| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2002;73:2014-2032
© 2002 The Society of Thoracic Surgeons
a Department of Cardiothoracic Surgery, Boston Medical Center, Boston, Massachusetts, USA
Accepted for publication March 2, 2002.
* Address reprint requests to Dr Shemin, Department of Cardiothoracic Surgery, Boston Medical Center, 88 E Newton St, Boston, MA 02118 USA
e-mail: richard.shemin{at}bmc.org
Abstract
Background. An appropriately sized physician workforce is an essential component for healthcare planning. Contemporary economic forces and the impact of managed care have stimulated renewed interest in understanding and monitoring workforce issues.
Methods. Between August and December of 1999, the Workforce Committee of the American Association for Thoracic Surgery and The Society of Thoracic Surgeons conducted a survey detailing demographic, geographic, and socioeconomic practice characteristics of the membership.
Results. The 2,515 returned surveys represented a 62.6% return rate (± 1.6% margin of error), providing a highly statistically valid sample. Active thoracic surgeons are a mean of 50 years old, 98% men, with a high level of career satisfaction. Length of training and educational debt has been escalating. Adult cardiac surgical case volumes have increased (mean, 225 cases) and 52% of adult cardiac surgeons perform general thoracic procedures. Surgeons work 67 hours/week and 47 weeks/year. Most are organized in single specialty groups (50%) in urban areas (59%) as for-profit corporations (77%). Half of the respondents will be retired a mean of 13 years from now corresponding to a median calendar year of 2011.
Conclusions. This survey highlights significant workforce retirement during the next 10 to 15 years. Currently, the workforce in thoracic surgery appears "right sized." These valuable data provide a profile of the specialty. The information enhances our ability to strategically plan.
The national healthcare debate during the past decade has stimulated an intense demand for data to provide a better understanding of our healthcare resources and practices. Healthcare economics have taken center stage as the nation focused on rising deficits, Medicare growth, the international comparisons of healthcare costs compared to health status outcomes, and the explosion in costly new technologies. Simultaneously the funding of graduate medical education and the size of the American physician workforce have become essential components for future healthcare planning [1–12].
The rapid growth in managed care during the past decade coupled with the reality that the baby boomer population has begun to approach Medicare status has complicated projections of the physician workforce. The ratio of family practice generalists to specialist has been a focus of debate in legislative and medical forums. Government policy and economic incentives have produced an unprecedented shift to the generalist [13–18].
In this setting the Workforce Committee of the American Association for Thoracic Surgery (AATS) and The Society of Thoracic Surgeons (STS) embarked on a membership survey. Previous surveys were performed in 1974, 1976, 1980, 1985, and 1992 [19–23]. The goal was to determine the memberships’ demographics, work volume, and practice patterns in thoracic surgery at the end of the twentieth century. In addition important new data were collected about length of training, work variations, educational debt, and practice-related economic variables allowing for a better understanding of our specialty. The survey provides a snapshot in time and comparison to previous surveys.
The Workforce Committee collaborated with The Center for Outcomes Research and Evaluation at Maine Medical Center and the Center for Evaluative Clinical Sciences at Dartmouth Medical School in a research effort that was published as The Dartmouth Atlas of Cardiovascular Healthcare [33]. This publication dealt primarily with cardiovascular regional practice variations in the Medicare population and projections for Medicare growth and the number of cardiovascular practitioners (cardiologists and cardiac surgeons). These data have provided complementary information and has enhanced predications about the thoracic surgery workforce into the new millennium.
Material and methods
The AATS/STS Workforce Committee designed a comprehensive seven-page self-administered questionnaire detailing demographic and geographic practice characteristics (Appendix). This survey instrument was considerably more complex than those used in previous surveys, containing questions that required the member to seek information that would not be readily available from memory.
