SAEM Emergency Medicine Faculty Salary and Benefits Survey

Transcription

1 ACAD EMERG MED December 2002, Vol. 9, No SAEM Emergency Medicine Faculty Salary and Benefits Survey Steven L. Kristal, MD, Karen A. Randall-Kristal, DO, Bruce M. Thompson, MD Abstract Objective: The Society for Academic Emergency Medicine (SAEM) commissioned an emergency medicine (EM) faculty salary and benefits survey for all 2001 residency review committee (RRC)-EM-accredited programs using the SAEM fifth-generation survey instrument. Responses were collected by SAEM and blinded from the investigators. Data represent compensation paid for the academic year. Seventy-six of 124 (61%) accredited programs responded, yielding usable data on 1,355 full-time faculty representing all four Association of American Medical Colleges (AAMC) regions. Methods: Blinded program and individual faculty data were entered into a customized version of Filemaker Pro, a relational database program with a built-in statistical package. Salary data were sorted by criteria such as program region, faculty title, American Board of Emergency Medicine (ABEM) certification, academic rank, years postresidency, program size, and whether data were reported to the AAMC. Demographic data were analyzed with regard to numerous criteria, including department staffing levels, emergency department (ED) volumes, ED length of stay, department income sources, salary incentive components, research funding, and specific type and value of fringe benefits offered. Data were compared with those from previous SAEM studies. Results: Mean salaries were reported as follows: all faculty, $180,913; first-year faculty, $147,746; programs reporting data to the AAMC, $174,354; programs not reporting data to the AAMC, $191,397. Mean salaries as reported by AAMC region: northeast, $178,593; south, $176,314; midwest, $200,095; west, $166,779. Full-time emergency medicine residency program faculty work an average of 1,129 clinical hours per year. Conclusions: Reported salaries for full-time EM residency faculty have risen approximately 8.7% since the last survey. Up to approximately 1,200 clinical hours worked per year, salary varies inversely with clinical hours worked. Total per-faculty patient contact time (overall workload) has grown approximately 13% since the last survey. Patient wait times have increased approximately 27% since the last survey. Significant regional differences in salaries have been present in all five SAEM surveys. Emergency medicine residency faculty continue to work at the upper extremes of patient encounters per physician, patient acuity levels, and department lengths of stay. Key words: Society for Academic Emergency Medicine; salary; benefits; income; emergency physicians; faculty. ACADEMIC EMER- GENCY MEDICINE 2002; 9: The Society for Academic Emergency Medicine (SAEM) has periodically conducted surveys of emergency medicine (EM) faculty salaries. The results of prior studies have been reported at national meetings 1 8 and in this journal. 9,10 SAEM commissioned the principal investigators of its four previous studies to conduct another faculty salary survey for the academic year. The results of this study and a comparison with those of prior surveys are reported here. METHODS Study Design. For this prospective study the investigators prepared a revised salary survey form From the SAEM Faculty Salary Task Force (SLK); William Beaumont Hospital, Royal Oak, MI (SLK); and Henry Ford Hospital, Detroit, MI (KAR-K, BMT). Received August 12, 2002; accepted August 19, Approved by the SAEM Board of Directors August 13, Address for correspondence: SAEM, 901 North Washington Avenue, Lansing, MI 48906; fax ; saem@saem. org. and presented it to SAEM for approval. The survey instrument included entry and exclusion criteria, and salary component descriptions, equivalent to prior studies. Study Population. The SAEM staff distributed the survey instrument and instructions to the chairs of each of the 124 EM-approved residency programs. Chairs were asked to provide information for each full-time faculty member, including faculty on sabbatical leave. Exclusion criteria included faculty for whom less than 50% of their compensation derived from EM, vacant positions, and housestaff and fellows in all ranks. Measurements. Salary component information was requested as follows: Base salary Compensation that is fixed, exclusive of fringe benefits, and is not influenced by practice earnings. Includes any guaranteed income provided by the institution, the department, or the medical practice plan. Expected/ non-guaranteed component Compensation that is

