PHYSICIAN WORKFORCE STUDY

B: S AMPLE C HARACTERISTICS 76 A PPENDIX C: D ETAILED R ESULTSFOR Q UESTIONS D ISCUSSEDIN R EPORT 80 A PPENDIX D:B IBLIOGRAPHY 86 I NDEXTO T ABLESINSIDEBACKCOVER Copyright © 2003 Massachusetts Medical Society. All rights reserved. For more information, contact the MMS Department of Health Policy/Health Systems at (781) 434-7222.<br><br> This report is also available online at http://www.massmed.org/pages/workforce.asp . T HE M ASSACHUSETTS M EDICAL S OCIETY 9 S P HYSICIAN W ORKFORCE S TUDY E XECUTIVE S UMMARY Introduction W ith the help of prominent labor economists, the Committee on Medical Service of the Massachusetts Medical Society completed a study that builds upon the results of the 2002 Physician Workforce Study. The results of the 2003 study indicate the following: All fourteen specialties surveyed are currently experiencing extreme- ly tight labor markets.<br><br> As a result physicians have been forced to react to these labor market shortages by increasing work hours (48%), adjusting professional staffing (37%), and altering the ser- vices they provide (31%). Five specialties are experiencing a critical physician shortage: anesthesiology, cardiology, gastroenterology, neurosurgery, and radiology. Three additional specialties are experiencing a severe shortage: general surgery, orthopedics, and vascular surgery.<br><br> Community hospitals are experiencing a much tighter labor market than teaching hospitals. Professional liability concerns are influencing physicians to contem- plate career changes. This is particularly true within high-risk spe- cialties such as obstetrics and gynecology (OB/GYN) and surgical subspecialties.<br><br> Thirty-two percent of practicing physician respondents are either planning on or considering leaving Massachusetts because of the current practice environment. The average number of months required to recruit a physician is roughly twelve; by specialty, it ranges from eight months for emer- gency medicine to twenty-seven months for neurosurgery. Regional disparities in the labor market exist.<br><br> This is particularly evident in the Springfield metropolitan area, where physician labor shortages are more acute. 1 The MMS Index 1 , which tracks the physician practice environment quantitatively, also indicates that physicians are facing a very diffi- cult and challenging environment in which to care for their patients. The results of this report support the conclusions of the 2002 Physician Workforce Study in terms of recruitment, retention, and staffing concerns.<br><br> In addition, this report contains additional information on professional liability, physician satisfaction, issues in community hospitals, and other topics that fur- ther reinforce the conclusion that the physician workforce in the Commonwealth of Massachusetts is experiencing a shortage. Report Summary In the 2002 Physician Workforce Study, the Committee on Medical Service found that the labor market situation among Massachusetts physicians could only be described as in a crisis state. The Massachusetts Medical Society (MMS), under the direction of the Committee on Medical Service and working with leading labor economists, James Howell, Ph.D., Carol Simon, Ph.D., and Andrew Sum, Ph.D., designed a comprehensive research methodology that included both primary (i.e., surveys) and secondary (i.e., existing databases, literature reviews) data collection.<br><br> The 2003 Physician Workforce Study validates our previous conclusions concerning the crisis in the physician community. The extraordinarily negative response to all of the questions relating to the current availability of physicians, as well as the degree of difficulty in recruiting and retaining physicians to fill posi- tions, highlights the problem. Methodology The MMS physician workforce study used information from seven primary data sources: 1.<br><br> Survey of practicing physicians in community and hospital settings throughout Massachusetts; 2 1 Massachusetts Medical Society website available at http://www.massmed.org/pages/mmsindex0302.asp (accessed 14 April 2003). 2. Survey of medical staff presidents in community hospitals; 3.<br><br> Survey of department chiefs from teaching hospitals; 4. Survey of residents and fellows in their last year of training; 5. Survey of residency and fellowship program directors; 6.<br><br> Survey of biotechnology companies; and 7. An opinion poll of patients in Massachusetts. The response rates for each of the surveys are described in Table 1.<br><br> Secondary sources of information included articles from peer-reviewed journals, existing databases, the 2002 Physician Workforce Study, and the MMS Index 2 , which tracks the physician practice environment quantitatively. Results Survey of Practicing Physicians Within the survey of practicing physicians were six questions designed to assess the degree of stress in physician labor markets. As we review our results, it is important to note that in similar studies of other professional labor markets responses indicating a shortage almost never exceed 10 percent.<br><br> Using this rate as a benchmark, results from this study indicate that Physician labor markets in Massachusetts are currently under extreme stress; the forces that pushed these markets into this unen- viable state are numerous and are not likely to be easily reversed. All fourteen specialties surveyed are currently experiencing suffi- ciently high labor-market vacancy rates that these labor markets are now confronting serious problems. Results from the comparative 3 2 Massachusetts Medical Society website available at http://www.massmed.org/pages/mmsindex0302.asp (accessed 14 April 2003).<br><br> T ABLE 1: R ESPONSE R ATE S UMMARY SURVEYRESPONSERATE Practicing Physicians27% Department Chiefs of Teaching Hospitals63% Medical Staff Presidents in Community Hospitals48% Residency and Fellowship Program Directors58% Residents and Fellows20% Biotechnology Companies25% analysis of specific workforce questions indicate that the following should be designated as Critical Shortages: Anesthesiology Cardiology Gastroenterology Neurosurgery Radiology Severe Shortages: General Surgery Orthopedics Vascular Surgery Approximately half of physician respondents (49%) felt that the pool of physician applicants is inadequate to fill vacant positions or expand one 9s practice. Moreover, more than two-thirds of all physician respondents (69%) indicated that they are currently experiencing some degree of diffi- culty filling vacant physician positions. In the following four spe- cialties, which are designated as having critical shortages, over 80% of physician respondents indicated that they were experiencing difficulty: Anesthesiology Gastroenterology Neurosurgery Radiology In nine additional specialties, 40% to 80% of physician respon- dents indicated that they are having difficulty: Cardiology Emergency Medicine Family Practice General Surgery Internal Medicine OB/GYN Orthopedics Psychiatry Vascular Surgery 4 Over forty percent of physician respondents (41%) reported a sig- nificant increase in the amount of time needed to recruit a new physician over the past three years.<br><br> 3 Relative to other professional occupations, these lengths of time are extraordinarily long and indi- cate lags in filling physician vacancies that can affect patients 9 access to and the availability of services. Over thirty percent of all physician respondents (31%) indicated that physician shortages have forced them to alter services. In addi- tion, 37% of all physician respondents have changed their profes- sional staffing patterns due to physician shortages.<br><br> Overall, 28% of physician respondents indicated that they are con- templating a career change due to the practice environment. The most concerning results are from the specialties of neurosurgery (56%), OB/GYN (40%), vascular surgery (39%), and emergency medicine (36%). Professional liability costs are a significant factor for physicians in their contemplation of a career change.