The first mass mailing of the confidential and blinded survey to the combined AATS and STS (4,018) memberships was on August 9, 1999. This mailing yielded a response of 1,454, a 36.2% return. A second mailing was sent to members who had not yet participated on November 3, 1999. The cumulative response was 2,012, yielding a 50.1% response rate. The third and final mailing was sent to the balance of AATS/STS members on December 16, 1999. The final effort yielded 2,515 responses (62.6%). This response rate provided a highly statistically significant survey. In May 2000, the decision not to pursue further responses was made and analysis was begun. The anonymity of the respondent was preserved by the two-envelope system used in previous surveys.
The data were carefully entered using quality control and verification measures into a comprehensive database. Each survey response was checked for inconsistencies and errors, and then cleaned. The survey data were downloaded and rigorously analyzed using the SPSS statistical software. The analysis was independently performed by the Market Research and Statistics Division of Smith Bucklin & Associates, Inc (Chicago, IL). The data elements were checked for accuracy to ensure the appropriate respondent groups were chosen according to their demographic information relative to answering specific questions. For example, Canadian or other international members and retired versus actively practicing surgeons were analyzed separately or excluded to have the appropriate and relevant group survey answers for a specific question. Career satisfactions, gender, length of training, educational debt, and projections about retirement age were among the important new data elements added to the current survey.
Results
The Workforce survey received 2,515 responses of a possible 4,018. This represents a ±1.6% margin of error providing a highly statistically valid sample (ie, 99% confident that a survey of 4,018 would reveal results within 1.6% of those shown in this report). Individual questions varied in the number of responses. Specific questions related to clinical practice volume or economics were analyzed from responses provided only by actively practicing surgeons and surgeons who retired within the past year. Active members practicing in the United States comprised 1,923 (76.5%) of the sample, retired members 546 (21.7%) and foreign members 46 (1.8%).
Age
The mean age of active thoracic surgeon respondents was 50 years (median, 49 years). The mean age of retired surgeons was 72 years (range, 42 to 96 years). Figure 1
depicts the age distribution of actively practicing thoracic surgeons in the United States. Figure 2
demonstrates the age trend over the five Thoracic Surgery Workforce studies. From 1976 to 1985, the mean age of actively practicing thoracic surgeons had been increasing. Age peaked in the 1992 survey at 52 years, decreasing to 50 years in the current survey.
|
|
Retired
Retired thoracic surgeons (546) represented 22% of the combined AATS/STS respondents, of these 12% were recent retirees having retired within the past year. Among recent retirees, 2% retired in their fourth decade, 19% in their fifth decade, 56% in their sixth decade, and 18% in their seventh decade, 5% in their eight decade (Fig 3).
|
|
Career satisfaction was also correlated with the number of hours a thoracic surgeon worked per week. The "extremely satisfied" comprised 52% of those surgeons working 40 hours or less, compared to 35% (41 to 50 hours), 38% (51 to 60 hours), 39% (61 to 70 hours), 37% (71 to 80 hours), and 36% (>80 hours). However, when the "extremely" and "very satisfied" respondents were combined the hours worked per week did not appreciatively affect career satisfaction: 78% (1 to 40 hours), 70% (41 to 50 hours), 72% (51 to 60 hours), 73% (61 to 70 hours), 73% (71 to 80 hours), and 68% (>80 hours).
The comparison between academic and nonacademic surgeons revealed 54% versus 34% responding as "extremely satisfied," respectively (p < 0.0001). Among full-time academic faculty, career satisfaction was highest for chairmen and directors with 55% responding as "extremely satisfied." Faculty members responding as either "extremely satisfied" or "very satisfied" were 93% for department chairmen, 81% professors, 79% associate professors, and 75% for assistant professors (p = 0.01).
Length of training
Figure 5A
depicts the number of years respondents trained after graduation from medical school. The mean was 8.3 years, with a median of 8.0 years. American Board of Surgery eligibility is a requirement for entering a thoracic surgery training program. The minimum length of general surgery training is 5 years added to the 2 to 3 years of required thoracic surgery training, establishing a minimum of 7 to 8 years to train a thoracic surgeon. However, 33% of respondents have trained 9 years or longer. The length of training has been trending upward. Of thoracic surgeons graduating from medical school between 1950 and 1959, 27% trained more than 8 years. Those graduating from medical school during the years 1970 to 1979 and who trained more than 8 years increased to 29%, and escalated to 42% for graduates from 1980 to 1989 (Fig 5B).