2 1436 Kristal et al SAEM SALARY SURVEY expected, but not guaranteed. The portion of staff income derived from per-hour work is included here. Bonus/incentive earnings This component is often highly variable and is usually estimated. It could be derived from any or all of the following: year-end bonus from a faculty practice plan, incentive earnings according to the practice plan, and/ or outside earnings where limited or controlled by the institution. The total expected cash payment is the sum of the base salary estimated nonguaranteed component bonus and incentive components. For the purposes of comparisons, total expected cash payment figures were used. Estimated cash value of fringe benefits A total dollar figure for any items provided in the section on benefits (i.e., Table 7). Chairs were asked to list an actual dollar figure, e.g., not 35% of base salary. Faculty time utilization definitions were as follows: Clinical hours are those worked in the clinical arena seeing patients and supervising residents seeing patients. Non-clinical hours include time spent on such things as administrative duties, research, lecture preparation, etc. Clinical hours per year indicates the total number of clinical hours per year a department chair expects each faculty to work after vacation, educational time off, etc., are factored out. Study Protocol. Completed survey forms were returned to the SAEM office. The SAEM staff contacted non-responding programs to encourage participation in the study. Survey forms were coded by SAEM staff to blind investigators to program and faculty identity. Data Analysis. Data were entered into a customized version of Filemaker Pro (Filemaker, Inc., Santa Clara, CA), a relational database program that permits searches and calculations of characteristic demographic and financial information. One hundred thirty-seven separate sort criteria were used to analyze the financial portion of the database with regard to 12 discrete variables. The demographic aspects of the database were similarly sorted and analyzed with regard to numerous criteria. Statistical analysis is given by the database statistical package, presented as mean standard deviation (SD). RESULTS Figure 1. The Association of American Medical Colleges (AAMC) region map. Of 124 residency review committee RRC-EM-approved programs, 76 (61%) returned the survey instrument. No surveys were unusable. A total of 1,361 faculty were reported. Six non-physician PhDs were excluded from the final database, resulting in 1,355 faculty from all Association of American Medical Colleges (AAMC) regions (Fig. 1). The database, sorted by 137 criteria for 12 key variables, is presented in comparison with previous SAEM studies. Table 1 compares the numbers and percentages of EM-approved programs participating in the five SAEM surveys. Table 2 compares EM residency faculty salaries and dollar estimates of benefits for the five study periods for variables of first-year faculty, AAMC coreporting, and AAMC regions. Table 3 reports demographic information for the responding programs for the last two study periods. Table 4 reports data collected for the last two study periods regarding shift length, hospital and emergency department (ED) size, and ED areas of specialization. Table 5 presents average ED staffing levels for the responding programs. Note that the numbers presented represent only those programs that reported >1 hour per week in that category. Table 6 reports work-hour data as well as demographic information for the faculty included in the two most recent surveys. Table 7 presents the number and percentages of TABLE 1. Percentage of Programs Reporting RRC = residency review committee.

3 ACAD EMERG MED December 2002, Vol. 9, No TABLE 2. Multiyear Comparisons AAMC = Association of American Medical Colleges. responding programs that offer fringe benefits for each of 22 categories included in the survey instrument. The corresponding data for the last survey are also listed. Table 8 presents data regarding non-clinical sources of department funding, including research and graduate medical education (GME) funding. This is the first time these data have been requested, so no comparative information is available. Table 9 reports the number of programs using each of various incentive components to determine salaries paid. The corresponding data for the last survey are also listed. Table 10 reports detailed salary and work-hour data for the academic year for faculty sorted by demographic factors. Table 11 presents salary and work-hour data for the academic year for faculty sorted by job type. DISCUSSION We have previously reported that salaries reported to the AAMC are significantly lower than those in programs not reporting. 9,10 The same result was obtained in this year s survey (Table 2). The consistency of this finding reiterates that although the

4 1438 Kristal et al SAEM SALARY SURVEY TABLE 3. Data by Programs AAMC = Association of American Medical Colleges. AAMC survey may accurately reflect salaries paid to medical school EM faculty, it underreports salaries paid to EM residency program faculty. This was the second time the survey instrument included an I don t know option for the question asking whether the responding program s data were also sent to AAMC. Once again, a surprisingly large percentage of program directors (37%) do not know whether their data are also sent to the AAMC. Reported salaries as a whole (Table 2) rose approximately 8.7% between the last two reporting periods, while first-year salaries rose approximately 5.8%. This compares with the 5.9% and 7.3%, respectively, rates of change reported in the 1998 survey. 9 Total work hours (Table 6) showed similar results for the last two reporting periods, with staff now working an average of 42.3 hours per week (note that these are hours as reported by department chairs, and that the total hours per week may differ slightly from the sum of the clinical and non-clinical means due to incomplete data on some of the survey forms). Of note this year are the data gathered for mean clinical hours per year (Table 6). The request for information regarding weekly clinical and non-clinical hours has been included in all five SAEM surveys, and mirrors information requested on the AAMC survey. Because of varying institutional practices regarding vacation and educational time off, it is difficult to extrapolate weekly work-hour figures into reliable data regarding clinical hours worked per year by each faculty (data increasingly requested of the investigators by department chairs). For that reason, this year s survey included a request for total clinical hours per year expected to be worked by each faculty member. The answer to the question what is the average number of clinical hours worked per year by emergency medicine residency program faculty? (Table 6) is 1,129 hours. Specific work-hour averages for each job type are included in Tables 11 and 12. Figure 2 demonstrates the extent to which salaries reflect clinical hours worked. The graph of mean yearly salary vs. yearly clinical work hours suggests that up until approximately 1,200 clinical hours per year, salary varies inversely with clinical hours worked. As the faculty counts on the same graph indicate, this is not a phenomenon restricted to a few highly paid individuals. Reports from department chairs indicate increasing pressure by hospital administrators to fund residency programs exclusively from clinically-generated revenue. Figure 2 suggests the extent to which programs fund a significant number of non-clinical activities. In a similar vein, Table 8 reports data regarding common, non-clinical, sources of department revenue. Of note is the small number of programs (27 of 76, 35%) reporting figures for GME funding. This figure represents revenue to departments from government sources by virtue of the hospital s having a residency program in EM. That so few depart- Figure 2. Salary vs. yearly clinical work hours.

