewer per year is 50% of those reporting.Academic centers have the highest average caseload (Fig.7B) and, regionally, Canada's mean number of cases was higher (42) than that forany region in the United States.
The average number of general thoracic procedures for the 2262individuals reporting that they do thoracic procedures are shown inFig 8.A; 67% of individuals responding did 50 cases orfewer per year. Fig. 8B shows the mean number of cases byage designation; the highest numbers are in the 50- to 70-year age group. Inthis analysis (Fig. 8C) practitioners in HMOs andacademic centers did more general thoracic procedures than practitioners inother types of practices. Regionally Fig. 8D), surgeonsin the Northeast (87) did slightly more than surgeons in other United Statesregions, and Canada had the highest mean total (100). Again, single-specialtygroups did more general thoracic surgery than surgeons in other types ofpractice (Fig. 8E).
In Fig 9.A are numbers of individuals doingperipheral vascular surgery (mean 49 per year). More than 50% of activelypracticing respondents, 1455 of 2677 (54%), did some peripheral vascular surgeryin 1992. Of those doing peripheral vascular surgery, 65% did 50 cases or fewerper year. Fig. 9B shows the average number of proceduresby age group, suggesting that older surgeons (50 to 70 years) do more peripheralvascular surgery. In analysis of type of practice, solo practice individuals domore peripheral vascular surgery, and the Southwest region has the highest meantotal (63).
Pacemaker procedures averaged 22 per year with a median of 17, whichreflects decreasing numbers done by surgeons and increasing numbers done bycardiologists. A total of 1690 individuals (63% of respondents) indicated theydid some pacemaker operations in 1992, and 1203 (71%) did 20 pacemakeroperations or fewer per year (Fig. 10). Regions with thehighest frequency of pacemaker implantations by thoracic surgeons were theMid-Atlantic states (29) and Canada (30). The number of procedures by type ofpractice setting was evenly distributed, but the lowest mean number was found inacademic centers (18).
Overall caseload including "other" (probably general surgery)averaged 249 cases per year with a median of 228 for all respondents.
DISCUSSION
This is the first comprehensive survey of workforce data for thoracicsurgeons in the United States and Canada since 1985. 1 These data have pointed out theincreasing workload in adult cardiac surgery and general thoracic surgery anddecreasing workload in peripheral vascular surgery and pacemaker implantationwhen compared with previous surveys in 1976, 21980, 3 and 1985. 1 The data reported here represents87% of the AATS/STS membership, a reasonably accurate estimate of the vastmajority of cardiothoracic operative activity in the United States and Canada in1992.
These data, although important, do have certain limitations First, thisis a questionnaire and actual numbers depend on accurate input from busysurgeons. Second, are the total numbers reasonably accurate and representativeof what thoracic surgeons do? We consulted Abt Associates of Cambridge,Massachusetts, a consultant group, to provide a check of these data. The 1992National Hospital Discharge Service (NHDS), designed to produce nationalestimates of volumes of in-patient discharges by DRG and ICD-9 procedure codes,* was used to validate this survey. Weselected thoracic and cardiac surgery DRGs and ICD-9 codes for cardiothoracicsurgery.
Table IV is a summary of cardiothoracic proceduresfrom NHDS for 1992 summarized by Abt Associates. Because 2103 respondents in our survey said they did adult cardiac surgery ata mean of 151 cases, this equaled approximately 317,000 cases. For DRG 104, 105,106, and 107, the most common adult cardiac DRGs from the NHDS, the total was330,000 cases. The discrepancy of about 13,000 cases may reflect respondentconservatism, miscoding by hospital discharge, a small number of procedures doneby nonmembers of either society, procedures done by osteopathic surgeons, andprocedures done by nonrespondents. For the general thoracic surgery survey, 2262respondents reported performing a mean of 64 procedures per year, equaling144,768 procedures in 1992. Calculation of the common general thoracic DRGs fromNHDS totaled 159,953. This difference of about 15,000 procedures is 10% of thetotal volume, less than the 14% estimated by Loop and associates 1 in 1985 as the number of proceduresdone by non-board-certified thoracic surgeons. Our 10% figure representsthoracic procedures done by general surgeons, nonsociety members, osteopathicsurgeons, and nonresponders. Complexity of cancer protocols and integratedstaging procedures have probably helped to increasingly attract patients intothoracic surgery clinics.