<br><br> This is particularly evident in the specialties of neurosurgery, OB/GYN, vascular surgery, gen- eral surgery, and orthopedics where responses were 10% to 40% higher than the overall question mean. (Note: Since the conclusion of this year 9s study, there has been an article in the Boston Globe on April 17, 2003. Professional liability rates for the largest insurer in Massachusetts will increase by 20 percent on average effective July 1, 2003.<br><br> 4 ) Physicians who are contemplating a career change are considering early retirement (31%), work in a non-health care setting (13%), and developing an entrepreneurial venture (14%). Seventy-nine percent of all physician respondents indicated that they would rate the profession of medicine as either rewarding or very rewarding, but 60% are dissatisfied or very dissatisfied with the current practice environment. Seven percent of physician respondents indicated that they are planning to leave Massachusetts to practice medicine elsewhere because of the current practice environment.<br><br> This extrapolates to 1,444 physicians of the approximately 20,628 practicing physicians in Massachusetts. 5 In addition, 25% of respondents indicated that they would consider doing so if the current situation does not 5 3 A review of the 2003 survey responses indicated only sixteen responses noted a decrease in the time to recruit. Due to the overwhelming response that time to recruit had increased, only those responses were included.<br><br> In 2002, there were no responses indicating a decrease. 4 Kowalczyk L. Premiums to Rise 20% for Mass.<br><br> Doctors, Boston Globe , 17 April 2003, sec. A1. 5 The Federation of State Medical Boards of the United States, Inc.<br><br> Summary of 2001 Board Actions, 9 April 2002, available at http://www.fsmb.org (accessed 16 April 2003). change. If the sample is indicative of the entire population, this extrapolates to 5,157 additional physicians contemplating leaving the state.<br><br> Potentially, this scenario could result in 6,601 physicians leaving the Commonwealth and would result in a patient to physi- cian ratio of 448 patients per physician in Massachusetts, as com- pared with a much lower patient to physician ratio of 392 patients per physician in the United States as a whole 6 , leaving us more con- cerned about access for our patients. In conclusion, we found that Massachusetts is experiencing a shortage of all physician specialties. Further stratification of the results allowed us to desig- nate the following specialties as experiencing Critical Shortages: Anesthesiology Cardiology Gastroenterology Neurosurgery Radiology Severe Shortages: General Surgery Orthopedics Vascular Surgery Many of the physician respondents expressed their satisfaction with the cnoble profession d they chose, but they are very dissatisfied with the current state of the health care environment.<br><br> The actual comments from several physicians are included below. cDespite the current meltdown of the system, the practice of medicine is the most rewarding calling I can imagine. d cI have four sons (ages 14, 12, 10, 7). I cannot recommend medicine to them.<br><br> My father was an M.D. d cI am only 35 years old, practicing for seven years and ready to stop. I would not encourage others to go into medicine. All of my medical school friends feel the same way.<br><br> We entered this field for altruistic reasons, and reality erased our idealism. d 6 6 Kaiser Family Foundation. State Health Facts Online Demographics and the Economy, Population Distribution by Age, state data 2000 32001, U.S. 2001, available at http://www.statehealthfacts.kff.org (accessed 15 April 2003).<br><br> Survey of Medical Staff Presidents in Community Hospitals and the Department Chiefs of Teaching Hospitals Community hospitals and teaching hospitals are both feeling the impact of physician shortages. Over half (54%) of the teaching hospital chiefs and 87% of the community hospitals reported that the pool of physician applicants is inadequate to fill their vacancies. Two-thirds (67%) of the community hospitals and more than half (55%) of the teaching hospitals indicated that their ability to retain existing staff has changed in the past three years.<br><br> Of those respond- ing, 100% of the community hospitals and 92% of the teaching hospitals stated that retaining physicians has become more difficult. Due to physician supply problems, alterations in patient services were necessary in both the teaching hospitals (38%) and communi- ty hospitals (53%). Additionally, adjustments in staffing patterns were reported to be necessary by 49% of the teaching hospitals and by 40% of the community hospitals.<br><br> Medical staff presidents in community hospitals singled out the following specialties as having a particularly severe shortage at their facilities: Anesthesiology General Surgery Gastroenterology Internal Medicine OB/GYN Radiology Survey of Residents and Fellows Only 35% of resident and fellow survey respondents indicated they are planning to pursue their medical career in Massachusetts. Overall, the residents and fellows considered the following factors as unfavorable in Massachusetts: Work hours Practice environment Salary Housing costs Cost of living Tax environment Most of the physician respondents indicated that the aforemen- tioned factors influenced their decisions to not practice in Massa- chusetts, despite the excellent clinical and research opportunities 7 available here. The intensity of these feelings should be fully rec- ognized, specifically 80% to 90% of residents and fellows believe these economic factors are very real and affect their locational decisions.<br><br> Survey of Residency and Fellowship Program Directors Results from the residency and fellowship program directors in Massachusetts who responded indicate that almost half of the resi- dents (44%) and fellows (49%) left Massachusetts to pursue med- ical careers elsewhere due to the poor practice environment and high cost of living in Massachusetts. Despite the large number of opportunities in Massachusetts, the residents and fellows continue to leave Massachusetts each year due to the cunfavorable d practice environment and personal factors, such as high cost of living. Survey of Biotechnology Companies In sharp contrast to the other survey respondents, approximately 60% of the biotechnology companies stated that they are able to recruit and fill a vacant physician position in less than three months.<br><br> Primary reasons include high compensation structures, such as salary and stock options, and relief from the day-to-day pressures of the practice environment. Access to Care Survey Seventeen percent of patients surveyed rated their access to medical care as difficult or extremely difficult, which extrapolates to approx- imately 680,000 adult Massachusetts residents. 7 Eighty-nine percent of patients surveyed were either very satisfied or somewhat satisfied with the health care they had received in the past 12 months.<br><br> In this context, there exists a certain amount of celasticity d in the health care workforce. This elasticity is particularly demonstrated in the physician community through increased work hours, staffing changes, and altered services, which has delayed a complete break- down in the high-quality health care services for which Massachu- setts is renowned. Based on a comparison of the 2002 and 2003 study results and the individual physician comments, one may sur- mise that the status may be nearing the limit of its elasticity.<br><br> 8 7 Based on U.S. Census 2001 estimates of the resident population between the ages of 18 and 65 in Massachusetts. 9 Implications for Patients 9 Access to Health Care Utilizing the Physician Workforce Study, the MMS has been monitoring physi- cian supply trends and changes in the practice environment for several years.