- Email: [email protected]
Satisfaction with practices: Emergency physicians versus internists
ORIGINAL CONTRIBUTION emergency medicine, practice; emergency physicians, satisfaction with practices
Satisfaction With Practices: Emergency Physicians Versus Internists No comparison has been made of satisfaction with practices of emergency physicians versus other specialists. We used a previously tested questionnaire that measures differences between current and preferred practice conditions, and with items regarding training and practice, and intention to change positions. It was sent to 250 physicians practicing emergency medicine and 250 practicing internal medicine in a metropolitan area. Both specialty types were defined by type of practice, rather than by specialty training or board certification. Ninety-five (38%) emergency physicians and 79 (32%) internists responded. Factor and correlation analyses identified six satisfaction indexes: resources, professional autonomy, administrative autonomy, patient relationships, professional relationships, and status. Physicians practicing emergency medicine were less satisfied, that is, reported more difference between current and preferred conditions, than physicians practicing internal medicine with professional autonomy, patient relations, and status (by t tests, P < .01 for each). Emergency physicians were more satisfied with professional relationships (P < .01). Only 40% of emergency physicians, versus 60% of internists, reported no intention to leave their present position within the next two years (P < .02). Expectation of position change by physicians practicing emergency medicine was predicted by dissatisfaction with professional autonomy, lack of board certification, recency of graduation from medical school, and belief that monetary compensation would be higher elsewhere (R square, 0.35; P < .01), while variables such as patient load, hours worked per week, and hospital size proved unimportant. We identified areas of dissatisfaction among emergency physicians that differed from those zlmong #lternists. If verified, educators and administrators may use our results to alter training and practice characteristics in ways that will enhance career satisfaction among emergency physicians. ]Murphy JG, Jacobson S: Satisfaction with practices: Emergency physicians versus internists. Ann Emerg Med March 1987;16:277-283.]
Jane G Murphy, PhD*? Sheldon Jacobson, MD, FACEP* Philadelphia, Pennsylvania From the Emergency Department, Section of General Medicine, Department of Medicine;* and Leonard Davis Institute of Health Economics,t University of Pennsylvania, Philadelphia, Pennsylvania. Received for publication May 23, 1986. Revision received August 25, 1986. Accepted for publication September 29, 1986. Presented at the University Association for Emergency Medicine Annual Meeting in Portland, Oregon, May 1986. Address for reprints: Jane G Murphy, PhD, Emergency Department, Silverstein/G1, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
INTRODUCTION The satisfaction of physicians with their practices has become a topic of increasing concern to administrators, health services researchers, and physicians themselves in recent years. The interest is more than academic. Dissatisfaction has been shown to be associated with both predicted and actual turnover among physicians in group practicesA,Z Satisfaction among internists has been linked to patient satisfaction, and to other patient-related variables such as continuity of care and no-show rates. 3 In addition, it has been suggested that job satisfaction may directly affect the performance of professionals, including physicians. 4 Much of the research to date has focused on physicians practicing in prepaid group practices or academically based clinics. Despite the continuing growth of the specialty of emergency medicine and the frequently discussed problem of "bum-out" among emergency physicians, there is virtually no information concerning how satisfied emergency physicians are with their practices. No systematic survey has addressed questions such as: With what aspects of their work are emergency physicians particularly satisfied and dissatisfied? What proportion of emergency physicians predict that they will change positions in the near future, and what elements of satisfaction or practice are related to those predictions? How do emergency physicians corn16:3 March 1987
Annals of Emergency Medicine
277/65
SATISFACTION WITH PRACTICES Murphy & Jacobson
pare w i t h p h y s i c i a n s p r a c t i c i n g in other specialties in regard to satisfaction and predicted position change? To address these issues, we used a previously tested questionnaire that measures differences between current and preferred practice conditions. Questionnaires were mailed to a sample of emergency and internal medic i n e p h y s i c i a n s in a m e t r o p o l i t a n area. I n t e r n i s t s were c h o s e n as t h e comparison group for two reasons. A significant p r o p o r t i o n of e m e r g e n c y physicians were initially trained as internists. Because p r a c t i c i n g i n t e r n a l m e d i c i n e is an alternative for many, the comparative satisfaction of internists w i t h their w o r k is particularly relevant. In addition, we w i s h e d to add to the work of Anwar et al, s,6 who compared the career patterns of emergency physicians, internists, and surgical residents.
MATERIALS AND METHODS Sample of Emergency Physicians and Internists T h e s a m p l e of p h y s i c i a n s w a s drawn from among those practicing emergency medicine or internal medicine in the Philadelphia metropolitan area. I n t e r n a l m e d i c i n e was defined broadly to include the practice of traditional internal medicine as well as general m e d i c i n e or family practice. Emergency m e d i c i n e was defined as the practice of emergency medicine, rather than by residency training or board eligibility or certification. The list of physicians eligible to participate was based primarily on a review of DorIand's Medical Directory. 7 D o r l a n d ' s p r o v i d e s n a m e s a n d addresses of physicians affiliated w i t h each area hospital by department. The roster of e m e r g e n c y p h y s i c i a n s was supplemented by the membership list of the Philadelphia Emergency Physicians' Society. A total of 355 emergency physicians and 1,233 internists was identified. S a m p l e s of 250 e m e r g e n c y p h y s i cians and 250 internists were chosen using a random n u m b e r table. N a m e s of p h y s i c i a n s w h o s e q u e s t i o n n a i r e s were returned as undeliverable were replaced by r a n d o m selection. Three mailings were conducted at six-week intervals between November 1984 and F e b r u a r y 1985. A n o n y m i t y of responses was guaranteed. A r a n d o m sample of 40% of nonrespondents was drawn. Year of gradu a t i o n from m e d i c a l school and cer-
66/278
TABLE 1. Factors and items
in final measures of physician satisfaction
Item Analysis* Factor 1: Satisfaction with Resources (a = 0.75)t Have nurses, aides, or technicians available Delegate specific clinical tasks Have secretaries available to clerical tasks Have administrators to handle administrative problems Have pharmaceuticals I want to prescribe available Have supplies I need available Obtain medical records I request Have examining room equipment in working order Have emergency equipment ready for use Have sufficient space to see patients Funds are available to make improvements Open positions for health staff are filled in two months Delegate routine clinical tasks to nurses or aides
0.61 0.47 0.63 0.58 0.55 0.62 0.51 0.68 0.55 0.51 0.62 0.54 0.46
Factor 2: Satisfaction with Professional Autonomy (a=0.47) Set the pace of my own work 0.73 Have inputs into decisions that affect my practice 0.84 Initiate changes in the way work is done 0.82 Factor 3: Satisfaction with Administrative Autonomy (a = 0.41) Must alter my practice to meet organizational rules 0.78 Report my activities to someone else 0.73 Receive direction from administrators 0.70 Factor 4: Satisfaction with Patient Relationships (a=0.55) Have doubts about whether patients are actually sick Deal with patients who are seriously disturbed or agitated Consider the possibility of legal reprisal from patients Consider the possibility of physical reprisal from patients Deal with patients who are trying to manipulate me
0.51 0.67 0.74 0.69 0.72
Factor 5: Satisfaction with Professional Relationships (a = 0.54) Participate in quality review activities 0.77 Discuss difficult cases with other physicians 0.76 Participate in teaching activities 0.68 Factor 6: Satisfaction with Status (a=0.38) Have input into major organizational decisions 0.72 Appreciated by nonclinical staff in the organization 0.62 Viewed as a colleague by others in local medical community 0,77 Praised by community residents for the nature of my work 0.83 *Correlation coefficients (Pearson's r's) between each item and the overall index. tCronbach's alpha coefficients. Annals of Emergency Medicine
16:3 March 1987
tification status were recorded from directories. 7,s
Questionnaire The questionnaire used was developed and t e s t e d by L i c h t e n s t e i n . 9. The instrument is based on the concept that a physician's "job dissatisfaction" is the difference between w h a t is experienced in his or her practice and what he or she would want or expect to find in an ideal position. The larger the difference between current experience and expectation, the more dissatisfied the physician is assumed to be with the current position. Following L i c h t e n s t e i n ' s m e t h o d , the questionnaire instructed each physician to compare attributes of his or her current position with an idealized position that "represents the best situation y o u w o u l d h a v e a r e a l i s t i c chance of o b t a i n i n g at t h e p r e s e n t time w i t h o u t a d d i t i o n a l t r a i n i n g . " Thirty-two attributes related to practice situations were listed. For each, the physician was asked to m a r k two points on a seven-point scale, one representing an evaluation of the current (C) position and one representing that of the preferred or best {B) state. For example: Set the pace of m y own work. Aimost n e v e i _ _ : C ~ : _ _ : _ _ : B : _ _ : - - - A l m o s t always 1 2 3 4 5 6 7
This set of r e s p o n s e s r e p r e s e n t s three units of dissatisfaction w i t h the current position (Best = 5, Current = 2, Dissatisfaction Score = 3). Respondents also assessed the likelihood t h a t they w o u l d change positions "during the next year or two" using a f i v e - p o i n t scale (0%, 25%, 50%, 75%, 100%). In addition, they estimated the a m o u n t more per year (in dollars) they would expect to earn from their "best possible job" if they worked the n u m b e r of hours devoted to their current position. Finally, there were questions con*Lichtenstein used a variety of methods, including Cronbach's alpha and correlation analyses, to test the reliability of the instrument. An ideal test of validity would have involved correlating measures of satisfaction with actual physician turnover. Since Lichtenstein's study was not longitudinal, validity was evaluated by relating specific elements of satisfaction to an overall job satisfaction rating and by correlating satisfaction with predicted turnover. 16:3 March 1987
cerning the physician's medical background ( m e d i c a l s c h o o l and year of graduation, specialty and certification status), and p r a c t i c e c h a r a c t e r i s t i c s (solo or type of group, salaried versus fee-for-service, n u m b e r s of p a t i e n t s s e e n a n d h o u r s w o r k e d per w e e k , length of t i m e in present position, and hospital size). D e m o g r a p h i c data included the physician's age, sex, and marital status.