Pediatric heart surgery was specifically addressed in some depth for thefirst time in this AATS/STS workforce report. With the exception of the largeacademic centers, the annual volume of pediatric surgery is low, with an overallmean of 27 cases per year and a median of 26; 50% of surgeons did fewer than 25cases per year. With 451 surgeons performing a mean of 27 cases per year, thetotal was 12,177 (23,622 were documented in 1992 by NHDS). Our data may be lowbecause of misdiagnoses and failure to report operations by nonrespondingmembers doing pediatric surgery, some operations done by nonmembers, inclusioninto the NHDS data of operations for congenital disease done in adults, ormisdiagnosis or miscoding of hospital charts. The ABTS requirement for residentperformance of congenital heart surgery is 20 operations per resident. It wouldappear that the major academic centers will be providing case material for mostresidents in training.
Analyses of peripheral vascular surgery indicated that these procedures,which include abdominal aortic surgery, infrainguinal arterial bypass, andcarotid endarterectomy, are done by the majority of practicing thoracicsurgeons, 1455 of 2677 or 54%. The 49/36 (mean/median) is virtually identicalto the report of 1985 but less than that of 1980 (mean = 56). Seven percent ofrespondents have special certification in vascular surgery under generalsurgery, and a small number of cardiothoracic resident training programs havethe ability to certify for the vascular certificate. Because all diplomates ofthe ABTS have American Board of Surgery certification, it is likely thatperipheral vascular surgery will continue to be a significant part of thepractice of cardiothoracic surgery, especially in private, single-specialtygroups.
Numbers of pacemaker operations done by thoracic surgeons are similar tothe numbers done in 1985 but were fewer than in 1980, apparently becauseincreasing numbers of pacemakers are inserted by electrophysiologiccardiologists. The number of pacemaker operations documented by the NHDS for1992 was 98,661. We documented approximately 37,000 cases from our respondents,which suggests that surgeons do about 40% of all pacemaker/implantablecardioverter defibrillator implants. The ABTS requirement of 10 pacemakeroperations per resident is now accomplished by a combination of implantablecardioverter defibrillator insertion and new transvenous pacemaker insertion.
This study documents the workforce patterns of thoracic surgery in 1992. It is only a snapshot in time, but it does provide some trends concerning thecomponents of thoracic surgical practice: cardiac, general thoracic, peripheralvascular, and pacemaker, when compared with workforce numbers of previousreports. In the last report, published in 1985, 1 detailed projections regardingfuture workforce needs were made correlating the procedure numbers, graduatingresidents, and number of training programs and thoracic operations done bynoncardiac surgeons. Projections of workforce needs in the United States at thistime are more difficult because of today's rapidly changing payment mechanismsand variation in economic credentialing. Capitation, increasing HMO marketpenetration, global fees, and fee reductions for all procedures may radicallyalter projections of workforce needs. Pricing competition in all areas will besevere, and general thoracic, pacemaker, and peripheral vascular procedures arealready being thrown into direct price competition between general and thoracicsurgeons. Also, increasing interventional cardiologic procedures such as stents,atherectomy, and percutaneous transluminal coronary angioplasty may alter whatsurgeons do. In 1992, about 100,000 more interventional cardiology procedureswere done than coronary bypass procedures 
(Table IV).However, the high incidence of lesion recurrence and paucity of good long-termresults of these procedures continue to be problems. In addition, the morecomplicated procedures such as coronary stenting may turn out to be more costlythan coronary bypass. Thus it is unclear whether coronary bypass will play agreater or lesser role in the future of treatment for coronary artery disease.
Stretching the "envelope" of operability has been increasingin all aspects of cardiothoracic care so that thoracic surgeons have beenoperating on more elderly patients, more acutely ill patients, and more patientswho are in compromised condition. Whether payors can or will sustain paymentfor the elderly, extremely ill, or very high-risk patient will be a struggle ofeconomic forces against the quest for access and quality care for all segmentsof the population.
How do the workforce numbers compare with those of other advancedcountries regarding use of certain procedures? Ebert, 4 in his address to the STS ManagedCare Conference in September 1994, pointed out that cardiac cases per 100,000population in France, Germany, and England were about one third the rate of thatof the United States, with one third the number of centers Izzat and associates 5 recently reported a practice volumeof coronary bypass in the United Kingdom. In 1992 United Kingdom surgeonsaveraged 243 coronary bypass cases per surgeon, compared with 151 adult cardiaccases per surgeon for the same period in the United States according to ourdata.
What do these numbers mean in reference to workforce needs and supply ofthoracic surgeons? A 40% growth rate in the numbers of thoracic surgeons hasbeen identified from 1965 to 1990 but a virtually flat curve since 1975, whereascardiology and gastroenterology disciplines have had 734% and 1038% growthrates, respectively, in a similar period Similarly, the number of thoracicsurgery residents has remained relatively stable: 294 in 1982 to 1983, 289 in1988 to 1989 (according to the Accreditation Council for Graduate MedicalEducation), and 333 in 1994 to 1995. The workforce data for 1992 and the rapidlychanging medical economic/reimbursement scene today obviously mandate no furtherincreases in the training of thoracic surgeons and may, in fact, suggestreduction in thoracic surgical trainees.