<br><br> This comprehensive report demonstrates that Massachusetts has clearly been experi- encing a crisis in the number of physicians available to deliver quality patient care for the past several years. The health care system is being further stressed by the continual exodus of new physicians who complete their training here and leave to practice elsewhere in the country. In addition, the loss of this clearn then leave d group is com- pounded by the loss of established physicians who leave to practice in another state, make a change in their careers, or retire early.<br><br> The MMS is concerned that the declining practice environment and short- ages of physicians across all specialties will threaten patients 9 access to the high- quality health care for which Massachusetts is renowned. I NTRODUCTION T he Committee on Medical Service (Committee) conducted the following primary research: a survey of medical staff presidents in community hospi- tals, a survey of residents and fellows in their last year of training, a survey of resi- dency and fellowship program directors, a survey of department chiefs from teaching hospitals, a survey of practicing physicians in community and hospital settings throughout Massachusetts, a survey of biotechnology companies, and an opinion poll of health care consumers. The Committee consulted with econo- mists James Howell, Ph.D., Carol Simon, Ph.D., and Andrew Sum, Ph.D., in the development of the survey tools and in the analysis of the results.<br><br> 10 11 B ACKGROUND I n 2002, the Massachusetts Medical Society (MMS) conducted a Physician Workforce Study 8 to evaluate the existing physician workforce and labor mar- kets within the Commonwealth of Massachusetts. Our study provided initial data suggesting that a critical shortage in the Massachusetts physician labor mar- ket existed. Based upon the data, we concluded that the Commonwealth was fac- ing a severe crisis in the number of physicians available to deliver patient care.<br><br> Since these findings were based upon data collection and analysis of only one year, we approached the results with a certain amount of caution and viewed them as the foundation of our longer-term analysis. The findings from the cur- rent study corroborate our initial results from last year and will demonstrate that the physician workforce and labor market are in a state of severe crisis. In an effort to further the discussions concerning the adequacy of the sup- ply of physicians in the Commonwealth, we continued to examine several of the influencing factors.<br><br> We found that despite the conflicting information regarding the oversupply/undersupply of the physician labor market put forth by govern- mental advisory agencies, such as the Council on Graduate Medical Education (COGME) 9 , there is an increasing amount of published literature supporting the argument that the existing methodology to determine adequacy of supply is flawed and that a shortage in the physician labor market exists throughout the country. While the evaluation of physician labor market supply by Cooper, et al., remains the most important paper 10 on this topic, additional studies are becom- ing available that scrutinize previous ideas and viewpoints concerning how to evaluate the physician labor market. Recently, additional studies, such as a paper by Sheldon in 2003, identify several flaws in the COGME workforce recommendations (e.g., using the same assumptions and ratios for the past twenty years, not defining cspecialists, d and 8 Massachusetts Medical Society, Physician Workforce Study, May 2002, available at http://www.massmed.org/pages/workforce.asp (accessed 15 April 2003).<br><br> 9 COGME, Update on the Physician Workforce, August 2000, available at http://www.cogme.gov/00_8726.pdf (accessed 14 April 2003). 10 Cooper RA, et al. Economic and Demographic Trends Signal an Impending Physician Shortage.<br><br> Health Affairs, January/February 2002. basing predictions upon a health care system expected to evolve along the organi- zational pattern of staff-model health maintenance organizations). 11 Additional supporting reports have been published, such as by Schubert, et al.<br><br> 12 , that demonstrate the national shortages of anesthesiologists. These national findings further corroborate the conclusions in our previous study, which referenced a manpower study conducted by the Massachusetts Society of Anesthesiologists. A bibliography of resources related to the physician workforce question is contained in Appendix D.<br><br> 12 11 Sheldon GF. Great Expectations: The 21st Century Health Workforce. Am J Surg 2003;185(1):35-41.<br><br> 12 Schubert A, Eckhout G, Tremper K. An Updated View of the National Anesthesia Personnel Shortfall. Anesth Analg 2003; 96(1): 207-214.<br><br> M ETHODOLOGY G iven the large scope of the project, the Committee felt that primary (i.e., surveys) and secondary data (e.g., a review of existing databases, literature) were needed to properly examine the Massachusetts physician workforce issues. The Committee conducted the following primary research: a survey of practicing physicians in community and hospital settings throughout Massachusetts, a sur- vey of residents and fellows in their final year of training, a survey of residency and fellowship program directors, a survey of teaching hospital department chiefs, survey of medical staff presidents of acute care community hospitals; a survey of biotechnology companies, and a phone-based survey of health care consumers. Survey of Practicing Physicians The largest component of this study was a survey mailed to 7,565 physicians in December 2002.<br><br> The survey was mailed to both MMS members and nonmem- bers who were randomly selected from 14 specialties (anesthesiology, cardiology, emergency medicine, family practice, gastroenterology, general surgery, internal medicine, neurosurgery, obstetrics and gynecology, orthopedics, pediatrics, psy- chiatry, radiology, and vascular surgery). Each survey was sent with a cover letter and a postage-paid return envelope. The surveys were serially numbered for a sec- ond follow-up mailing to nonresponders that occurred in January 2003.<br><br> The sample size of this survey is almost twice the sample size surveyed in this portion of the 2002 Physician Workforce Study. The survey asked physician respondents to provide information regarding physician vacancies, recruitment efforts, alteration of services, or adjustment to staffing due to physician vacancies, shortages in specific specialties, and retention. In addition, questions were asked to measure physician perceptions about the practice environment in Massachusetts.<br><br> By tracking responders and nonresponders, it was possible to aggregate the results by metropolitan statistical area (MSA), allowing for statistical analysis by region. The MSA grouping methodology was based on the Dartmouth Atlas on Health Care methodology. 13 Survey of Community Hospitals Due to the importance of community hospitals to the provision of health care services, a survey was mailed to the medical staff presidents of 62 acute care com- munity hospitals throughout Massachusetts.<br><br> This survey asked respondents to provide information regarding physician vacancies, recruitment efforts, alteration of services or adjustments to staffing due to physician vacancies, shortages in spe- cific specialties, and retention at their facility. The questions asked were written to be comparable to questions asked in the surveys of practicing physicians and teaching hospital department chiefs. Survey of Residents/Fellows and Residency/Fellowship Program Directors In focusing on the factors affecting the residents 9 and fellows 9 locational decisions, two groups were surveyed: residents and fellows in their last year of training and the residency and fellowship program directors.<br><br> The survey mailings targeted individuals in fourteen specialties at ten teaching hospitals: Specialties: anesthesiology, cardiology, emergency medicine, family practice, gastroenterology, general surgery, internal medicine, neurosurgery, obstet- rics and gynecology, orthopedics, pediatrics, psychiatry, radiology, and vas- cular surgery. Teaching Hospitals: Boston Medical Center, Massachusetts General Hos- pital, Brigham and Women 9s Hospital, Beth Israel Deaconess Medical Center, Children 9s Hospital, Baystate Medical Center, UMass Medical Center, Caritas St. Elizabeth 9s Medical Center, Lahey Clinic, and New England Medical Center.