|
|
Specialty self-designation
The largest self-designated group of thoracic surgeons was as adult cardiac surgeons (75%). The self-designation as general thoracic surgeon was 64%, 10% as pediatric cardiac surgeons, and 31% as peripheral vascular surgeons. Obvious overlap was present as many surgeons chose multiple designations. This has been an issue in previous workforce surveys. Further analysis was possible by regrouping respondents by the case volume they actually reported in each specialty category. For example, those respondents who only reported performing pediatric cardiac procedures were one group, those only performing adult cardiac were another, whereas those performing adult cardiac and general thoracic were a third group, and surgeons performing only general thoracic procedures composed a fourth group. Other combinations were not analyzed in detail because the numbers were too small.
Overall, respondents had a median case volume of 262 patients, a mean of 279 (Fig 7A). The distribution was 7% performing less than 100 cases, 21% performing 101 to 200 cases, 36% performing 201 to 300 cases, and 36% performing more than 300 cases. Survey data are unaudited and have inherent inaccuracy. The discrepancy between the mean and median coupled with some surgeons claiming very large caseloads, led to a reanalysis of the caseload volumes setting a upper limit of 400 patients. The new mean was 237 patients/year.
|
The mean case volume among respondents who performed only pediatric cardiac operations (n = 70) was 217 patients/year (median, 200 patients/year). The case volume distribution was 13% performing 0 to 100 cases, 41% performing 101 to 200 cases, 35% performing 201 to 300 cases, 6% performing 301 to 400 cases, and 4% performing more than 400 cases (Fig 7B). Reanalysis without the four surgeons performing more than 400 cases/year, reduced the mean to 200 cases/year (median, 200 cases/year).
The number of thoracic surgical respondents performing adult cardiac procedures was 926. Only 408 (44%) performed only adult cardiac procedures. The adult cardiac surgeons also performing general thoracic procedures were 442 (48%); 49 surgeons performed adult cardiac and pediatric procedures (5%); and 27(3%) performed adult cardiac, general thoracic, and pediatric procedures. Peripheral vascular surgical procedures were performed by 214 adult cardiac surgeons, with a mean annual case volume of 50 patients.
The caseload for cardiac surgeons reporting only adult cardiac operations was a mean of 225 patients/year (median, 200 patients/year). The distribution was 8% performing 0 to 100 cases, 44% performing 101 to 200 cases, 34% performing 201 to 300 cases, 8% performing 301 to 400 cases, and 5% performing more than 400 cases (Fig 7B). Reanalysis without the 33 surgeons performing more than 400 cases/year, reduced the mean to 201 cases/year (median, 200 cases/year).
The annual caseload for adult cardiac surgeons also performing general thoracic surgery was a mean of 251 cases (median, 250 cases). The mean numbers of adult cardiac cases were 174 and 63 (mean) general thoracic cases. The total caseload distribution for these surgeons was 4% (0 to 100 patients), 23% (101 to 200 patients), 39% (201 to 300 patients), 24% (301 to 400 patients), and 10% (>400 patients).
The respondents reporting only a general thoracic caseload (n = 195) reported a mean of 211 patients/year (median, 200 patients/year. The volume distribution was 31% of surgeons performing 0 to 100 cases, 23% performing 100 to 200 cases, 20% performing 201 to 300 cases, 11% performing 301 to 400 cases, and 10% performing more than 400 cases (Fig 7B).
Work intensity
Respondents worked a mean of 67 hours/week (median, 65 hours/week). The work distribution showed that 16% of surgeons worked 50 hours or less, 30% worked 51 to 60 hours, 24% worked 61 to 70 hours, 21% worked 71 to 80 hours, and 9% more than 80 hours/week.
The mean number weeks per year worked by respondents were 47 (median, 48 weeks/year). The distribution was that 72% of surgeons worked 46 to 50 weeks/year and 7% worked 51 to 52 weeks/year.
Career satisfaction did not significantly vary by work hours per week or weeks per year.