Workforce needs will be significantly affected by changes in health carefinancing and need for primary care physicians versus specialists 6 It has been predicted that by 2000,more than 75% of the United States population will be covered by managed careprograms, which place a major emphasis on primary care, rather than thespecialist. Managed care programs selectively contract with a limited number ofproviders who meet certain minimum volume levels. Thus these alternativereimbursement methods, such as capitation, will affect volume of servicesperformed. For example, whereas HMOs tend to serve younger patients, evenMedicare age-adjusted data show lower use in patients older than 65 years. 7 In a recent report, the Council onGraduate Medical Education 8estimated that by the year 2000 there will be a shortage of 35,000 generalistphysicians and a surplus of 115,000 specialist physicians if the presentpatterns of specialty choice and number of graduates persist.
The workforce numbers reported here are important for the profile ofthoracic surgery. Although much will change because of economic pressures inthe medical marketplace before the next report is written, our specialty shouldcontinue its challenging quest for the highest quality training of men and womento satisfy the thoracic surgical needs of the United States.
Acknowledgments
Special thanks go to The Society of Thoracic Surgeonsfor statistical services and to Don Turney for consultation and guidance.
Footnotes
*Chairman (AATS/STS). 
**Representing the STS.
***Representing the AATS.
*DRG = Diagnosis-related group: ICD = International
Classification of Disease.
References
-
Loop FD, Wilcox BR, Cunningham JN, et al.Thoracic surgery manpower: The fourth manpower study of thoracic surgery: 1985report of the Ad Hoc Committee on Manpower of The American Association forThoracic Surgery and the Society of Thoracic Surgeons. Ann Thorac Surg 1987;44:450-61.[Abstract]
-
Adkins PC, Orthner HF. The Society ofThoracic Surgeons Manpower Survey for 1976: a summary. Ann Thorac Surg 1979;28:410-2.
-
Cleveland RJ, Orthner HF, Bahnson HT, et al.Thoracic surgery manpower. The third manpower study of thoracic surgery: 1980report of the Ad Hoc Committee on Manpower of The American Association forThoracic Surgery and The Society of Thoracic Surgeons. J THORACCARDIOVASC SURG 1982;84:921-32.[Abstract]
-
Ebert P. Presentation. Society of ThoracicSurgeons Managed Care Conference. Atlanta: September 1994.
-
Izzat MB, West RR, Bryan AJ, et al. Coronaryartery bypass surgery: current practice in the United Kingdom. Br Heart J 1994;71:382-5.[Abstract/Free Full Text]
-
Physician Payment Review Commission AnnualReport, 1992.
-
Group Health Association of AmericaSourcebook on HMO Utilization Data, 1992.
-
Fourth Report--Council on Graduate MedicalEducation: Recommendation to improve access to health care through physicianworkforce reform. Washington DC: US Department of Health and Human Services,January 1994.
This article has been cited by other articles:
|
|
|
|
 
M. L. Jacobs, C. Mavroudis, J. P. Jacobs, C. I. Tchervenkov, and G. J. Pelletier
Report of the 2005 STS Congenital Heart Surgery Practice and Manpower Survey
Ann. Thorac. Surg.,
September 1, 2006;
82(3):
1152 - 1152.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
R. J. Shemin, S. W. Dziuban, L. R. Kaiser, J. E. Lowe, W. C. Nugent, M. C. Oz, D. A. Turney, and J. K. Wallace
Thoracic surgery workforce: snapshot at the end of the twentieth century and implications for the new millennium
Ann. Thorac. Surg.,
June 1, 2002;
73(6):
2014 - 2032.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
J. A. DeWeese
Thoracic surgical involvement in vascular surgery
Ann. Thorac. Surg.,
October 1, 2001;
72(4):
1116 - 1117.
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. L. Early and S. R. Roberts
Excellence and low case volume: an example of the inapplicability of volume-based credentialing
Ann. Thorac. Surg.,
January 1, 2000;
69(1):
146 - 150.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
F. L. Grover
The Bright Future of Cardiothoracic Surgery in the Era of Changing Healthcare Delivery
Ann. Thorac. Surg.,
February 1, 1996;
61(2):
499 - 510.
[Abstract]
[Full Text]
|
|
|
|
|
|
|
F. A. Crawford Jr, R. P. Anderson, R. E. Clark, F. L. Grover, N. T. Kouchoukos, J. A. Waldhausen, and B. R. Wilcox
Volume Requirements for Cardiac Surgery Credentialing: A Critical Examination
Ann. Thorac. Surg.,
January 1, 1996;
61(1):
12 - 16.
[Abstract]
[Full Text]
|
|
|