<br><br> A four-page, six-question survey was developed to ask about post-training employment decisions, whether or not those surveyed were planning to seek employment in Massachusetts, and how respondents rated Massachusetts (favor- ably/unfavorably) with respect to professional and personal factors that influence locational decisions. This survey was very similar to the resident and fellow 14 survey used in the 2002 Physician Workforce Study with the inclusion of addi- tional factors that influence locational decisions. The survey for program directors asked historical questions about program openings and applications over the past year, as well as the number of trainees who have stayed or left Massachusetts between 1997 and 2002.<br><br> Residency and fellowship program directors were also asked how Massachusetts rated (favor- ably/unfavorably) with respect to professional and personal factors. Both surveys were mailed in December 2002. Staff contacted each of the residency and fellowship programs, using the American Medical Association 9s Graduate Medical Education Directory, to explain the goal of our study and to reemphasize that only graduating residents and fellows were eligible to partici- pate.<br><br> Working with the program coordinator to determine the number of resi- dents and fellows in their last year of training, a package of surveys was sent to each program containing surveys for residents, fellows, and the program director. Each survey packet included a cover letter, survey, and postage-paid return enve- lope. In an effort to increase the survey response rate, an additional follow-up mailing was sent to each program.<br><br> Survey of Department Chiefs of Teaching Hospitals This survey asked department chiefs of anesthesiology, cardiology, emergency medicine, primary care, general surgery, gastroenterology, neurosurgery, obstetrics and gynecology, orthopedics, pediatrics, psychiatry, radiology, and vascular surgery at nine teaching hospitals questions regarding physician full-time equiva- lents (FTE) currently employed, FTE vacancies, new hires, and separations dur- ing the last six months. It also asked for the department chiefs 9 experience with the adequacy of the physician applicant pool, recruiting time to fill a physician vacancy, alteration of services and adjustments to staffing due to unfilled vacan- cies, and retention of existing staff physicians. The survey expanded upon the 2002 Physician Workforce Study by increasing the number of specialties sur- veyed from four to thirteen.<br><br> Surveys were sent with cover letters and postage- paid return envelopes. Additional follow-up mailings were also sent. Results from the surveys were kept in the aggregate to maintain the confidentiality of the respondents.<br><br> 15 Survey of Patients 4 Access to Care A telephone-based opinion poll of 400 patients was conducted by Opinion Dynamics Corporation in January 2003. The survey gathered information about the accessibility of health care services and respondents 9 satisfaction with the care provided. Survey of Biotechnology Companies In conjunction with the Massachusetts Biotechnology Council (MBC), a survey was sent in December 2002 to 135 MBC members to gather information about the employment of physicians and the use of physician consultants within their organizations.<br><br> The surveys were sent to the human resources department within each company. Surveys were sent with cover letters and postage-paid return envelopes. Additional follow-up mailings were also sent in January 2003.<br><br> Data Entry and Analysis All returned surveys were logged in, and responses were entered into an Microsoft Access database for cleaning and categorization. The databases were imported into SPSS, a statistical software package, for analysis. Sample Characteristics Please see Appendix B for a detailed discussion of sample characteristics.<br><br> Study Methodology The most significant characteristic of the MMS 2002 and 2003 Physician Work- force Studies was the extraordinarily high frequency of negative responses to all questions relating to the current availability of physicians to fill positions, as well as the degree of difficulty in recruiting and retaining physicians. These results provide strong empirical support to the conclusion that physician labor markets in Massachusetts are currently under extreme stress. In the results of the 2003 study, the mean response ranged from over 25 to 50 percent for each of the rele- vant survey questions, indicating that there are serious problems with the labor market in which they are operating.<br><br> Customarily, responses indicating shortages in labor markets rarely exceed 5 to 10 percent. The existence of these significantly 16 higher-than-expected response rates to questions concerning labor shortage raises questions about how these results should be interpreted. After considerable discussion, we believe that the best way to interpret these results is to conclude that because all fourteen specialties surveyed are currently experiencing sufficiently high labor-market vacancy rates, these labor markets are now confronted with serious problems.<br><br> This represents confirmation of the findings of the 2002 Physician Work- force Study report. We noted the very high physician-vacancy rates in the 2002 survey data but chose to approach their magnitude with a certain amount of cau- tion. With the results of the 2003 Physician Workforce Study survey, the high 2002 vacancy rates are confirmed.<br><br> Indeed, in some cases, they describe labor market conditions that have actually worsened. Thus, we reiterate our primary conclusion: Physician labor markets in Massachusetts are currently under extreme stress; the forces that have pushed these markets into this unenviable state are numerous and they are not likely to be easily reversed. Having established this primary conclusion, we fully recognize that the adverse impacts of labor market tightness have not been felt in all markets with the same degree of intensity.<br><br> Thus, we wanted a way to quantify differences in the labor markets for the fourteen physician specialties selected for detailed study. 13 The most straightforward way to address this issue is to analyze by specialty the variances from the sample means of the responses to each of the six relevant survey questions for the practicing physician survey and compare these results to a fixed level of greater than 20% and greater than 50% to identify a shortage. This analysis satisfies our needs for two reasons.<br><br> First, data limitations resulting from the availability of only two years of data necessarily meant that a time-series analysis could not be undertaken. Second, an analysis of variances and comparing 17 13 A transition comment needs to be made about the concept clabor-market shortages d and cvacancy rates. d In most studies of this nature, it is commonplace to estimate statistically the percent of job vacancies that exist across business firms and in the aggregate across an industry. For these rates to be estimated, the industry and firm structure must be easily defined; that is, vacancies are estimated as a percentage of the total firm employ- ment.<br><br> Given the splintering of the physician practice industry, and the reality that patient services must be met whatever the difficulty, it has been virtually impossible to define vacancy rates in this study. Of necessity, this means that a survey of this nature can only describe conditions in labor markets through a series of questions that tend to describe the various parameters of the market. This statistical problem must be recognized but in no way diminishes the conclusions discussed in this report.