Analyses Data were analyzed in four stages. First, c h a r a c t e r i s t i c s of r e s p o n d e n t s and n o n r e s p o n d e n t s were c o m p a r e d using chi-square and t tests. Next, satisfaction indexes were created and tested. To do so, the difference between a physician's rating of his or her c u r r e n t job and the preferred state was calculated for each job attribute. This resulted in a "dissatisfaction score" for each p h y s i c i a n on each attribute. These scores were transformed to "satisfaction scores" by subtracting each dissatisfaction score from 6, the m a x i m u m possible dissati s f a c t i o n s c o r e for e a c h e l e m e n t . Thus, each satisfaction score ranged from a low of 0 to a high of 6.* Satisfaction scores were factor analyzed to identify relationships among the 32 satisfaction items. (Factor analysis is a multivariate m e t h o d that allows one to represent a n u m b e r of intercorrelated variables in terms of a few conceptually reasonable and relatively i n d e p e n d e n t factors, lo The results of the factor analyses were used to gicoup items into satisfaction indexes. Indexes were tested for scalability using correlation analyses and for internal reliability using Cronbach's alpha test. Third, emergency physicians and internists were compared w i t h respect to p e r s o n a l a n d p r a c t i c e c h a r a c teristics and w i t h regard to scores on satisfaction indexes. Chi-square tests, M a n n - W h i t n e y U tests, and t t e s t s were used to evaluate the statistical significance of differences. Regression analyses were used to control for potential confounding by covariates of qf a physician rated the current position as I and the best position as 7, the dissatisfaction score for that element was 6 and the satisfaction score was 0 (6-6=0}. Conversely, if a physician rated the best and current positions equally, there was no discrepancy between ratings, the dissatisfaction score was 0 and the satisfaction score was 6 {6-0=6). Annals of Emergency Medicine
relationships between physician group and scores on satisfaction indexes. Finally, physicians' predictions conceming changing positions in the next year or two was used as a dependent variable. Differences b e t w e e n groups were evaluated using chi-square tests and r e g r e s s i o n analysis. R e g r e s s i o n a n a l y s e s also were used to i d e n t i f y variables related to these predictions among emergency physicians and internists.
RESULTS Ninety-five (38%) emergency physicians and 79 (32%) i n t e r n i s t s completed questionnaires. Respondents were more recent medical school graduates than were nonrespondents (1,966 _+ 10.3 years versus 1,961 + 11.8 years) (P < .01}. Respondents also were more likely to be board certified, but the difference was not s t a t i s t i c a l l y significant. In the original study, 9 Lichtenstein used factor analyses to evaluate relat i o n s h i p s a m o n g t h e 32 i t e m s included in our questionnaire. Six facets of s a t i s f a c t i o n w e r e i d e n t i f i e d : resources, professional autonomy (which he termed "self-directed autonomy"), administrative autonomy ("other-directed autonomy"), p a t i e n t relationships, professional relationships, and professional status. Factor analyses of our data confirmed the existence of the six elements of satisf a c t i o n . (One of L i c h t e n s t e i n ' s 32 items was dropped because it contributed to none of the factors in our analysis.) Factors and i t e m s are s h o w n (Table 1}. Indexes corresponding to the six sati s f a c t i o n e l e m e n t s were f o r m e d by s u m m i n g responses across items. The results of correlation analyses of relationships between items and indexes, and of Cronbach's alpha test for internal reliability also are included (Table 1). (Cronbach's alpha ranges from 0.0 to 1.0; a score of 0.5 or higher is considered desirable.) The high correlations between items and indexes are comparable to those found by Lichtenstein and indicate that the manner in which items were grouped into indexes was reasonable. Lichtenstein did not report' Cronbach's alpha levels for individual indexes. Our results suggest that the indexes reflecting satisfaction w i t h administrative a u t o n o m y and w i t h status are less reliable than the others. Results reported w i t h regard to t h e s e i n d e x e s s h o u l d be re279/67
SATISFACTION WITH PRACTICES Murphy & Jacobson
garded as tentative. Finally, intercorrelations among indexes were calculated. As would be expected of indexes developed using factor analyses, corr e l a t i o n coefficients were r e l a t i v e l y low, ranging from 0.10 to 0.56, with 13 of the 15 correlation coefficients lower than 0.40. Characteristics of emergency physicians and internists are shown (Table 2). Emergency physicians were significantly younger, more recent graduates of medical school, and were in their current positions for shorter periods of time. More than one-quarter of emergency physicians were board certified in emergency medicine; an additional 22% were certified in internal medicine or family practice. The pattern of certification between emergency medicine and internal m e d i c i n e respondents was significantly different; however, l o o k i n g o n l y at c e r t i f i c a t i o n versus noncertification, there was no significant difference between groups. There was no significant difference between emergency medicine and internal medicine respondents with respect to average n u m b e r s of h o u r s worked per week (between 42 and 44 hours). The average emergency physician practiced in a hospital w i t h 308 beds and in an emergency department seeing approximately 530 patients per week. More than two-thirds were salaried and one-third were hospital emp l o y e e s . T h e average i n t e r n i s t saw nearly 100 patients per week in his or her practice. Internists tended to be in solo practice (63%) and to be paid on a fee-for-service basis (70%). Scores on satisfaction indexes are c o m p a r e d (Table 3). Overall, p h y s i cians reported high levels of satisfaction w i t h current positions. All m e a n s a t i s f a c t i o n scores were m o r e t h a n t w o - t h i r d s of t h e h i g h e s t p o s s i b l e scores. For example, the m e a n score on the professional a u t o n o m y index for e m e r g e n c y physicians was 12.24, 68% of the m a x i m u m possible score of 18. The m e a n score on the status i n d e x a m o n g i n t e r n i s t s was 20.52, 86% of the total possible score of 24. Despite the high levels of satisfaction overall, there were statistically significant differences between groups with respect to four indexes. Emergency p h y s i c i a n s w e r e m o r e s a t i s f i e d w i t h professional relationships than were their internal medicine counterp a r t s . I n t e r n i s t s were s i g n i f i c a n t l y more satisfied with professional autonomy, patient relationships, and 68/280
TABLE 2. Comparison of characteristics of emergency physicians
and internists
Characteristics
Age (in years) Year graduated from medical school Years in current position Hours worked/week Patients seen/week: In ED In internal medicine practice Beds in hospital with ED
Marital status Degree Certification
Payment
40.0 _+ 9.8
50.2 + 13.5"
1968 -- 1 0 . 0 6.1 _+ 5.7 42.2 ± 7.6
1964 ± 13.1t 16.5 ± 12.2t 44.3 ± 16.2
No. (%)
Emergency medicine Internal medicine/ family practice Other None Salaried Fee-for-service Mix
Practice type: Emergency medicine Hospital employee Personal contract Group contract No contract Internal medicine Solo practice Group practice *P < .01. tp < .05. 1Comparison between groups not appropriate.