Organizations and membership
Thoracic surgeons belonged to a mean of 4.5 organizations per member with a range of 1 to 13. Because the study was designed to query members of the AATS/STS, it is not surprising that 96% of respondents were STS members. Membership in other selected national organizations was 85% American College of Surgeons, 47% American College of Chest Physicians, 45% American Medical Association, 41% American Heart Association, 39% American College of Cardiology, 21% American Association for Thoracic Surgery, 19% International Heart and Lung Transplant Society, and 6% in the American Surgical Association.
The mean number of professional meetings attended per year was 3.0 (40% attended two meetings per year, whereas 46% attend three or more). Fifty-one percent of the retired group attends two meetings and 41% attended three or more meetings per year.
Practice organization
The most common mode of practice was the single specialty group practice (50%), followed by 21% full-time academic, 15% multispecialty group practice, and 13% solo practitioners. Most new practices started in the past 5 years were single specialty groups (46%), academic (14%), multispecialty groups (16%), and solo (22%).
The number of years respondents had been in their current practices was 14% for less than 5 years, 20% for 6 to 10 years, and 65% for 11 years or more.
Practice location was 59% urban, 18% suburban, 21% small to median sized community, and 2% rural. The regional map illustrates the states that were analyzed within each region (Fig 8). Urban practice is most common in the North East and Far West regions, but ranged from 45% to 71% in all regions. Rural practice was more common in the Rocky Mountain and Southeast regions.
|
The most common hospitals starting a new program during the past 5 years were community hospital (58%), private medical school affiliated hospital (20%), private nonmedical school affiliated (8%), and university teaching hospitals (4%).
The numbers of surgeons comprising practice groups were four or less in 50% of practices. The distribution of surgical group sizes was 24% had 1 to 2 surgeons, 26% had 3 to 4, 19% had 5 to 6, 22% had 7 to 12, and 9% had 13 or more. The multispecialty group practice was most likely to have 13 or more members (45%).
The legal forms of practice organization were 9% sole proprietorship, 23% partnership, 60% corporation, and 7% other (Fig 9A). The sole proprietorship had dropped from 23% in the 1992 survey. A corporation was the most common organization in all regions (range, 55% to 75%). Sole proprietorships were a minority in all regions, ranging from 3% to 13%. Most partnerships were in the Southeast with 37% (Fig 9B).
|
Hospital
The sizes of hospitals where respondents practiced were less than 300 beds in 30%, 301 to 500 beds in 39%, and more than 500 beds in 31%. The 100- to 300-bed hospitals were the most common hospital size in the Rocky Mountain region (55%), Far West (44%), and Southwest (37%) regions. The 300- to 500-bed hospitals predominated in all other regions except the Southeast, where the more than 500-bed institution slightly predominated.
The respondents operated at a mean of 2.5 hospitals (median, 2.0 hospitals). The distribution was 36% operate at one hospital, 30% at two, 16% at three, and 18% four or more. The Northeast region stands out as the only region where more than half of the thoracic surgeons practice in a single hospital (55%), second is the Mid-Atlantic with 42%, and all other regions range from 40% to as low as 25%. The Central, South Central, Far West, and Southwest regions stand out with a large number of thoracic surgeons (approximately 25%) traveling to four or more hospitals.
Academia
Among surgeon respondents 35% were associated with teaching in a cardiothoracic surgical training program. The Northeast and Mid-Atlantic regions stand out with 57% and 50% of respondents, respectively, indicating that they were teaching in a cardiothoracic program. The Southeast and Rocky Mountain regions had the greatest numbers of nonteaching surgeons (80% and 77%) (Fig 10).
|
The academic faculty’s legal form of organization is 49% as a corporation, 21% as a partnership, and 7% as a sole proprietorship. This is similar to the distribution among nonacademic faculty 63%, 24%, and 10%, respectively. Both academic and nonacademic had "other forms" of organization 23% and 3%, respectively.
The distribution of academic ranks for self-designated pediatric cardiac, adult cardiac, and general thoracic surgeons were 7%, 12%, and 11% department chairman, 30%, 34%, and 34% professor, 37%, 27%, and 25% associate professor, and 26%, 27%, and 30% assistant professor.