5 ACAD EMERG MED December 2002, Vol. 9, No TABLE 4. Demographic Information Summaries ICU = intensive care unit; ED = emergency department; EM = emergency medicine; Drs. = physicians. TABLE 5. Average Staffing ED = emergency department; EM = emergency medicine. ment chairs are able to report a figure in this category suggests the lack of clarity that exists regarding the flow of these funds to departments. The 27 programs that did report data in this category represented 449 faculty for an average of 16.6 faculty per program, yielding approximately $25,011 in GME funding per faculty. Looked at another way, these 27 programs reported an average of 29.9 residents per program, yielding approximately $13,886 in GME funding per resident. With regard to patient volumes, data calculated from the results shown in Tables 4 and 5 suggest that EM residency faculty are seeing considerably more new patients per hour (3.34) than the com-

6 1440 Kristal et al SAEM SALARY SURVEY TABLE 6. Faculty Attributes the 12.9% national average. 12 Of note is that the admission percentage for academic emergency programs held steady between the last two surveys, while the national average rose approximately 1.9% during the same period. 12,13 Other selected findings of the most recent salary survey include the following: Female faculty at RRC-EM-accredited programs earn, in aggregate, approximately 14.4% less than their male counterparts (Table 10, Group 2). Females work approximately 2.8% fewer to- TABLE 7. Fringe Benefits EM = emergency medicine; ABEM = American Board of Emergency Medicine; EMS = emergency medical services. TABLE 8. Funding Amounts munity average of Patient volume has grown approximately 3% since the last survey. Patient wait times have increased approximately 27% since the last survey. Total department patient contact time (overall workload) has grown approximately 31% since the last survey. Academic ED faculty clinical coverage has increased approximately 16% since the last survey. Total per-faculty patient contact time (overall workload) has grown approximately 13% since the last survey. Acuity levels in the centers surveyed continue to be significantly higher than community norms, with a 22% average admission rate (Table 4) versus GME = graduate medical education.

7 ACAD EMERG MED December 2002, Vol. 9, No TABLE 9. Incentive Components tal hours and approximately 6.6% more clinical hours per year than males. The gap between DOs and MDs has narrowed to approximately 3% compared with the 8% edge DOs held in the last survey. Salaries show an average approximately 1.8% increase per year postresidency for the first 15 years (Table 10, Group 4). ABEM-certified faculty earn approximately 17% more than non-abem-certified faculty (Table 10, Group 5). This is comparable to the 15% difference seen in the 1998 survey. Non-EM-trained faculty again show a slight advantage (3%) in average salary over their EMtrained colleagues (Table 10, Group 6). This advantage has been seen in all but the 1995 survey. Fellowship-trained faculty continue to earn less than non-fellowship-trained faculty (Table 10, Group 7). Average numbers of years practicing and total hours worked were again similar for the two groups. We suggested in the last two surveys that the difference was possibly accounted for by the fewer clinical hours worked by fellowship-trained faculty, again seen in this year s study. 108 of the 1,355 reported faculty were in tenured positions (Table 10, Group 8), roughly the same percentage as reported in the last survey. As expected, tenured faculty earned more than their non-tenured counterparts. Similarly expected, salaries increase as one moves up the academic ladder (Table 10, Group 9) from instructor through full professor, with the biggest increase occurring as one moves from assistant to associate professor. City population and ED volume again do not appear to correlate directly with salaries (Table 10, Groups 10 and 11). Regional differences (Table 11, Group 15) continue to be significant. In all five salary surveys, the midwest has had the highest salaries, with the west always being the lowest (Table 2). Pediatrics and low-acuity staff continue to represent the bottom of the salary scale within the survey (Table 11, Groups 30 and 31). A new job-type category, observation unit director, was added this year. Only two faculty, from different programs, were reported in this category, and neither was reported to be working any clinical hours. LIMITATIONS Overall, we believe that the current SAEM faculty salary survey represents a significant improvement over the data available from the AAMC or physician-recruiter organization surveys. The SAEM study applies strictly to EM residency program faculty and surveys important variables not deemed relevant to other surveys. The chief limitation of our study is that it is not a complete representation of every RRC-approved program in EM. We are extremely grateful to the program directors who participated in the study and encourage all program directors to participate in the next SAEM salary survey. Once again, however, the final issue we would like to discuss is the nature of the data we collected. Most of us working in academic positions regard fringe benefits as a significant portion of our income. But fringe benefit data are the most difficult component to quantify. Although a program usually offers a standard package to its faculty, the value of that package will differ considerably from faculty member to faculty member, and from one year to the next by a single faculty member, depending on which components are actually utilized. Therefore, although the total expected cash payment is the most comparable figure to use when comparing salaries between faculty, it excludes a probably significant portion of true income. Faculty are advised to keep this in mind when attempting to draw conclusions from the data presented here.