Annals of Emergency Medicine
Mean __ SD
307.6 ± 144.5
Male Female Now married Not now married MD DO
status. There were no significant differences between groups with respect to satisfaction w i t h resources or with administrative autonomy. Regression a n a l y s e s were used to t e s t w h e t h e r t h e r e l a t i o n s h i p s between p h y s i c i a n group and satisfaction w i t h professional relationships, p r o f e s s i o n a l a u t o n o m y , p a t i e n t relat i o n s h i p s , and s t a t u s w o u l d r e m a i n
Internists
Mean ___ SD
529.7 ± 244.6
Categories
Sex
Emergency Physicians
75 20 73 22
(79.0) (21.0) (76.8) (23.2)
81 (85.3) 14 (14.7) 25 (26.3) 21 11 38 65 23 7
(22.1) (11.6) (40.0) (68.4) (24.2) (7.4)
31 23 38 3
(32.6) (24.2) (40.0) (3.2)
99.5 ± 51.0, N/A No. (%)
68 11 71 8
(86.1) (13.9) (89.9) (10.1)t
69 (87.3) 10 (12.7) 0 47 10 22 16 55 8
(59.5) (12.7) (27.8)* (20.3) (69.6) (10.1)*
50 (63.3) 29 (36.7)*
once physician age, year of graduation from medical school, and years in current position were controlled. (Emergency physicians and internists were significantly different w i t h respect to these three variables, suggesting their inclusion as covariates.) Group membership was the most significant predictor of scores on each satisfaction index. When age, year of graduation, 16:3 March 1987
TABLE 3. Comparison of the satisfaction of emergency physicians and internists
Satisfaction Index (Potential Range) Resources (0-78) Professional autonomy (0-18) Administrative autonomy (0-18) Patient relationships (0-30) Professional relationships (0-18) Status (0-24) *P < .01; t tests and Mann-Whitney U tests.
Emergency Physicians Mean % of Total ___ SD Possible 58.17 74.6 _+ 11,17 12.24 68,0 _+ 4.27 14.06 78.1 _+ 3.58 21,57 71.9 _+ 5,41 14.72 81.8 -+ 3.09 17,23 71.8 + 4.62
Internists _+ _+ + _+ _+ _+ _+
Mean SD 61,05 10.94 14.03" 3.61 14.39 3.82 23.98* 4.33 13.1 O* 3.87 20.52* 3.41
% of Total Possible 78.3 77.9 79.9 79.9 72.8 85.5
TABLE 4. Comparison of expectation of position change in next two years (excluding those planning to retire)*
Emergency Physicians (%) Internists (%) *P < .001.
and years in position were controlled, the relationships between group and satisfaction with patient relations and status remained significant, while those between group and satisfaction with professional relationships and with professional autonomy did not. As described previously, an additional element of satisfaction, satisfaction with pay, also was measured. Emergency physicians and internists each estimated that they could earn, on average, a p p r o x i m a t e l y $23,500 more in the idealized "best" position (difference between groups not significant). Finally, p r e d i c t i o n s of p o s i t i o n change were analyzed. Forty percent of emergency physicians and 60% of internists reported no intention to leave their current position within the next two years (P < .02). Excluding those considering retirement, the difference was even more striking (Table 4). One emergency physician and 11 internists reported considering retire16:3 March 1987
0% 38 (40.4) 47 (69.1)
Likelihood of Changing Positions 25% 50% 75% 23 12 10 (24.5) (12.8) (10.6) 9 12 (13.2) (17.6)
ment. Among the remaining, more than one-fifth of emergency physicians but no internists estimated that their chances of leaving their current positions were 75% or more. When age, year of graduation from medical school, and years in current position were controlled, the significant relationship between intention to change position and group remained. Notably, the vast majority of those considering changes apparently intend to remain within their chosen specialty. Seven emergency physicians (7.4%) described "best" positions in other than emergency settings. Just two internists (2.9%) described "best" positions in settings outside internal medicine. Regression analyses were used to relate likelihood of position change to elements of satisfaction, and emergency physicians' and internists' personal and practice characteristics. Among emergency physicians not considering retirement, likelihood was predicted Annals of Emergency Medicine
100% 11 (11.7)
Total 94
68
(in order of importance) by dissatisfaction with professional autonomy, lack of board certification, recency of graduation from medical school, and belief that m o n e t a r y compensation would be greater elsewhere (R square, 0.35; P < .01). Given the clustering of responses among internists, likelihood of change was dichotomized (no intention of leaving versus 25% to 50% chance). D i s s a t i s f a c t i o n w i t h resources was the sole predictor of change (R square, 0.29; P < .01). Variables such as p a t i e n t load; hours worked per week; years in current position; payment or practice type; physician's sex or age; and, for emergency physicians, hospital size, were not significant predictors of expectations of changing position a m o n g either emergency physicians or internists.