Malpractice expense
A major practice expense was the cost of malpractice insurance. The mean was $48,930 (median, $30,000). The distribution of malpractice premiums varied widely in each region. Highest mean malpractice costs appeared in the Mid-Atlantic region at $53,000. The premium distribution in each region is demonstrated in Figure 11A.
|
The premium costs by mode of practice were noteworthy for academic full-time faculty, with 34% paying the lowest premiums (less than $20,000) and 21% paying the highest premiums (more than $50,000) (Fig 11C).
Incme
The impact of physician payment reform, Medicare budget neutrality, and practice expense adjustments, managed care contracting, and capitation has had a significant impact on surgeon income. During the past 3 years 48% of surgeons had a more than 10% reduction in income without adjusting for inflation.
The single specialty group practices experienced the greatest gains and loss in each category. Among surgeons with a 16% to 25% reduction in income, 64% were in a single specialty group. Salary increases by academic rank was greatest for the assistant professors. They comprised 70% of those with a more than 25% increase in income. All other categories experienced losses, which were greatest among professors and associate professors.
Salary income was structured as fee for service in 41%, straight salary in 38%, and salary with incentive in 20% of respondents. Fee for service was most prevalent in the Southwest region (55%), but predominated in all regions except the Northeast (31%) and Mid-Atlantic (33%). In the Northeast (46%) and Mid-Atlantic (45%) regions straight salaries predominated. Fee for service predominated (77%) in sole proprietorships, 34% of multispecialty groups, 46% single specialty groups, and 11% of academic groups. Straight salary predominated in academic practices (54%) and multispecialty groups (46%).
Retirement
Three survey questions asked for retirement information. Active surgeons projected a mean of 12.9 (median, 11) additional years to remain in practice (Fig 12).
Among respondents, 11% planned to retire at less than 55 years of age, 26% between the ages of 56 to 60 years, 40% at 61 to 65 years, 18% at 66 to 70 years, and 5% at more than 71 years of age. In a separate question the calendar year the surgeon expected to retire was requested. The median calendar year the respondents projected for retirement is 2011.
|
In academia the mean number of additional years of practice by academic rank revealed 11 years for chairmen, 9 years for professors, 14 years for associate professors, and 20 years for assistant professors.
Additional years until retirement by mode of practice was 10 years for solo practitioners, 13 years for multispecialty groups, 13 years for single-specialty groups, and 15 years for academic practice.
A regional analysis of remaining practice years revealed a mean range of 12 to 14 years with similar distributions in all regions.
Analysis of years until retirement by self-designation revealed 14 years for pediatric cardiac surgeons, 13 years for general thoracic surgeons, and 13 years for adult cardiac surgeons.
Practice revenue
Practice revenue came from two main sources: office visits 6.7% and surgical procedures 91%. Other sources (3%) included administration and teaching stipends.
The major third party payers contributing to the practice revenue were Medicare 47%, private insurance 21%, HMO contracts 19%, Medicaid 8%, and others 2%. The clinical volume that contributed to the respondent’s practice by payer type was 51% Medicare, 18% privately insured, 17% HMO patients, 3% self pay, and others 9%. Capitated Medicare contracts were held by only 32% of surgeons. These contracts contributed only 8% of these surgeons’ total practice revenues. The growth in managed care was indicated by 45% of respondents indicating they had a "significant increase" in managed care contracts and 33% responding that they had a "somewhat increase." The impact of managed care clinical decisions have been a theoretical concern. A compromise in usual standards when caring for patients enrolled in managed care plans was reported as "always" (4%), "often" (15%), "sometimes" (27%)," rarely" (33%), and "never" (21%) by surgeons operating on these patients.