8 1442 Kristal et al SAEM SALARY SURVEY TABLE 10. Sort by Demographic Information AAMC = Association of American Medical Colleges; ABEM = American Board of Emergency Medicine; EM = emergency medicine.

9 ACAD EMERG MED December 2002, Vol. 9, No TABLE 11. (below and next page). Sort by Job Category AAMC = Association of American Medical Colleges; EMS = emergency medical services.

10 1444 Kristal et al SAEM SALARY SURVEY TABLE 11. (cont.). Sort by Job Category CONCLUSIONS The SAEM emergency medicine faculty salary survey represents the most comprehensive and focused salary survey for EM residency faculty available. The demographic data suggest that residency faculty are the emergency physicians working at the upper extremes of patient encounters per physician, patient acuity levels, and department lengths of stay. The same data demonstrate that, in spite of declining government compensation for our services, our individual workloads continue to increase. References 1. Thompson B, Kristal S. Emergency medicine faculty salaries: a study of data submitted to SAEM ( ). Society for Academic Emergency Medicine annual meeting, Toronto, Ontario, Canada, Kristal S, Thompson B SAEM emergency medicine faculty salary survey. Society for Academic Emergency Medicine annual meeting, Washington, DC, Kristal S, Thompson B, Marx J SAEM emergency medicine faculty salary survey. Society for Academic Emergency Medicine annual meeting, Denver, CO, Kristal S, Randall-Kristal KA, Thompson B, Marx JA SAEM emergency medicine faculty salary survey. Society for Academic Emergency Medicine annual meeting, Boston, MA, Kristal S, Randall-Kristal KA, Thompson B, Marx JA. Academic ED staffing levels and workloads: results from the SAEM faculty salary survey. Society for Academic Emergency Medicine annual meeting, Boston, MA, Kristal S, Randall-Kristal KA, Thompson B, Marx JA. Academic emergency department areas of specialization results from the SAEM faculty salary survey. American College of Emergency Physicians annual meeting, Las Vegas, NV, Kristal S, Randall-Kristal KA, Thompson B, Marx JA. Academic emergency department funding sources and incentives results from the SAEM faculty salary survey. American College of Emergency Physicians annual meeting, Las Vegas, NV, Kristal S, Randall-Kristal KA, Thompson B, Marx JA. Female faculty salary and work hours results from the SAEM faculty salary survey. American College of Emergency Physicians annual meeting, Las Vegas, NV, Kristal SL, Randall-Kristal KA, Thompson BM, Marx JA SAEM emergency medicine faculty salary and benefits survey. Acad Emerg Med. 1999; 6: Kristal SL, Thompson BM, Marx JA SAEM emergency medicine faculty salary/benefits survey. Acad Emerg Med. 1998; 5: Emergency medicine management and physician compensation report Pittsburgh, PA: Daniel Stern and Associates, McCaig LF, Burt CW. National Hospital Ambulatory Medical Care Survey: 1999 Emergency Department Survey. Vital and Health Statistics of the Centers for Disease Control and Prevention/National Center for Health Statistics. Number 320, Jul McCaig L, Stussman B. National Hospital Ambulatory Medical Care Survey: 1996 Emergency Department Survey. Vital and Health Statistics of the Centers for Disease Control and Prevention/National Center for Health Statistics. Number 293, Dec 1997.