DISCUSSION This survey of emergency physicians and internists practicing in the 281/69
SATISFACTION WITH PRACTICES Murphy & Jacobson
Philadelphia area was designed to address three questions: With what aspects of their work are emergency physicians particularly satisfied and dissatisfied? What proportion of emergency physicians predict that they will change positions in the near future, and what elements of satisfaction or practice are related to those predictions? How do emergency physicians compare w i t h p h y s i c i a n s practicing in other specialties with regard to satisfaction and predicted position change? The findings can be s u m m a r i z e d with respect to these questions and can be compared with results of related research. Emergency physicians reported relatively high levels of satisfaction with all facets of practice measured. Average scores were more than two-thirds of possible scores, ranging from 68% to 82%. Variability in levels of satisfaction appeared to be related to the nature of emergency practice. The highest level was reported with respect to professional relationships. Emergency medicine requires physicians to work together, giving them ample opportunity to discuss cases both with their colleagues and with consultants. In many settings, emergency physicians also teach medical students and/or residents. T h e l o w e s t level of s a t i s f a c t i o n among emergency physicians was reported with regard to professional autonomy. The hospitals in which they practice are bureaucratic settings. Previous studies of prepaid group practices by Mechanicll, lz and othersl3,14 have suggested that bureaucratization is inversely related to physicians' satisfaction with autonomy. Thus, it is not surprising t h a t the e m e r g e n c y physicians were relatively dissatisfied with their abilities to have input into decisions affecting their practices, and with their abilities to initiate changes in the way they work. Twenty-two percent of emergency physicians reported that there was at least a 75% chance that they would change positions within the next year or two. This compares with Demlo's ~s report that 10% of physicians in large prepaid groups and just 6.6% of solo practitioners predicted a similar likelihood of changing positions. Excluding those planning to retire, no internists in our s a m p l e m a d e similar predictions. 70/282
In her survey of residency-trained emergency physicians, Anwar 6 noted that many emergency medicine positions were available and that emergency physicians had changed jobs relatively often in an effort to "find the perfect position." A review of advertisements for emergency physicians confirms that m a n y postions remain available. The variety of opportunities may encourage some emergency physicians to change jobs - - or at least to consider changing jobs - - relatively frequently. A n o t h e r possible e x p l a n a t i o n for the high level of interest in changing positions is the relative youth of the emergency medicine respondents (mean age = 40.0 _+ 9.8 years). Herman I6 and Tilson I7 found that younger physicians were more likely than their older counterparts to leave group practices. H e r m a n conjectured that the y o u n g e r d o c t o r s h a d n o t y e t reached a "stable career stage." It may be that as the emergency physicians in this study become older, they will express interest in changing positions less frequently. Intention to change positions was related to level of satisfaction w i t h professional autonomy, certification status, year of graduation from medical school, and a s s e s s m e n t of the monetary rewards related to an idealized position. As noted above, the issue of professional autonomy has been shown to be particularly salient for p h y s i c i a n s in group practices and other b u r e a u c r a t i c settings. H-14 It would have been surprising if emergency physicians were not sensitive to this aspect of their practices. Because y e a r of g r a d u a t i o n f r o m m e d i c a l school is highly correlated with age, and because younger emergency physicians still may be seeking certificat i o n status, t h e career stage phenomenon described by Herman 16 may explain the relationship between predicted position change and these variables. Tilson I7 also reported a significant positive association between dissatisfaction with pay and physician turnover. That association probably is especially strong a m o n g emergency physicians, for w h o m changing positions, rather t h a n e x p a n d i n g one's practice, is generally the only way in which to increase income. Comparing emergency physicians' responses w i t h those of internists from the same geographic area is instructive. It is important to note that Annals of Emergency Medicine
average levels of satisfaction were high for all physicians with respect to all aspects of practices measured. Nevertheless, even when age, year of graduation from medical school, and years in c u r r e n t p o s i t i o n were controlled, emergency physicians were significantly less satisfied with patient relationships and status than were their internal medicine counterparts. With respect to p a t i e n t relationships, emergency physicians are dedicated to treating emergently ill patients. But, they often find themselves caring for patients with nonurgent problems. Because internists expect to see patients with nonemergency conditions, they do not c o m m o n l y face the gap between expectations regarding patients and the realities of patient care with w h i c h emergency physicians must contend. In addition, internists have an opportunity to develop long-lasting relationships with their patients; the continuity of care that often makes for satisfying physicianpatient relationships is not usually an aspect of emergency medicine. The relative dissatisfaction of emergency physicians with status probably reflects the fact that emergency medicine is still a developing specialty. Anwar s found that emergency medicine residents chose their specialty despite the "normal patterns of influence" in m e d i c a l school. O n c e in practice, however, emergency physicians in this study were less satisfied with their standing in their professional communities than were their counterparts in m o r e traditional i n t e r n a l medicine practices. The results of this study m u s t be evaluated with respect to the limitations of the research design. The response rate was relatively low, 35% overall, and the study was conducted in a single g e o g r a p h i c area. Even t h o u g h respondents and nonrespondents were similar with respect to certification status, more broadly based studies are needed to test whether, or how, either the response rate or the selection of a single geographic area affected the generalizability of the findings. A larger s t u d y also could define e m e r g e n c y m e d i c i n e and i n t e r n a l medicine more precisely. For example, while we defined emergency medicine broadly by practice location, it is possible that the satisfaction of emergency-residency-trained or certified emergency physicians differs from that of 16:3 March 1987
the wider group of p h y s i c i a n s w i t h other training who practice in emergency settings. Likewise, physicians certified in internal medicine or who practice in traditional departments of internal m e d i c i n e m a y differ f r o m those certified in family practice or who practice in relatively n e w departments-of family practice. Our sample was too small to stratify in ways that would allow us to test these possibilities. In addition, it is not k n o w n w h a t proportion of physicians who predict that they will change positions actually will do so. A longitudinal study would be required to test w h e th e r elements of satisfaction are indeed related to position changes among emergency physicians. Such a study also could evaluate the effects of changes in the career stages of the (now relatively young) emergency physicians. Further research also should include additional variables, the potential importance of wh i ch is suggested by the present work. For example, the residency t r a i n i n g status of e m e r g e n c y physicians s h o u l d be i n c l u d e d as a variable that might help explain variations in physicians' satisfaction w i t h patient relations or status. Variables that could suggest the effect of variation in satisfaction on clinical practice also should be adde& Finally, the possibility that "satisfaction" should be redefined for emerging medical specialties such as emergency medicine should be considered in further research. The data collection ins t r u m e n t used addressed the issues traditionally i m p o r t a n t to physicians in the past. However, it did not include i t e m s r e l a t e d t o s a t i s f a c t i o n with flexibility of scheduling, with opportunities for mobility, or w i t h variety of clinical experience. It may be
16:3 March 1987
that the e m e r g e n c y physicians place different values on aspects of t h ei r careers t h a n do t h o se in o t h e r spec i a l ti e s , and t h a t t h e y are, by and large, well satisfied w i t h dimensions not measured in the present research. Despite its limitations, the research is important because it is the first syst e m a t i c effort t o a s s e s s the satisfact i o n of e m e r g e n c y p h y s i c i a n s w i t h their practices, and to compare levels of satisfaction with those of a set of th e i r colleagues. Moreover, satisfact i o n was e v a l u a t e d w i t h an instrum e n t whose reliability had been tested, and t h a t a l l o w e d d i m e n s i o n s of satisfaction to be measured. CONCLUSION It is reassuring to find that emergency physicians are relatively well satisfied w i t h their practices. But, there is room for improvement. If verified, educators and administrators m a y use these results to alter training and practice characteristics in ways that will further enhance career satisfaction a m o n g e m e r g e n c y p h y s i c i a n s . Increased satisfaction may, in turn, reduce turnover, and limit the n u m b e r of e m e r g e n c y physicians w h o "burn out" or b e c o m e alienated f r o m their chosen specialty. The authors thank Stephanie Abbuhl, MD~ GW Crooks, MD; Gladys S Fenichel, MD; C William Hanson, Jr, MD; and Alison McDonald, MD, for their helpful comments.
REFERENCES 1. Prybil L: Characteristics, career patterns and opinions of physicians who practice in large muhispecialty groups. Medical Group Management I974;21:22-26. 2. Ross A: A report on physician termination in group practice. Medical Group Management 1969;I6:I5-21.
Annals of Emergency Medicine
3. Linn L8, Brook RH, Clark VA: Physician and patient satisfaction as factors related to the organization of internal medicine group practices. Medical Care 1985;23:1171-1178. 4. Likert R: N e w Patterns of Management. New York, McGraw-Hill, 1961. 5. Anwar RAH: Trends in training: Focus on emergency medicine. Ann Emerg Med 1980;9: 60-71. 6. Anwar RAH: A longitudinal study of residency-trained emergency physicians. Ann Emerg Med 1983;12:20-24. 7. Dorland's Medical Directory: Philadelphia and Metropolitan Area. Philadelphia, Packard
Press, 1983. 8. American Medical Directory. Chicago, American Medical Association, 1984. 9. Lichtenstein R: Measuring the job satisfaction of physicians in organized settings. Medical Care 1984;22:56-66. 10. Kleinbaum DG, Kupper LL: Applied Regression Analysis and Other Multivariate Methods.
North Scituate, Massachusetts, Duxbury Press, 1978. 11. Mechanic D: General medical practice: Some comparisons between the work of primary care physicians in the United States and England and Wales. Medical Care 1972;10:402-420. 12. Mechanic D: The organization of medical practice and practice orientations among physicians in prepaid and non-prepaid primary care settings. Medical Care I975i13:189-204. I3. Ben-David J: The professional role of the physician in bureaucratized medicine: A study in role-conflict. Human Relations 1958;2: 255-274. 14. McElrath DC: Perspectives and participation of physicians in prepaid group practice. American Sociological Review 1961;26:596-607. 15. Demlo L: The relationship of physician practice patterns to organizationai and personal characteristics {Unpublished doctoral dissertation). New Haven, Connecticut, Yale University, 1975. 16. Herman JG, Dunham RB, Hulin CL: Organizational structure, demographic characteristics, and employee responses. Organizational Behavior and Human Performance
1975;I3:206-232. I7. Tilson HH: Stability of physician employment in Neighborhood Health Centers. Medical Care 1973;11:384-400.