Billing, benefits, full time equivalents
The majority (75%) of surgeons bill directly to third party carriers, 17% use an outside billing service, 5% do both, and 3% responded "other." Practice benefits included malpractice coverage (86%), medical insurance (85%), retirement plan (72%), life insurance (53%), dental insurance (55%), long-term disability (54%), education and training allowances (52%), travel allowances (50%), profit sharing (37%), automobile (17%), and a dependent educational allowance (12%). The mean number of surgeons in a practice was 7 (median, 4). The mean FTEs employed by a practice was 11.5 full-time and 6.1 part-time employees.
Comment
This workforce study only provides a snapshot in time and ignores the dynamics of multiple interacting variables affecting the numbers of actively practicing thoracic surgeons, their distribution, and workload. Demographics and case volume have been extensively studied in previous Thoracic Surgery Workforce studies. However, survey responses are highly dependent on how a question is worded. These differences limited opportunities for comparison and trending.
This study was significantly more comprehensive then previous surveys. This fact is undoubtedly the reason for a reduction in the percentage of returned responses. However, the 63% return rate yielded a highly statistically valid survey with a ±1.6 margin of error. This survey provided valuable new data on career satisfaction, length of training, educational debt, memberships and meeting attendance, academia, practice modes, organizational formats, board certification, and projections for retirement. The new information affords a better understanding of our specialty and provides the leadership of our professional societies with valuable data when providing testimony to regulatory or legislative committees and for strategic planning.
To predict the number of surgeons needed in thoracic surgery during the next 10 to 20 years is complex and fraught with potential error [5, 24–26, 33]. The data from this survey are helpful, but insufficient to make accurate predictions. The current workforce is performing more operations than in previous surveys. Adult cardiac surgeons continue to perform a significant number of general thoracic procedures in addition to adult cardiac procedures and occasional pediatric cases. This finding has been observed in previous surveys. Our policy of providing broad training and not fragmenting the specialty of thoracic surgery has been wise.
General thoracic surgeons enjoy significant caseloads and remain an area for future growth. This area is increasingly popular among female surgeons [27]. There also is a significant volume of vascular procedures being performed by adult cardiac surgeons. This observation requires further future analysis as cardiac and vascular procedures are particularly strong in several regions and in private practice. The specialty of vascular surgery is extremely vulnerable in the current environment of aggressive catheter-based interventional techniques used by cardiologists and radiologists. The practice of cardiovascular surgery may become an increasingly popular trend. During the past two decades thoracic surgeons have allowed the high volumes of cardiac surgery to lure themselves into complacency and many traditional procedures of the specialty have been given away to medical colleagues (eg, pacemakers, automatic implantable cardioverter-defibrillators, bronchoscopy, chest tube placement) [28].
The data show the mean age of our workforce is constant at 50 years old. Most surgeons are projecting a retirement before or by the age 65 years. In the last survey, many surgeons practiced past the age of 70 years, performing significant caseloads [23]. The reason for an earlier planned retirement is not known. Better retirement planning, radical changes in reimbursement, raising overhead, the medical-legal environment, and career satisfaction are certainly important factors affecting attrition from an active thoracic surgeons workforce. It appears by the year 2010, 50% of currently practicing thoracic surgeons project retirement. This finding is similar in all regions and among academic surgeons. However, the recent changes in the stock market effecting retirement portfolios and the net worth may significantly change the projections for retirement, especially among those planning retirement within the short term.
The numbers of training programs that are approved by the Residency Review Committee on Thoracic Surgery are approximately 90, producing approximately 142 new graduates yearly. The American Board of Thoracic Surgery awards approximately 142 new certificates yearly. The disease burden in our society requiring the services of a thoracic surgeon is unlikely to diminish as our population ages and baby boomers enter their senior years. One cannot predict the impact future advances in medical therapy or nonsurgical interventional procedures may have on the thoracic surgical workload. The use of new stent technology and aggressive lipid therapy with plaque stabilization may result in a significant reduction in coronary artery bypass procedures, which would impact caseload. These changes could acutely affect workload and may not allow time for the specialty to adjust the workforce because of the long "pipeline" of the thoracic surgical training.
The rapid decline in per case reimbursement is likely to stabilize as the major reductions due to the decrease in practice expense to thoracic surgery is in the final year of implementation. The decade of the 1990s produced a more than 50% reduction in Medicare reimbursement for cardiac and thoracic procedures when the value of the dollar was adjusted for inflation [29].