<br><br> the results to the fixed shortage levels was the preferred route toward producing a final document that is simultaneously statistically robust and relatively easy to understand. In order to differentiate the degree of intensity of labor market shortages, we established the following criteria: For a physician specialty to be considered ccritical d in terms of its labor-market tightness, responses to the six key questions must meet all the following criteria: Responses to at least two out of six questions must exceed an absolute rate of 50%. Responses to remaining questions must exceed an absolute rate of 20%.<br><br> Individual responses to all six questions must be greater than the mean for each question. For a physician specialty to be considered csevere d in terms of its labor-market tightness, the responses to the six key questions must meet all the following criteria: Responses to one out of six questions must exceed an absolute rate of 50%. Responses to at least five questions out of six must exceed an absolute rate of 20%.<br><br> Individual responses to any three out of six questions must be greater than the mean for each question. The Six Survey Questions First, we will examine the categories we identified as critical to labor market con- ditions, the steps taken to determine if shortages existed, and, if so, for which specialties. Adequacy of physician applicant pool to fill vacant positions Question 16: Is the current pool of physician applicants ade- quate to fill your vacant positions or expand your practice?<br><br> Specialties where filling existing vacancies is difficult Question 17: Are you currently experiencing difficulty in fill- ing physician vacancies? Specialties where recruitment time and the average time it takes to recruit a physician have increased 18 19 Question 22: Over the past three years, has the amount of time needed to recruit physicians changed? If YES, by how much time?<br><br> (increased/decreased by # months). Specialties where staff retention is more difficult Question 23: Over the past three years, has your ability to retain your existing staff of physicians changed? If YES, has retaining physicians in your practice become more difficult or easier?<br><br> Specialties where supply problems make it necessary to alter services or adjust professional staffing patterns Question 18: Have physician supply problems made it neces- sary for you to alter the services you provide? Question 19: Have physician supply problems made it neces- sary for you to adjust your professional staffing patterns? Each of these questions is addressed sequentially in the discussion that fol- lows.<br><br> In addition, when available, examples of verbatim answers will follow appropriate questions and tables. The next focus will be on the results of the detailed structural analysis. The specific answers for each of the specialties across the six questions are organized into three separate tables, each reflecting the extent that each of the results meet the criteria established above.<br><br> Shown in Table 2 are the five specialties that are defined as currently experiencing critical physician shortages. A review of these results reveals a fairly consistent pattern: All five specialties have rates in excess of 50% for at least two of six questions. T ABLE 2: R ESULTSFOR S PECIALTIES E XPERIENCING C RITICAL P HYSICIAN S HORTAGES Q16:Q22:Q23:Q17:Q18:Q19: 2003 INADEQUATEINCREASESIGNIFICANTSIGNIFICANTSUPPLYPROBLEMSSUPPLYPROBLEMS SUMMARYOFPOOLOFINTIMEDIFFICULTYDIFFICULTYINCAUSEALTERATIONCAUSECHANGESIN ACTUALRATESPHYSICIANSTORECRUITTORETAINFILLINGVACANCIESOFSERVICESPROFESSIONALSTAFFING Anesthesiology72%50%43%44%39%63% Cardiology60%60%24%41%42%48% GI72%65%24%53%49%46% Neurosurgery72%33%33%57%53%63% Radiology83%68%32%58%31%66% SAMPLEMEAN 49%41%23%32%31%37% In three specialties (gastroenterology, neurosurgery, and radiology), these high vacancy rates are present in more than three cases.<br><br> The responses for all five specialties for all six questions exceed the 20% threshold and are consistently greater than their respective sample means. Before proceeding with the analysis, it is important to establish what these responses are telling us about the physician labor market in Massachusetts. First, the responses in excess of 50% are to be viewed as rough proxies of the intensity of the shortages in each of the specialty labor markets.<br><br> Second, the responses that satisfy each of the second and third criteria are indicative of the pervasive nature of physician shortages in labor markets. The second group of physician shortages are those that are currently experi- encing severe labor market shortages. The responses to the six questions are shown in Table 3.<br><br> A careful review of these responses shows a somewhat less definitive pattern than in the responses in Table 2, but these results are important in defining physician specialties where severe shortages exist. While it is true that there are responses in excess of 50% for at least one of the six questions, six of the responses are below the mean rates for four of the questions. This part alone is consistent with a certain amount of labor-market slack, vis-à-vis the conditions in the critical shortage labor markets outlined above.<br><br> In only one case, vascular surgery, was the response rate below the 20% threshold, and this could be considered to be in the more normal range. 20 T ABLE 3: R ESULTSFOR S PECIALTIES E XPERIENCING S EVERE P HYSICIAN S HORTAGES Q16:Q22:Q23:Q17:Q18:Q19: 2003 INADEQUATEINCREASESIGNIFICANTSIGNIFICANTSUPPLYPROBLEMSSUPPLYPROBLEMS SUMMARYOFPOOLOFINTIMEDIFFICULTYDIFFICULTYINCAUSEALTERATIONCAUSECHANGESIN ACTUALRATESPHYSICIANSTORECRUITTORETAINFILLINGVACANCIESOFSERVICESPROFESSIONALSTAFFING General Surgery51%45%28%37%26%25% Orthopedics58%47%23%39%37%34% Vascular Surgery63%52%14%30%38%42% SAMPLEMEAN 49%41%23%32%31%37% The specific response rates for the remaining six specialties are displayed in Table 4. To reemphasize the general conclusions stated above, shortages exist across all fourteen specialties; it is the differential intensity of the shortage rates that we are discussing here.<br><br> A review of these responses tells us that while the shortages in these physi- cian labor markets are very real 4 that is, they are considerably greater than what is to be expected in terms of the normal workings of labor markets 4 they are certainly less severe than those outlined in Tables 2 and 3. Two specific inter- pretative comments for this table are as follows: Note that only six of the 36 responses to the six questions are greater than their respective survey question means. Ten responses are below the 20% threshold; two of these are in the single-digit range.<br><br> Structure of the Report The following sections will be discussed in detail in the report: Section 1: Determination of Shortage by Specialty Section 2: Evaluating the Results Among Medical Staff Presidents in Community Hospitals and the Department Chiefs of Teaching Hospitals Section 3: Survey Results Concerning the Opinions of Residents, Fellows, and Program Directors Section 4: Analysis of Questions Relating to the Issue of Profession- al Liability Expenses 21 T ABLE 4: R ESULTSFOR S PECIALTIES E XPERIENCING P HYSICIAN S HORTAGES Q16:Q22:Q23:Q17:Q18:Q19: 2003 INADEQUATEINCREASESIGNIFICANTSIGNIFICANTSUPPLYPROBLEMSSUPPLYPROBLEMS SUMMARYOFPOOLOFINTIMEDIFFICULTYDIFFICULTYINCAUSEALTERATIONCAUSECHANGESIN ACTUALRATESPHYSICIANSTORECRUITTORETAINFILLINGVACANCIESOFSERVICESPROFESSIONALSTAFFING Emergency Medicine50%38%16%18%23%43% Family Practice38%25%16%21%27%24% Internal Medicine50%39%20%29%36%34% OB/GYN29%24%21%20%19%25% Pediatrics19%14%7%9%14%12% Psychiatry39%35%24%31%36%32% SAMPLEMEAN 49%41%23%32%31%37% Section 5: Patients 9 Access to Care and Physicians 9 Attitudes Toward Their Profession Section 6: Regional Disparities Across the Principal Urban Labor Markets in Massachusetts Section 7: Conclusions 22 S ECTION 1: D ETERMINATIONOF S HORTAGEBY S PECIALTY 1.1 4 Adequacy of the Physician Labor Pool Question 16: Is the current pool of physician applicants adequate to fill your vacant positions or expand your practice? Question 17: Are you currently experiencing difficulty in filling physician vacancies? When considered together, these two questions identify those physician specialties where shortages are critical in terms of the perceptions of adequacy of the current labor pool and where day-to-day difficulties in recruiting specialists are most troublesome.