283/71
Satisfaction With Practices: Emergency Physicians Versus Internists No comparison has been made of satisfaction with practices of emergency physicians versus other specialists. We used a previously tested questionnaire that measures differences between current and preferred practice conditions, and with items regarding training and practice, and intention to change positions. It was sent to 250 physicians practicing emergency medicine and 250 practicing internal medicine in a metropolitan area. Both specialty types were defined by type of practice, rather than by specialty training or board certification. Ninety-five (38%) emergency physicians and 79 (32%) internists responded. Factor and correlation analyses identified six satisfaction indexes: resources, professional autonomy, administrative autonomy, patient relationships, professional relationships, and status. Physicians practicing emergency medicine were less satisfied, that is, reported more difference between current and preferred conditions, than physicians practicing internal medicine with professional autonomy, patient relations, and status (by t tests, P < .01 for each). Emergency physicians were more satisfied with professional relationships (P < .01). Only 40% of emergency physicians, versus 60% of internists, reported no intention to leave their present position within the next two years (P < .02). Expectation of position change by physicians practicing emergency medicine was predicted by dissatisfaction with professional autonomy, lack of board certification, recency of graduation from medical school, and belief that monetary compensation would be higher elsewhere (R square, 0.35; P < .01), while variables such as patient load, hours worked per week, and hospital size proved unimportant. We identified areas of dissatisfaction among emergency physicians that differed from those zlmong #lternists. If verified, educators and administrators may use our results to alter training and practice characteristics in ways that will enhance career satisfaction among emergency physicians. ]Murphy JG, Jacobson S: Satisfaction with practices: Emergency physicians versus internists. Ann Emerg Med March 1987;16:277-283.]
Jane G Murphy, PhD*? Sheldon Jacobson, MD, FACEP* Philadelphia, Pennsylvania From the Emergency Department, Section of General Medicine, Department of Medicine;* and Leonard Davis Institute of Health Economics,t University of Pennsylvania, Philadelphia, Pennsylvania. Received for publication May 23, 1986. Revision received August 25, 1986. Accepted for publication September 29, 1986. Presented at the University Association for Emergency Medicine Annual Meeting in Portland, Oregon, May 1986. Address for reprints: Jane G Murphy, PhD, Emergency Department, Silverstein/G1, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
INTRODUCTION The satisfaction of physicians with their practices has become a topic of increasing concern to administrators, health services researchers, and physicians themselves in recent years. The interest is more than academic. Dissatisfaction has been shown to be associated with both predicted and actual turnover among physicians in group practicesA,Z Satisfaction among internists has been linked to patient satisfaction, and to other patient-related variables such as continuity of care and no-show rates. 3 In addition, it has been suggested that job satisfaction may directly affect the performance of professionals, including physicians. 4 Much of the research to date has focused on physicians practicing in prepaid group practices or academically based clinics. Despite the continuing growth of the specialty of emergency medicine and the frequently discussed problem of "bum-out" among emergency physicians, there is virtually no information concerning how satisfied emergency physicians are with their practices. No systematic survey has addressed questions such as: With what aspects of their work are emergency physicians particularly satisfied and dissatisfied? What proportion of emergency physicians predict that they will change positions in the near future, and what elements of satisfaction or practice are related to those predictions? How do emergency physicians corn16:3 March 1987
Annals of Emergency Medicine
277/65
SATISFACTION WITH PRACTICES Murphy & Jacobson
pare w i t h p h y s i c i a n s p r a c t i c i n g in other specialties in regard to satisfaction and predicted position change? To address these issues, we used a previously tested questionnaire that measures differences between current and preferred practice conditions. Questionnaires were mailed to a sample of emergency and internal medic i n e p h y s i c i a n s in a m e t r o p o l i t a n area. I n t e r n i s t s were c h o s e n as t h e comparison group for two reasons. A significant p r o p o r t i o n of e m e r g e n c y physicians were initially trained as internists. Because p r a c t i c i n g i n t e r n a l m e d i c i n e is an alternative for many, the comparative satisfaction of internists w i t h their w o r k is particularly relevant. In addition, we w i s h e d to add to the work of Anwar et al, s,6 who compared the career patterns of emergency physicians, internists, and surgical residents.
MATERIALS AND METHODS Sample of Emergency Physicians and Internists T h e s a m p l e of p h y s i c i a n s w a s drawn from among those practicing emergency medicine or internal medicine in the Philadelphia metropolitan area. I n t e r n a l m e d i c i n e was defined broadly to include the practice of traditional internal medicine as well as general m e d i c i n e or family practice. Emergency m e d i c i n e was defined as the practice of emergency medicine, rather than by residency training or board eligibility or certification. The list of physicians eligible to participate was based primarily on a review of DorIand's Medical Directory. 7 D o r l a n d ' s p r o v i d e s n a m e s a n d addresses of physicians affiliated w i t h each area hospital by department. The roster of e m e r g e n c y p h y s i c i a n s was supplemented by the membership list of the Philadelphia Emergency Physicians' Society. A total of 355 emergency physicians and 1,233 internists was identified. S a m p l e s of 250 e m e r g e n c y p h y s i cians and 250 internists were chosen using a random n u m b e r table. N a m e s of p h y s i c i a n s w h o s e q u e s t i o n n a i r e s were returned as undeliverable were replaced by r a n d o m selection. Three mailings were conducted at six-week intervals between November 1984 and F e b r u a r y 1985. A n o n y m i t y of responses was guaranteed. A r a n d o m sample of 40% of nonrespondents was drawn. Year of gradu a t i o n from m e d i c a l school and cer-
66/278
TABLE 1. Factors and items
in final measures of physician satisfaction
Item Analysis* Factor 1: Satisfaction with Resources (a = 0.75)t Have nurses, aides, or technicians available Delegate specific clinical tasks Have secretaries available to clerical tasks Have administrators to handle administrative problems Have pharmaceuticals I want to prescribe available Have supplies I need available Obtain medical records I request Have examining room equipment in working order Have emergency equipment ready for use Have sufficient space to see patients Funds are available to make improvements Open positions for health staff are filled in two months Delegate routine clinical tasks to nurses or aides
0.61 0.47 0.63 0.58 0.55 0.62 0.51 0.68 0.55 0.51 0.62 0.54 0.46
Factor 2: Satisfaction with Professional Autonomy (a=0.47) Set the pace of my own work 0.73 Have inputs into decisions that affect my practice 0.84 Initiate changes in the way work is done 0.82 Factor 3: Satisfaction with Administrative Autonomy (a = 0.41) Must alter my practice to meet organizational rules 0.78 Report my activities to someone else 0.73 Receive direction from administrators 0.70 Factor 4: Satisfaction with Patient Relationships (a=0.55) Have doubts about whether patients are actually sick Deal with patients who are seriously disturbed or agitated Consider the possibility of legal reprisal from patients Consider the possibility of physical reprisal from patients Deal with patients who are trying to manipulate me
0.51 0.67 0.74 0.69 0.72
Factor 5: Satisfaction with Professional Relationships (a = 0.54) Participate in quality review activities 0.77 Discuss difficult cases with other physicians 0.76 Participate in teaching activities 0.68 Factor 6: Satisfaction with Status (a=0.38) Have input into major organizational decisions 0.72 Appreciated by nonclinical staff in the organization 0.62 Viewed as a colleague by others in local medical community 0,77 Praised by community residents for the nature of my work 0.83 *Correlation coefficients (Pearson's r's) between each item and the overall index. tCronbach's alpha coefficients. Annals of Emergency Medicine
16:3 March 1987
tification status were recorded from directories. 7,s
Questionnaire The questionnaire used was developed and t e s t e d by L i c h t e n s t e i n . 9. The instrument is based on the concept that a physician's "job dissatisfaction" is the difference between w h a t is experienced in his or her practice and what he or she would want or expect to find in an ideal position. The larger the difference between current experience and expectation, the more dissatisfied the physician is assumed to be with the current position. Following L i c h t e n s t e i n ' s m e t h o d , the questionnaire instructed each physician to compare attributes of his or her current position with an idealized position that "represents the best situation y o u w o u l d h a v e a r e a l i s t i c chance of o b t a i n i n g at t h e p r e s e n t time w i t h o u t a d d i t i o n a l t r a i n i n g . " Thirty-two attributes related to practice situations were listed. For each, the physician was asked to m a r k two points on a seven-point scale, one representing an evaluation of the current (C) position and one representing that of the preferred or best {B) state. For example: Set the pace of m y own work. Aimost n e v e i _ _ : C ~ : _ _ : _ _ : B : _ _ : - - - A l m o s t always 1 2 3 4 5 6 7
This set of r e s p o n s e s r e p r e s e n t s three units of dissatisfaction w i t h the current position (Best = 5, Current = 2, Dissatisfaction Score = 3). Respondents also assessed the likelihood t h a t they w o u l d change positions "during the next year or two" using a f i v e - p o i n t scale (0%, 25%, 50%, 75%, 100%). In addition, they estimated the a m o u n t more per year (in dollars) they would expect to earn from their "best possible job" if they worked the n u m b e r of hours devoted to their current position. Finally, there were questions con*Lichtenstein used a variety of methods, including Cronbach's alpha and correlation analyses, to test the reliability of the instrument. An ideal test of validity would have involved correlating measures of satisfaction with actual physician turnover. Since Lichtenstein's study was not longitudinal, validity was evaluated by relating specific elements of satisfaction to an overall job satisfaction rating and by correlating satisfaction with predicted turnover. 16:3 March 1987
cerning the physician's medical background ( m e d i c a l s c h o o l and year of graduation, specialty and certification status), and p r a c t i c e c h a r a c t e r i s t i c s (solo or type of group, salaried versus fee-for-service, n u m b e r s of p a t i e n t s s e e n a n d h o u r s w o r k e d per w e e k , length of t i m e in present position, and hospital size). D e m o g r a p h i c data included the physician's age, sex, and marital status.