The forces affecting the striking reduction in number of young physicians attracted to our specialty are a serious concern. In the 2001 residency match, 88% of the 149 active applicants matched in 131 of 141 available positions. Only 112 applicants were graduates of American medical schools. Thus, the ratio of applications to available training slots is nearly 1:1. The specialty’s attraction to women, who comprise 50% of medical school graduates and 25% of general surgical residents, continues to be inadequate. Our future workforce may be plagued by an undersupply of best-qualified candidates graduating from US medical schools instead of an oversupply of surgeons. Much of the data from the current study, which documents the increasing length of training, increasing educational debt, professional effort reflected in the hours per week and weeks per years worked, declining income, and erosion in professional satisfaction provides insight into many of the factors contributing to this disturbing trend [30, 31]. The recent decision to eliminate the requirement for American Board of Surgery certification will help shorten the time to train a thoracic surgeon.
The Dartmouth Atlas of Cardiovascular Healthcare furthers our understanding of regional variations in practice patterns, case volume per 1,000 Medicare enrollees, numbers of surgeons in various regions and their case volume [33]. One can model the number of surgeons needed to practice in a referral region by taking into account case volume and the disease burden in the age- and sex-adjusted population.
This model seeks to produce optimal health outcomes with the appropriate number of practitioners performing procedures using evidence-based criteria and national benchmarking to best practices [32]. Population-based benchmarking offers practical advantages to the more traditional needs-based or demand-based planning when estimating a reasonably sized per capita workforce.
During the past 25 years, the US physician workforce has grown significantly in large part as a response to federal policies that responded to a perceived physician shortage in the 1950 to 1960s. A more rapid rate of growth in specialists than in primary care physicians occurred. The number of cardiologists (1975 to 1995) grew from 6,000 to 15,500, whereas the number of cardiothoracic surgeons remained stable at about 4,000. During this time, all specialists grew by 35% and primary care physicians by 25%.
The Dartmouth Atlas investigators attempted to project the future supply of cardiovascular specialists adjusting for population demographics, age, and sex of the professional workforce. The workforce involved in active patient care must account for new physicians entering the workforce, the number of physicians that will die or retire, the partial work effort reflected by female practitioners (ie, less work hours and time away while raising a family) and the physicians or surgeons involved in nonclinical activities (eg, teaching, research, and administration) [33]. This latter group accounts for the active practitioners performing very few to no operations. Nevertheless, they fill important roles in the thoracic surgical workforce. Adequate funding of teaching and administration may help keep senior surgeons from earlier retirement to fill these important roles.
The number of cardiothoracic surgeons per 100,000 people is projected to experience minimal growth from 1.1 to 1.3 surgeons per 100,000 from 1995 to 2005 and then decline to 0.9 by 2020. The impact of an arbitrary 50% reduction in approved thoracic surgery residency positions (once a serious consideration by the US Congress) would severely reduce the cardiothoracic workforce and markedly impact the number of surgeons within 5 years. Even doubling the number of positions would only result in a steady increase in the number of surgeons to 1.8 per 100,000 by 2010; afterward the number would remain constant. This number is still significantly lower than the current 2.0 surgeons per 100,000 currently practicing in the Medicare regional referral region of Metairie, LA [33].
Workforce projections for noncardiac (general thoracic) surgeons are complicated by the significant volume of procedures being performed by general surgeons. The regional distribution of general thoracic surgeons is extremely variable. They range from more than 30 per hospital referral region to none. The highest density is in Boston (30), eastern Long Island and New York (30), Manhattan (22), and Los Angeles (16). The significant maldistribution requires other surgeons to provide thoracic surgical services especially in urgent medical situations.