<br><br> The primary conclusion that may be drawn from the data are that there were significant variances by physician specialty in the adequacy of the available labor market pool in 2002, and these variances continue to exist in 2003. The data are shown in Table 5 on page 24. A careful review of the data clearly shows that the eight specialties designat- ed as having either severe or critical shortages have responses that are significantly above the sample means with regard to supply inadequacies in the existing labor pool in both 2002 and 2003.<br><br> In addition, it is clear that the supply inadequacies of the physician labor pool are increasing, and it is becoming more difficult to recruit physicians. A review of the data shown in Table 5 also provides additional insight into the performance of physician labor markets in Massachusetts. Of particular inter- est are the survey responses to Question 17.<br><br> In addition to the data displayed in Table 5 on page 24, we have also plotted the responses to Question 17 in a chart. Reading from left to right one is able to see those specialties where the practicing physician survey respondents reported significant difficulties in filling existing vacancies. Of interest are the significant dispersions around the aggregate sample 23 24 T ABLE 5: R ESPONSESTO Q UESTIONS 16 AND 17 OFTHE P RACTICING P HYSICIANS S URVEY INADEQUATESIGNIFICANTDIFFICULTY LABORPOOLTORECRUIT 2003200220032002 RESULTSRESULTSRESULTSRESULTS Anesthesiology72%84%44%52% Cardiology60%57%41%33% Emergency Medicine50%55%18%24% Family Practice38%24%21%9% General Surgery51%29%37%22% GI72%71%53%48% Internal Medicine50%34%29%20% Neurosurgery72%58%57%47% OB/GYN29%27%20%11% Orthopedics58%41%39%28% Pediatrics19%14%9%8% Psychiatry39%26%31%18% Radiology83%84%58%62% Vascular Surgery63%44%30%14% SAMPLEMEAN 49%38%32%24% Q16: I STHECURRENTPOOLOFPHYSICIAN APPLICANTSADEQUATETOFILLYOURVACANT POSITIONSOREXPANDYOURPRACTICE ?<br><br> Q17: A REYOUCURRENTLYEXPERIENCING DIFFICULTYINFILLINGPHYSICIANVACANCIES ? 0% 10% 20% 30% 40% 50% 60% 70% Pediatrics Emergency Med. OB/GYN Family Practice Internal Med.<br><br> Vascular Surg. Psychiatry Sample Mean General Surgery Orthopedics Cardiology Anesthesiology GI Radiology Neurosurgery 9% 18% 20% 21% 29% 30% 31% 32% 37% 39% 41% 44% 53% 57% 58% Chart 1 4 Question 17: Percentage of physician respondents reporting significant difficulty in fillling physician vacancies means. The means by specialty range from 9% to 58%, with an overall mean of 32%.<br><br> In Chart 1, we have displayed the data from the 2003 survey. It is interesting to note that the specialties having critical shortages are the five specialties (anesthesiology, cardiology, gastroenterology, neurosurgery, radiol- ogy) that reported the most difficulty in filling physician vacancies. These two questions were also asked of the department chiefs in nine teaching hospitals for thirteen specialties.<br><br> Table 6 provides the responses to ques- tions 16 and 17 from the chiefs of selected specialty departments at the nine teaching hospitals. These responses provide additional support for the results in Chart 1. In addition to the specialties already identified as having critical shortages, three other specialties have a value significantly greater than the mean: OB/GYN, pediatrics, and primary care.<br><br> While the shortages in these specialties are not as significant as in the practicing physician survey, these results may point to a future concern. Two relevant responses to Question 17 are included below. cIdealism and zeal are being diminished by a flawed system. d cDifficult to attract MDs to Massachusetts due to the poor and unrealistic reimbursement that increasingly leads to patients being denied access to care. d 25 T ABLE 6: S ELECTED R ESPONSESTO Q UESTIONS 16 AND 17 OFTHE D EPARTMENT C HIEFSOF T EACHING H OSPITALS S URVEY 2003 SURVEYRESULTS INADEQUATEDIFFICULTTO LABORPOOLRECRUIT Anesthesiology50%75% Cardiology100%100% GI83%67% Neurosurgery100%100% OB/GYN67%57% Pediatrics75%75% Primary Care75%75% Radiology67%78% SAMPLEMEAN 55%62% Q16: I STHECURRENTPOOLOFPHYSICIANAPPLICANTSADEQUATE TOFILLYOURVACANTPOSITIONSOREXPANDYOURPRACTICE ?<br><br> Q17: A REYOUCURRENTLYEXPERIENCINGDIFFICULTYINFILLINGPHYSICIANVACANCIES ? 1.2 4 Alterations to Services and Adjustments to Staffing Question 18: Have physician supply problems made it necessary for you to alter the services you provide? Question 19: Have physician supply problems made it necessary for you to adjust your professional staffing patterns?<br><br> Questions 18 and 19 were designed to determine the impact of physician short- ages on the provision of services and/or the need to adjust professional staffing patterns to provide patient care. The responses to these two questions are shown in Table 7. Before commenting on the disaggregated details across the fourteen special- ties, it is important to recognize that 31% of the physicians surveyed are finding it necessary to alter the services they provide and 26 T ABLE 7: R ESPONSESTO Q UESTIONS 18 AND 19 OFTHE P RACTICING P HYSICIAN S URVEY YESYES , ADJUSTED ALTEREDSERVICESSTAFFINGPATTERNS 2003200220032002 RESULTSRESULTSRESULTSRESULTS Anesthesiology39%53%63%72% Cardiology42%29%48%40% Emergency Medicine23%30%43%50% Family Practice27%17%24%17% General Surgery26%23%25%35% GI49%56%46%44% Internal Medicine36%28%34%25% Neurosurgery53%38%63%36% OB/GYN19%22%25%28% Orthopedics37%34%34%31% Pediatrics14%16%12%14% Psychiatry36%35%32%31% Radiology31%40%66%76% Vascular Surgery38%0%42%25% SAMPLEMEAN 31%30%37%32% Q18: H AVEPHYSICIANSUPPLYPROBLEMSMADEITNECESSARY FORYOUTOALTERTHESERVICESYOUPROVIDE ?<br><br> Q19: H AVEPHYSICIANSUPPLYPROBLEMSMADEITNECESSARY FORYOUTOADJUSTYOURPROFESSIONALSTAFFINGPATTERNS ? 37% have been forced to change professional staffing patterns because of the current physician shortage in Massachusetts. From a managerial and organizational standpoint, these high ratios speak volumes about the disruptive context in which physicians maintain their prac- tices.<br><br> In addition, the 2003 results are an increase over the results from 2002. Now, we 9ll turn our attention to the details. Based on the statistical criteria discussed at the beginning of this section, it should be noted that among seven of the eight specialties identified as experienc- ing a critical or severe shortage 4 anesthesiology, cardiology, gastroenterology, neurosurgery, radiology, orthopedics, and vascular surgery 4 physician shortages have resulted in adjustments in professional staffing patterns to satisfy patient demands and/or altered physician services.<br><br> But, in the final analysis, all of the ratios shown in Table 7 are sufficiently large enough to raise the question as to how this trend will affect the future abili- ty of health care organizations to continue providing the quality medical care that Massachusetts physicians want to deliver. Alternatively, physicians are con- tinually challenged with managing staffing and providing services in order to meet growing patient demands. Relevant verbatim responses to Questions 18 and 19 from the practicing physician survey are below.<br><br> cIt is five-week wait to see me (I don 9t think it should be like this). d cWe are having to say 8no 9 to even newborn patients at times. d cNow, we are in a cross-coverage 8on call 9 schedule with another group. This means that an individual may be delivered by an OB she has never met. d When the same questions were asked of the department chiefs of teaching hospitals, they responded that 38% had altered services and 49% had adjusted professional staffing patterns. These overall rates for the teaching hospital department chiefs are above the rates for the practicing physicians, and they outline the serious situation in hospi- tals.