Analyses Data were analyzed in four stages. First, c h a r a c t e r i s t i c s of r e s p o n d e n t s and n o n r e s p o n d e n t s were c o m p a r e d using chi-square and t tests. Next, satisfaction indexes were created and tested. To do so, the difference between a physician's rating of his or her c u r r e n t job and the preferred state was calculated for each job attribute. This resulted in a "dissatisfaction score" for each p h y s i c i a n on each attribute. These scores were transformed to "satisfaction scores" by subtracting each dissatisfaction score from 6, the m a x i m u m possible dissati s f a c t i o n s c o r e for e a c h e l e m e n t . Thus, each satisfaction score ranged from a low of 0 to a high of 6.* Satisfaction scores were factor analyzed to identify relationships among the 32 satisfaction items. (Factor analysis is a multivariate m e t h o d that allows one to represent a n u m b e r of intercorrelated variables in terms of a few conceptually reasonable and relatively i n d e p e n d e n t factors, lo The results of the factor analyses were used to gicoup items into satisfaction indexes. Indexes were tested for scalability using correlation analyses and for internal reliability using Cronbach's alpha test. Third, emergency physicians and internists were compared w i t h respect to p e r s o n a l a n d p r a c t i c e c h a r a c teristics and w i t h regard to scores on satisfaction indexes. Chi-square tests, M a n n - W h i t n e y U tests, and t t e s t s were used to evaluate the statistical significance of differences. Regression analyses were used to control for potential confounding by covariates of qf a physician rated the current position as I and the best position as 7, the dissatisfaction score for that element was 6 and the satisfaction score was 0 (6-6=0}. Conversely, if a physician rated the best and current positions equally, there was no discrepancy between ratings, the dissatisfaction score was 0 and the satisfaction score was 6 {6-0=6). Annals of Emergency Medicine
relationships between physician group and scores on satisfaction indexes. Finally, physicians' predictions conceming changing positions in the next year or two was used as a dependent variable. Differences b e t w e e n groups were evaluated using chi-square tests and r e g r e s s i o n analysis. R e g r e s s i o n a n a l y s e s also were used to i d e n t i f y variables related to these predictions among emergency physicians and internists.
RESULTS Ninety-five (38%) emergency physicians and 79 (32%) i n t e r n i s t s completed questionnaires. Respondents were more recent medical school graduates than were nonrespondents (1,966 _+ 10.3 years versus 1,961 + 11.8 years) (P < .01}. Respondents also were more likely to be board certified, but the difference was not s t a t i s t i c a l l y significant. In the original study, 9 Lichtenstein used factor analyses to evaluate relat i o n s h i p s a m o n g t h e 32 i t e m s included in our questionnaire. Six facets of s a t i s f a c t i o n w e r e i d e n t i f i e d : resources, professional autonomy (which he termed "self-directed autonomy"), administrative autonomy ("other-directed autonomy"), p a t i e n t relationships, professional relationships, and professional status. Factor analyses of our data confirmed the existence of the six elements of satisf a c t i o n . (One of L i c h t e n s t e i n ' s 32 items was dropped because it contributed to none of the factors in our analysis.) Factors and i t e m s are s h o w n (Table 1}. Indexes corresponding to the six sati s f a c t i o n e l e m e n t s were f o r m e d by s u m m i n g responses across items. The results of correlation analyses of relationships between items and indexes, and of Cronbach's alpha test for internal reliability also are included (Table 1). (Cronbach's alpha ranges from 0.0 to 1.0; a score of 0.5 or higher is considered desirable.) The high correlations between items and indexes are comparable to those found by Lichtenstein and indicate that the manner in which items were grouped into indexes was reasonable. Lichtenstein did not report' Cronbach's alpha levels for individual indexes. Our results suggest that the indexes reflecting satisfaction w i t h administrative a u t o n o m y and w i t h status are less reliable than the others. Results reported w i t h regard to t h e s e i n d e x e s s h o u l d be re279/67
SATISFACTION WITH PRACTICES Murphy & Jacobson
garded as tentative. Finally, intercorrelations among indexes were calculated. As would be expected of indexes developed using factor analyses, corr e l a t i o n coefficients were r e l a t i v e l y low, ranging from 0.10 to 0.56, with 13 of the 15 correlation coefficients lower than 0.40. Characteristics of emergency physicians and internists are shown (Table 2). Emergency physicians were significantly younger, more recent graduates of medical school, and were in their current positions for shorter periods of time. More than one-quarter of emergency physicians were board certified in emergency medicine; an additional 22% were certified in internal medicine or family practice. The pattern of certification between emergency medicine and internal m e d i c i n e respondents was significantly different; however, l o o k i n g o n l y at c e r t i f i c a t i o n versus noncertification, there was no significant difference between groups. There was no significant difference between emergency medicine and internal medicine respondents with respect to average n u m b e r s of h o u r s worked per week (between 42 and 44 hours). The average emergency physician practiced in a hospital w i t h 308 beds and in an emergency department seeing approximately 530 patients per week. More than two-thirds were salaried and one-third were hospital emp l o y e e s . T h e average i n t e r n i s t saw nearly 100 patients per week in his or her practice. Internists tended to be in solo practice (63%) and to be paid on a fee-for-service basis (70%). Scores on satisfaction indexes are c o m p a r e d (Table 3). Overall, p h y s i cians reported high levels of satisfaction w i t h current positions. All m e a n s a t i s f a c t i o n scores were m o r e t h a n t w o - t h i r d s of t h e h i g h e s t p o s s i b l e scores. For example, the m e a n score on the professional a u t o n o m y index for e m e r g e n c y physicians was 12.24, 68% of the m a x i m u m possible score of 18. The m e a n score on the status i n d e x a m o n g i n t e r n i s t s was 20.52, 86% of the total possible score of 24. Despite the high levels of satisfaction overall, there were statistically significant differences between groups with respect to four indexes. Emergency p h y s i c i a n s w e r e m o r e s a t i s f i e d w i t h professional relationships than were their internal medicine counterp a r t s . I n t e r n i s t s were s i g n i f i c a n t l y more satisfied with professional autonomy, patient relationships, and 68/280
TABLE 2. Comparison of characteristics of emergency physicians
and internists
Characteristics
Age (in years) Year graduated from medical school Years in current position Hours worked/week Patients seen/week: In ED In internal medicine practice Beds in hospital with ED
Marital status Degree Certification
Payment
40.0 _+ 9.8
50.2 + 13.5"
1968 -- 1 0 . 0 6.1 _+ 5.7 42.2 ± 7.6
1964 ± 13.1t 16.5 ± 12.2t 44.3 ± 16.2
No. (%)
Emergency medicine Internal medicine/ family practice Other None Salaried Fee-for-service Mix
Practice type: Emergency medicine Hospital employee Personal contract Group contract No contract Internal medicine Solo practice Group practice *P < .01. tp < .05. 1Comparison between groups not appropriate.