The data from the Dartmouth Atlas estimated that 44% of the 23,452 lung procedures performed on Medicare enrollees were performed by cardiac surgeons, 31% by general thoracic surgeons, and 25% by general surgeons. These data are consistent with the data from the current workforce survey where 48% of adult cardiac surgeons perform general thoracic procedure with a mean of 63 thoracic and 174 cardiac cases annually. General surgeons made up 70% of the surgeons billing Medicare for fewer than five pulmonary procedures annually but less than 17% of those surgeons billing for more than 10 procedures annually. General thoracic surgeons perform and bill Medicare for the highest volumes as individual surgical practitioners.
In summary, this report provides valuable data to profile our specialty at the end the century as we deal with the challenges of the new millennium. The specialty appears to be "right sized" for the predictable next decade. We need to focus on recruitment of the superb talent our specialty has attracted in the past. Reaching out and mentoring medical students will allow us to share the unique work we perform in the healthcare system. The problems of increasing educational debt, length of training, and lifestyle issues related to long work hours in the milieu of radical reductions in reimbursement will not have easy solutions.
Future workforce studies must continue to document the trends that have already been identified so that timely corrective actions can be planned and implemented. Major and expensive workforce surveys like the current study requires extensive effort precluding the ability to perform them frequently. The challenge is to evolve to a Web-based annual survey that can be rapidly analyzed. In this way, timely data can identify the areas to focus our efforts. To improve the reliability of survey data, large numbers of respondents accurately answering the questions are essential. The membership must develop the habit of viewing the Web pages of our professional societies and respond to surveys to become an active participant in forging our future.
Appendix. Sample of survey used by the workforce
|
|
|
|
|
|
|
References
This article has been cited by other articles:
|
|
 |
|
  M. B. Ratcliffe, C. Howard, M. Mann, and P. del Nido National Institutes of Health funding for cardiothoracic surgical research. J. Thorac. Cardiovasc. Surg., August 1, 2008; 136(2): 392 - 397. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. L. Grover At Last, Inequities in Reimbursement Modified by Real Evidence-Based Data Ann. Thorac. Surg., January 1, 2007; 83(1): 9 - 11. [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
|
|
V. L. Gott, N. D. Patel, S. C. Yang, and W. A. Baumgartner Attracting outstanding students (premedical and medical) to a career in cardiothoracic surgery. Ann. Thorac. Surg., July 1, 2006; 82(1): 1 - 3. [Full Text] [PDF]
|
|
|
|
|
|
F. A. Crawford Jr Thoracic Surgery Education-Past, Present, and Future Ann. Thorac. Surg., June 1, 2005; 79(6): S2232 - S2237. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Mayer and A. C. Skinner Too Many, Too Few, Too Concentrated?: A Review of the Pediatric Subspecialty Workforce Literature Arch Pediatr Adolesc Med, December 1, 2004; 158(12): 1158 - 1165. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Guezuraga and D. Y. Steinbring View from industry Eur. J. Cardiothorac. Surg., December 1, 2004; 26(Suppl_1): S19 - S24. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. J. Gardner Residency Training for the Future, Not the Past Ann. Thorac. Surg., November 1, 2004; 78(5): 1519 - 1521. [Full Text] [PDF]
|
|
|
|
|
|
J. D. Salazar, R. Lee, G. H. Wheatley III, and J. R. Doty Are There Enough Jobs in Cardiothoracic Surgery? The Thoracic Surgery Residents Association Job Placement Survey for Finishing Residents Ann. Thorac. Surg., November 1, 2004; 78(5): 1523 - 1527. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. B. Fye Introduction: the origins and implications of a growing shortage of cardiologists J. Am. Coll. Cardiol., July 21, 2004; 44(2): 221 - 232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Baughman, C. J. Pepine, J. Abrams, T. M. Bashore, R. M. Califf, A. M. Feldman, W. B. Fye, B. P. Griffin, R. J. Myerburg, G. V. Naccarelli, et al. Working group 1: How to increase the output of cardiologists J. Am. Coll. Cardiol., July 21, 2004; 44(2): 233 - 237. [Full Text] [PDF]
|
|
|
|
|
|
F. A. Crawford Jr Presidential address: thoracic surgery education--responding to a changing environment J. Thorac. Cardiovasc. Surg., November 1, 2003; 126(5): 1235 - 1242. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