<br><br> In Table 8, the responses to Questions 18 and 19 for selected specialties of 27 the hospital chiefs survey are displayed, which reinforce the result of the practic- ing physician survey. Several verbatim responses to Questions 18 and 19 from physicians are included below. cWe have not been able to increase our coverage to keep pace with the increasing volume and acuity.<br><br> As a result, patients wait 50 per- cent longer than they did two to three years ago, and we spend less time with each patient. d cPotentially dangerous conditions exist, such as obstetricians having to cover office when patients are in labor, etc. Working harder and always in a rush to get everything done. d cTwo physicians now cover the same number of patients and night call once covered by six physicians. d 28 T ABLE 8: S ELECTED R ESPONSESTO Q UESTIONS 18 AND 19 OFTHE D EPARTMENT C HIEFSOF T EACHING H OSPITALS S URVEY HOSPITALDEPARTMENTCHIEFSSURVEY ADJUSTED ALTEREDSERVICESPROFESSIONAL STAFFINGPATTERNS Anesthesiology50%75% Cardiology33%100% Neurosurgery67%100% OB/GYN43%29% Orthopedics40%60% Pediatrics75%75% Primary Care75%50% Radiology56%78% SAMPLEMEAN 38%49% Q16: H AVEPHYSICIANSUPPLYPROBLEMSMADEITNECESSARYFORYOU TOALTERTHESERVICESYOUPROVIDE ? Q17: H AVEPHYSICIANSUPPLYPROBLEMSMADEITNECESSARYFORYOU TOADJUSTYOURPROFESSIONALSTAFFINGPATTERNS ?<br><br> 1.3 4 Time Required to Recruit Physicians Question 21: Based on your current experience, how long does it take to recruit a physician to your practice? Question 22: Over the past three years, has the time needed to recruit physicians changed? If yes, by how much has the time need- ed to recruit increased or decreased?<br><br> These two questions were designed to collect opinions and attitudes on the changing dynamics of physician recruitment. Question 21 collects information on the specific amount of time, measured in months, required to recruit a physi- cian. Question 22 asks whether the amount of time to recruit has changed and by how much time the change either increased or decreased.<br><br> A review of all responses to this question showed that an overwhelming number indicated the amount of time had increased. Accordingly, the data shown in Table 9 (see page 30) includes only those responses that expressed increases in recruitment time. The survey results must be judged together, but before we examine the details, it will be important to note what is perhaps the most important overall conclusion supported by these survey data: The average amount of time required to recruit a physician amounts to over one year, based on the results of the 2003 practic- ing physician survey.<br><br> Among the 2003 survey respondents, there were only two specialties (emergency medicine and pediatrics) where the amount of time was significantly less than one year. At the other extreme, there were three specialties (gastroen- terology, neurosurgery, and orthopedics) where the recruit- ment time was eighteen months or longer, all of which are designated as having a severe or critical physician labor shortage. These ratios tell us a great deal about the difficulty inherent in physician labor markets in Massachusetts.<br><br> The loss of a single physician from a small prac- tice 4 particularly when the individual is in a critically short-supply specialty and in a practice with only one or two other physicians 4 means that the prac- tice will not likely be able to provide the same volume of professional services for 29 a twelve to eighteen month period. Such events can cause a significant disconti- nuity in patient care. Data from a related MMS/Massachusetts Biotechnology Council survey provide results that are in sharp contrast to the data from the MMS Physician Workforce Study.<br><br> For example, when biotech CEOs were asked: cHow long does it take to recruit a physician? d roughly 60 percent indicated that it would take three months or less. There are two obvious reasons for this. First, the higher salary levels and relief from the day-to-day pressures on the practicing physician make recruitment easier.<br><br> Second, the widespread availability of stock options has also served as an attractive incentive. We may now turn our attention to the specific responses to Questions 21 and 22. The survey data are summarized below in Table 9.<br><br> 30 T ABLE 9: R ESPONSESTO Q UESTIONS 21 AND 22 OFTHE P RACTICING P HYSICIANS S URVEY AMOUNTOFTIMEREQUIREDTOMEANNUMBEROFMONTHS RECRUITHASINCREASEDSIGNIFICANTLYREQUIREDTORECRUIT 20032002 20032002 ( INMONTHS )( INMONTHS ) Anesthesiology50%63%9.6%10.6% Cardiology60%41%14.4%12.7% Emergency Medicine38%23%8.4%8.6% Family Practice25%19%10.1%11.7% General Surgery45%33%13.7%12.8% GI65%69%19.1%22.3% Internal Medicine39%30%12.5%10.3% Neurosurgery33%44%26.5%22.9% OB/GYN24%20%11.7%12.8% Orthopedics47%42%17.4%14.6% Pediatrics14%18%8.8%7.8% Psychiatry35%34%11.9%11.1% Radiology68%65%13.5%14.9% Vascular Surgery52%13%11.8%17.0% SAMPLEMEAN 41%35%12.4%11.9% Q21: B ASEDONYOURCURRENTEXPERIENCE , HOWLONG DOESITTAKETORECRUITAPHYSICIANTOYOURPRACTICE ? Q22: O VERTHEPASTTHREEYEARS , HASTHEAMOUNTOF TIMENEEDTORECRUITPHYSICIANSCHANGED ? ( INCREASES ONLY 14 ) 14 A review of the 2003 survey responses indicated only sixteen responses noted a decrease in the time to recruit.<br><br> Due to the overwhelming response that time to recruit had increased, only those response were included. In 2002, there were no responses indicating a decrease. A careful review of the data supports a number of important generaliza- tions about changes in labor markets for physicians and the amount of time required to recruit the physicians: First, 41% of the physician respondents indicated that the time needed to recruit a physician has increased significantly.<br><br> This is a considerable increase from the 2002 result of 35%. Second, there are two interesting conclusions that may be drawn from the juxtaposition of the 2002 and 2003 responses to Ques- tions 21 and 22. Among the four specialties listed below, there was a consistency between increases in the amount of time required to recruit a physician and an increase in the average number of months required to successfully secure the replacement physician: Cardiology General Surgery Internal Medicine Orthopedics At the same time, it should be noted that among the fourteen spe- cialties there were also several contradictions across the paired observations.<br><br> Respondents in the following specialties indicated that the amount of time required to recruit a physician had increased but reported a decrease in the number of months required to accomplish this goal: Family Practice Obstetrics/Gynecology Radiology Vascular Surgery An additional but opposite contradiction was seen in the neurosurgery responses. The average number of months required to recruit increased, but the percentage of respondents who indicated that the time required to recruit had increased was significantly less than in 2002. The survey of department chiefs of teaching hospitals also included Question 22 regarding a change in recruiting time and if the change was an increase or decrease.<br><br> In response to this question: 66% of hospital department chiefs indicated that the amount of time needed to recruit physicians has changed. Of those who said recruiting time changed, 94% indicated that recruiting time increased. 31 The specialties with considerably higher responses that indicated an increased recruiting time were as follows: Anesthesiology (75%) Cardiology (100%) Gastroenterology (83%) OB/GYN (71%) Primary Care (75%) Radiology (89%) With the exception of OB/GYN and primary care, all of the spe- cialties above have been determined to be a critical shortage special- ty.<br><br> The results for OB/GYN and primary care illustrate that they may be approaching a crisis situation. 1.4 4 Retention of Physicians Question 23: Over the past three years, has your ability to retain your existing staff of physicians changed? If yes, has retaining physi- cians in your practice become easier or more difficult?