Annals of Emergency Medicine
Mean __ SD
307.6 ± 144.5
Male Female Now married Not now married MD DO
status. There were no significant differences between groups with respect to satisfaction w i t h resources or with administrative autonomy. Regression a n a l y s e s were used to t e s t w h e t h e r t h e r e l a t i o n s h i p s between p h y s i c i a n group and satisfaction w i t h professional relationships, p r o f e s s i o n a l a u t o n o m y , p a t i e n t relat i o n s h i p s , and s t a t u s w o u l d r e m a i n
Internists
Mean ___ SD
529.7 ± 244.6
Categories
Sex
Emergency Physicians
75 20 73 22
(79.0) (21.0) (76.8) (23.2)
81 (85.3) 14 (14.7) 25 (26.3) 21 11 38 65 23 7
(22.1) (11.6) (40.0) (68.4) (24.2) (7.4)
31 23 38 3
(32.6) (24.2) (40.0) (3.2)
99.5 ± 51.0, N/A No. (%)
68 11 71 8
(86.1) (13.9) (89.9) (10.1)t
69 (87.3) 10 (12.7) 0 47 10 22 16 55 8
(59.5) (12.7) (27.8)* (20.3) (69.6) (10.1)*
50 (63.3) 29 (36.7)*
once physician age, year of graduation from medical school, and years in current position were controlled. (Emergency physicians and internists were significantly different w i t h respect to these three variables, suggesting their inclusion as covariates.) Group membership was the most significant predictor of scores on each satisfaction index. When age, year of graduation, 16:3 March 1987
TABLE 3. Comparison of the satisfaction of emergency physicians and internists
Satisfaction Index (Potential Range) Resources (0-78) Professional autonomy (0-18) Administrative autonomy (0-18) Patient relationships (0-30) Professional relationships (0-18) Status (0-24) *P < .01; t tests and Mann-Whitney U tests.
Emergency Physicians Mean % of Total ___ SD Possible 58.17 74.6 _+ 11,17 12.24 68,0 _+ 4.27 14.06 78.1 _+ 3.58 21,57 71.9 _+ 5,41 14.72 81.8 -+ 3.09 17,23 71.8 + 4.62
Internists _+ _+ + _+ _+ _+ _+
Mean SD 61,05 10.94 14.03" 3.61 14.39 3.82 23.98* 4.33 13.1 O* 3.87 20.52* 3.41
% of Total Possible 78.3 77.9 79.9 79.9 72.8 85.5
TABLE 4. Comparison of expectation of position change in next two years (excluding those planning to retire)*
Emergency Physicians (%) Internists (%) *P < .001.
and years in position were controlled, the relationships between group and satisfaction with patient relations and status remained significant, while those between group and satisfaction with professional relationships and with professional autonomy did not. As described previously, an additional element of satisfaction, satisfaction with pay, also was measured. Emergency physicians and internists each estimated that they could earn, on average, a p p r o x i m a t e l y $23,500 more in the idealized "best" position (difference between groups not significant). Finally, p r e d i c t i o n s of p o s i t i o n change were analyzed. Forty percent of emergency physicians and 60% of internists reported no intention to leave their current position within the next two years (P < .02). Excluding those considering retirement, the difference was even more striking (Table 4). One emergency physician and 11 internists reported considering retire16:3 March 1987
0% 38 (40.4) 47 (69.1)
Likelihood of Changing Positions 25% 50% 75% 23 12 10 (24.5) (12.8) (10.6) 9 12 (13.2) (17.6)
ment. Among the remaining, more than one-fifth of emergency physicians but no internists estimated that their chances of leaving their current positions were 75% or more. When age, year of graduation from medical school, and years in current position were controlled, the significant relationship between intention to change position and group remained. Notably, the vast majority of those considering changes apparently intend to remain within their chosen specialty. Seven emergency physicians (7.4%) described "best" positions in other than emergency settings. Just two internists (2.9%) described "best" positions in settings outside internal medicine. Regression analyses were used to relate likelihood of position change to elements of satisfaction, and emergency physicians' and internists' personal and practice characteristics. Among emergency physicians not considering retirement, likelihood was predicted Annals of Emergency Medicine
100% 11 (11.7)
Total 94
68
(in order of importance) by dissatisfaction with professional autonomy, lack of board certification, recency of graduation from medical school, and belief that m o n e t a r y compensation would be greater elsewhere (R square, 0.35; P < .01). Given the clustering of responses among internists, likelihood of change was dichotomized (no intention of leaving versus 25% to 50% chance). D i s s a t i s f a c t i o n w i t h resources was the sole predictor of change (R square, 0.29; P < .01). Variables such as p a t i e n t load; hours worked per week; years in current position; payment or practice type; physician's sex or age; and, for emergency physicians, hospital size, were not significant predictors of expectations of changing position a m o n g either emergency physicians or internists.
DISCUSSION This survey of emergency physicians and internists practicing in the 281/69
SATISFACTION WITH PRACTICES Murphy & Jacobson
Philadelphia area was designed to address three questions: With what aspects of their work are emergency physicians particularly satisfied and dissatisfied? What proportion of emergency physicians predict that they will change positions in the near future, and what elements of satisfaction or practice are related to those predictions? How do emergency physicians compare w i t h p h y s i c i a n s practicing in other specialties with regard to satisfaction and predicted position change? The findings can be s u m m a r i z e d with respect to these questions and can be compared with results of related research. Emergency physicians reported relatively high levels of satisfaction with all facets of practice measured. Average scores were more than two-thirds of possible scores, ranging from 68% to 82%. Variability in levels of satisfaction appeared to be related to the nature of emergency practice. The highest level was reported with respect to professional relationships. Emergency medicine requires physicians to work together, giving them ample opportunity to discuss cases both with their colleagues and with consultants. In many settings, emergency physicians also teach medical students and/or residents. T h e l o w e s t level of s a t i s f a c t i o n among emergency physicians was reported with regard to professional autonomy. The hospitals in which they practice are bureaucratic settings. Previous studies of prepaid group practices by Mechanicll, lz and othersl3,14 have suggested that bureaucratization is inversely related to physicians' satisfaction with autonomy. Thus, it is not surprising t h a t the e m e r g e n c y physicians were relatively dissatisfied with their abilities to have input into decisions affecting their practices, and with their abilities to initiate changes in the way they work. Twenty-two percent of emergency physicians reported that there was at least a 75% chance that they would change positions within the next year or two. This compares with Demlo's ~s report that 10% of physicians in large prepaid groups and just 6.6% of solo practitioners predicted a similar likelihood of changing positions. Excluding those planning to retire, no internists in our s a m p l e m a d e similar predictions. 70/282
In her survey of residency-trained emergency physicians, Anwar 6 noted that many emergency medicine positions were available and that emergency physicians had changed jobs relatively often in an effort to "find the perfect position." A review of advertisements for emergency physicians confirms that m a n y postions remain available. The variety of opportunities may encourage some emergency physicians to change jobs - - or at least to consider changing jobs - - relatively frequently. A n o t h e r possible e x p l a n a t i o n for the high level of interest in changing positions is the relative youth of the emergency medicine respondents (mean age = 40.0 _+ 9.8 years). Herman I6 and Tilson I7 found that younger physicians were more likely than their older counterparts to leave group practices. H e r m a n conjectured that the y o u n g e r d o c t o r s h a d n o t y e t reached a "stable career stage." It may be that as the emergency physicians in this study become older, they will express interest in changing positions less frequently. Intention to change positions was related to level of satisfaction w i t h professional autonomy, certification status, year of graduation from medical school, and a s s e s s m e n t of the monetary rewards related to an idealized position. As noted above, the issue of professional autonomy has been shown to be particularly salient for p h y s i c i a n s in group practices and other b u r e a u c r a t i c settings. H-14 It would have been surprising if emergency physicians were not sensitive to this aspect of their practices. Because y e a r of g r a d u a t i o n f r o m m e d i c a l school is highly correlated with age, and because younger emergency physicians still may be seeking certificat i o n status, t h e career stage phenomenon described by Herman 16 may explain the relationship between predicted position change and these variables. Tilson I7 also reported a significant positive association between dissatisfaction with pay and physician turnover. That association probably is especially strong a m o n g emergency physicians, for w h o m changing positions, rather t h a n e x p a n d i n g one's practice, is generally the only way in which to increase income. Comparing emergency physicians' responses w i t h those of internists from the same geographic area is instructive. It is important to note that Annals of Emergency Medicine
average levels of satisfaction were high for all physicians with respect to all aspects of practices measured. Nevertheless, even when age, year of graduation from medical school, and years in c u r r e n t p o s i t i o n were controlled, emergency physicians were significantly less satisfied with patient relationships and status than were their internal medicine counterparts. With respect to p a t i e n t relationships, emergency physicians are dedicated to treating emergently ill patients. But, they often find themselves caring for patients with nonurgent problems. Because internists expect to see patients with nonemergency conditions, they do not c o m m o n l y face the gap between expectations regarding patients and the realities of patient care with w h i c h emergency physicians must contend. In addition, internists have an opportunity to develop long-lasting relationships with their patients; the continuity of care that often makes for satisfying physicianpatient relationships is not usually an aspect of emergency medicine. The relative dissatisfaction of emergency physicians with status probably reflects the fact that emergency medicine is still a developing specialty. Anwar s found that emergency medicine residents chose their specialty despite the "normal patterns of influence" in m e d i c a l school. O n c e in practice, however, emergency physicians in this study were less satisfied with their standing in their professional communities than were their counterparts in m o r e traditional i n t e r n a l medicine practices. The results of this study m u s t be evaluated with respect to the limitations of the research design. The response rate was relatively low, 35% overall, and the study was conducted in a single g e o g r a p h i c area. Even t h o u g h respondents and nonrespondents were similar with respect to certification status, more broadly based studies are needed to test whether, or how, either the response rate or the selection of a single geographic area affected the generalizability of the findings. A larger s t u d y also could define e m e r g e n c y m e d i c i n e and i n t e r n a l medicine more precisely. For example, while we defined emergency medicine broadly by practice location, it is possible that the satisfaction of emergency-residency-trained or certified emergency physicians differs from that of 16:3 March 1987
the wider group of p h y s i c i a n s w i t h other training who practice in emergency settings. Likewise, physicians certified in internal medicine or who practice in traditional departments of internal m e d i c i n e m a y differ f r o m those certified in family practice or who practice in relatively n e w departments-of family practice. Our sample was too small to stratify in ways that would allow us to test these possibilities. In addition, it is not k n o w n w h a t proportion of physicians who predict that they will change positions actually will do so. A longitudinal study would be required to test w h e th e r elements of satisfaction are indeed related to position changes among emergency physicians. Such a study also could evaluate the effects of changes in the career stages of the (now relatively young) emergency physicians. Further research also should include additional variables, the potential importance of wh i ch is suggested by the present work. For example, the residency t r a i n i n g status of e m e r g e n c y physicians s h o u l d be i n c l u d e d as a variable that might help explain variations in physicians' satisfaction w i t h patient relations or status. Variables that could suggest the effect of variation in satisfaction on clinical practice also should be adde& Finally, the possibility that "satisfaction" should be redefined for emerging medical specialties such as emergency medicine should be considered in further research. The data collection ins t r u m e n t used addressed the issues traditionally i m p o r t a n t to physicians in the past. However, it did not include i t e m s r e l a t e d t o s a t i s f a c t i o n with flexibility of scheduling, with opportunities for mobility, or w i t h variety of clinical experience. It may be
16:3 March 1987
that the e m e r g e n c y physicians place different values on aspects of t h ei r careers t h a n do t h o se in o t h e r spec i a l ti e s , and t h a t t h e y are, by and large, well satisfied w i t h dimensions not measured in the present research. Despite its limitations, the research is important because it is the first syst e m a t i c effort t o a s s e s s the satisfact i o n of e m e r g e n c y p h y s i c i a n s w i t h their practices, and to compare levels of satisfaction with those of a set of th e i r colleagues. Moreover, satisfact i o n was e v a l u a t e d w i t h an instrum e n t whose reliability had been tested, and t h a t a l l o w e d d i m e n s i o n s of satisfaction to be measured. CONCLUSION It is reassuring to find that emergency physicians are relatively well satisfied w i t h their practices. But, there is room for improvement. If verified, educators and administrators m a y use these results to alter training and practice characteristics in ways that will further enhance career satisfaction a m o n g e m e r g e n c y p h y s i c i a n s . Increased satisfaction may, in turn, reduce turnover, and limit the n u m b e r of e m e r g e n c y physicians w h o "burn out" or b e c o m e alienated f r o m their chosen specialty. The authors thank Stephanie Abbuhl, MD~ GW Crooks, MD; Gladys S Fenichel, MD; C William Hanson, Jr, MD; and Alison McDonald, MD, for their helpful comments.
REFERENCES 1. Prybil L: Characteristics, career patterns and opinions of physicians who practice in large muhispecialty groups. Medical Group Management I974;21:22-26. 2. Ross A: A report on physician termination in group practice. Medical Group Management 1969;I6:I5-21.
Annals of Emergency Medicine
3. Linn L8, Brook RH, Clark VA: Physician and patient satisfaction as factors related to the organization of internal medicine group practices. Medical Care 1985;23:1171-1178. 4. Likert R: N e w Patterns of Management. New York, McGraw-Hill, 1961. 5. Anwar RAH: Trends in training: Focus on emergency medicine. Ann Emerg Med 1980;9: 60-71. 6. Anwar RAH: A longitudinal study of residency-trained emergency physicians. Ann Emerg Med 1983;12:20-24. 7. Dorland's Medical Directory: Philadelphia and Metropolitan Area. Philadelphia, Packard
Press, 1983. 8. American Medical Directory. Chicago, American Medical Association, 1984. 9. Lichtenstein R: Measuring the job satisfaction of physicians in organized settings. Medical Care 1984;22:56-66. 10. Kleinbaum DG, Kupper LL: Applied Regression Analysis and Other Multivariate Methods.
North Scituate, Massachusetts, Duxbury Press, 1978. 11. Mechanic D: General medical practice: Some comparisons between the work of primary care physicians in the United States and England and Wales. Medical Care 1972;10:402-420. 12. Mechanic D: The organization of medical practice and practice orientations among physicians in prepaid and non-prepaid primary care settings. Medical Care I975i13:189-204. I3. Ben-David J: The professional role of the physician in bureaucratized medicine: A study in role-conflict. Human Relations 1958;2: 255-274. 14. McElrath DC: Perspectives and participation of physicians in prepaid group practice. American Sociological Review 1961;26:596-607. 15. Demlo L: The relationship of physician practice patterns to organizationai and personal characteristics {Unpublished doctoral dissertation). New Haven, Connecticut, Yale University, 1975. 16. Herman JG, Dunham RB, Hulin CL: Organizational structure, demographic characteristics, and employee responses. Organizational Behavior and Human Performance
1975;I3:206-232. I7. Tilson HH: Stability of physician employment in Neighborhood Health Centers. Medical Care 1973;11:384-400.
283/71