<br><br> These two questions were designed to gain insight into the important issue of physician retention and the efficient functioning of a practice. The responses to the first part of the question 4 namely, how has the retention of physicians in your practice changed over the past three years 4 was designed to collect infor- mation about the most recent local labor market adjustments to physician demand. The results for the fourteen specialties collected on those two questions are shown in Table 10.<br><br> These questions are important to the Physician Workforce Study for at least two reasons. First, answers to the question concerning experiences over the past three years tell us a great deal about the physicians 9 perceptions about the stability or instability of their labor markets. Second, responses indicate whether conditions are making it easier or more difficult to retain physicians.<br><br> For nine of the fourteen specialties surveyed, approximately two-thirds to three-fourths of the respondents indicated that their ability to retain physicians had changed over the past three years. For five specialties, these ratios were some- what less. The sample responses were equally scattered around the means of 57 32 percent.<br><br> Thus, for the majority of physicians surveyed, labor markets, in terms of retention, have changed over the past year. More importantly, the responses to Question 23 make it very clear that these changes are making it more, not less, difficult to retain physicians. When compared with other labor market reports, it is extremely rare to see responses that were so overwhelmingly consistent; it indicates that the cumulative impact of changes in physician labor markets over the past three years has made it more difficult to retain physicians in all specialties.<br><br> This conclusion alone captures the very essence of a hard reality that physi- cian labor markets in Massachusetts are in a crisis stage. The verbatim responses as to why the ability to retain existing staff has changed have been grouped into the following categories along with the per- centages of responses each category was of the total verbatim responses to Question 23: Low salary compared to other states and/or for hours worked (40%) 33 T ABLE 10: R ESPONSESTO Q UESTION 23 OFTHE P RACTICING P HYSICIANS S URVEY RETENTIONINRETENTIONOFPHYSICIANS LABORMARKETHASBECOME HASCHANGEDMOREDIFFICULT 2003200220032002 Anesthesiology75%77%96%100% Cardiology58%61%100%100% Emergency Medicine55%68%84%96% Family Practice47%50%93%96% General Surgery61%58%100%100% GI50%44%100%92% Internal Medicine63%62%94%96% Neurosurgery67%69%100%100% OB/GYN53%47%99%86% Orthopedics56%50%97%96% Pediatrics32%35%92%88% Psychiatry55%55%100%100% Radiology71%89%100%97% Vascular Surgery43%14%100%100% SAMPLEMEAN 57%57%97%96% Q23: O VERTHEPASTTHREEYEARS , HASYOURABILITYTO RETAINYOUREXISTINGSTAFFOFPHYSICIANSCHANGED ? I FYES , HASRETAININGPHYSICIANSINYOUR PRACTICEBECOMEEASIERORMOREDIFFICULT ?<br><br> Increased workload (12%) Increased hours (9%) Cost of living in Massachusetts too high (8%) Practice environment in Massachusetts too bad (7%) Miscellaneous (24%) Specific verbatim responses for Question 23 are included below. cThey cannot make enough money to sustain reasonable living especially for candidates who are fresh out of training, have loans to pay, and have had poor salaries throughout their residency (which can last five to seven years). d cNoncompetitive environment for physicians in Massachusetts. Why beat your head against a wall when you can work for 50 per- cent more in another state and have more help with call? d cBecause of difficulty recruiting, we are working more hours per week and seeing more patients per hour than the national average. d cI think many people are positioning themselves to leave.<br><br> The work load is too heavy, especially off hours. d cMobility 4 move to other states with 8friendlier 9 environment toward physicians. d cCompetitive income from other practices/career changes (e.g., pharmaceutical careers). d cNoncompetitive salaries, increasing work volume, aging staff. d When the department chiefs of teaching hospitals were asked these same questions, 55% responded that retention rates in the labor market have changed, and an overwhelming majority of 92% responded that retention of physicians has become more difficult. These results correspond to the responses of the prac- ticing physicians. Table 11 displays the results of the hospital department chiefs survey for selected specialties.<br><br> 34 1.5 4 Determination of Shortage by Specialty Conclusion In conclusion, having identified the eight physician specialties that are currently most stressed in Massachusetts labor markets and discussing specific question results, we will elaborate on the analysis by examining additional survey details in terms of the comparing the results of the practicing physician survey to the results from the community hospital medical staff presidents and teaching hospital department chiefs surveys, the impact of professional liability costs, physicians 9 attitudes toward their practices, and finally regional disparities in shortages. 35 T ABLE 11: S ELECTED R ESPONSESTO Q UESTION 23 OFTHE D EPARTMENT C HIEFSOF T EACHING H OSPITALS S URVEY HOSPITALDEPARTMENTCHIEFSSURVEY RETENTIONINLABORMARKETRETENTIONOFPHYSICIANSHAS HASCHANGEDBECOMEMOREDIFFICULT Anesthesiology75%67% Cardiology67%100% GI100%83% Neurosurgery67%100% Pediatrics100%100% Radiology67%100% SAMPLEMEAN 55%92% Q23: O VERTHEPASTTHREEYEARS , HASYOURABILITYTORETAIN YOUREXISTINGSTAFFOFPHYSICIANSCHANGED ? I FYES , HASRETAININGPHYSICIANSINYOURPRACTICEBECOME MOREDIFFICULTOREASIER ?<br><br> S ECTION 2: E VALUATINGTHE R ESULTS A MONG M EDICAL S TAFF P RESIDENTSIN C OMMUNITY H OSPITALSANDTHE D EPARTMENT C HIEFSOF T EACHING H OSPITALS 2.1 4 Introduction T here are two additional surveys that are integral components of the larger MMS 2003 Physician Workforce Study. One was a carryover from the 2002 survey and included a series of questions sent directly to teaching hospital department chiefs. The second was a new component of the 2003 survey; it was a separate mail-out, mail-back survey to the medical staff presidents in community hospitals.<br><br> Before we analyze the survey data, a brief comment about the functions and institutional structures of community hospitals and teaching hospitals. Community Hospitals Community hospitals are not only the front line for primary and emergency care services, they also provide all but the most sophisticated services, such as trans- plant operations that are found only at teaching hospitals. The pervasive nature of the system of community hospitals means that all areas of the Commonwealth benefit from their economic impact.<br><br> The survey was sent to the medical staff presidents of the sixty-two acute care community hospitals, which are scattered throughout the Commonwealth among centers of populations and economic activity. Teaching Hospitals Massachusetts teaching hospitals or academic health centers are also critically important to our economy. An academic health center (AHC) is an institution 36 organized around patient care, sophisticated research, and teaching medical stu- dents.<br><br> At the heart of each AHC are the area 9s medical schools. The Boston 3Worcester Corridor includes four medical schools: Boston University, Harvard University, Tufts University, and the University of Massachusetts. The Baystate Medical Center is also a major AHC with a significant teaching role in the western region.<br><br> 2.2 4 Comparing the Results from Surveys of Community Hospitals, Teaching Hospital Departments, and Practicing Physicians In the 2003 MMS Physician Workforce Study, surveys were mailed to sixty-two community hospitals and thirteen departments at nine teaching hospitals. The key characteristics of these two samples are shown in Table 12. There was one change in the distribution of this year 9s survey to teaching hospital department chiefs.<br><br> The survey was sent to only thirteen of the fourteen physician specialties contained in the broader practicing physicians survey dis- cussed above; the specialties of internal medicine and family practice were com- bined into a survey of primary care. This represents a much more intensive survey from last year, where informatio

less