Physician career satisfaction has been reported to be declining during the last 30 years.1 Physician dissatisfaction has been significantly associated with decreased quality of care and patient dissatisfaction, and the importance of increasing physician satisfaction has been established in recent research (Bettes BA, Chalas E, Coleman VH, Schulkin J. Heavier workload, delivery, less personal control: impact of delivery on obstetrician/gynecologists’ career satisfaction. Am J Obstet Gynecol. In press).2,3 Potential factors that affect satisfaction level may be professional liability and the cost of professional liability insurance. More than other specialties, obstetrician-gynecologists have been adversely impacted by the current unavailability or high cost of professional liability insurance4 and have reported greater dissatisfaction with their careers.5 Obstetrician-gynecologists in the most acutely affected states are coping by curtailing or giving up gynecologic surgery, stopping high-risk obstetrics, dropping obstetric practice entirely, relocating, and in some cases, ceasing practice altogether.6
It is important to examine the link between overall career satisfaction and liability insurance concerns among obstetrician-gynecologists who continue obstetric practice. It cannot be denied that other factors exist, such as diminished reimbursement,7 interference by managed care with medical decision-making,8,9 fear of litigation,10 and defensive medicine.11 Interestingly, Kushnir et al12 found that participation in continuing medical education activities was positively correlated with career satisfaction for family physicians, and this also may be important to examine. However, liability insurance costs seem to be a primary component of the decision-making process in altering obstetric practice.4 This study assessed the impact of liability insurance costs, managed care pressures, and continuing medical education requirements on satisfaction with overall professional life and job-specific activities of obstetrician-gynecologists.
MATERIALS AND METHODS
Written questionnaires were mailed in the first week of June 2001 to 1,500 American College of Obstetricians and Gynecologists (ACOG) Fellows. In an effort to maximize the response rate, a second mailing was sent out in the last week of June 2001 as follow-up to the nonrespondents. Fellows were selected randomly by computer from those who had not been surveyed in the previous year. The purpose of random computerized selection was to avoid overburdening individual ACOG fellows with questionnaire surveys while maximizing the probability that the sample is representative of the general ACOG membership.
The career satisfaction component of the questionnaire was developed in consultation with Daniel Kahneman of Princeton University and adapted from Peter Warr’s13 conceptualization of job-specific satisfaction. Warr’s research points to 10 specific aspects of professional life that are highly indicative of the primary factors associated with overall career satisfaction.13,14 These 10 aspects of professional life are personal control, utilization of special skills and abilities, workload/demand, variety, feedback, income level, physical safety, supportive/effective leadership and supervision, interpersonal contact, and status/meaningfulness/prestige. Satisfaction level with each specific aspect was ranked, and scores were combined to form an overall career satisfaction construct for obstetrician-gynecologists.
Career satisfaction ratings were made on a scale from -5 to 5, where -5 represented “extremely unsatisfactory,” 0 represented “neutral,” and 5 represented “extremely satisfied” across 10 aspects of professional life.
Career pressures were divided into 3 categories of interest: liability insurance, managed care, and continuing medical education. Clinicians responded to 2 items in each category based on how closely the statement “corresponds with how you feel” on a 7-point scale from 0 (not at all) to 6 (greatly). With regard to the cost of liability insurance, clinicians rated the degree to which they felt liability insurance 1) “decreases the amount of time that I can spend with patients” and 2) “will affect the duration of my career in OB/GYN.” With regard to managed care, clinicians rated the degree to which they felt that it 3) “affects my medical decision making” and 4) “limits my diagnostic and treatment options.” And with regard to continuing medical education (CME), clinicians rated the degree to which they felt that 5) “obtaining CME is a financial burden” and 6) “keeping up with educational requirements is a time burden.”
The 7 job-specific professional activities in the survey were selected based on common activities of obstetrician-gynecologists and included grand rounds, surgery, vaginal delivery, planned cesarean delivery, office visits, in-hospital/on-call, and rounds. Respondents were asked how frequently they perform each activity and then to rate their level of satisfaction with each activity on a scale from 1 to 4, where 1 represented “not at all satisfied” and 4 represented “very satisfied.” Satisfaction scores were computed for each of the 7 professional activities.
The total number of hours clinicians work per week was computed by summing the number of hours clinicians reported for office hours, in-hospital hours, rounds hours, administrative hours, and secondary hours. Total number of work hours per week was examined across the career pressures groups to investigate the relationship between work hours and perceived impact level of liability insurance costs, managed care, and continuing medical education.
Data were analyzed by using a personal computer-based version of SPSS 10.0 (SPSS Inc, Chicago, IL). Differences in ratings on the dependent measures (job-specific satisfaction ratings and overall career satisfaction) were examined with multivariate analysis of variance. For those analyses for which there was a significant multivariate effect, univariate F tests were conducted to determine which, if any, satisfaction ratings were significantly different from each other. Analysis of covariance was used to examine effects of a physician’s sex (with age as a covariate) on satisfaction ratings for each of the professional activities so that all the respondents who engage in each activity could be included in the analyses. Age was included in these analyses as a covariate because of the possibility that the relationship between the independent and dependent variables varies as a function of age, as has been observed in similar measures. Any significant main effects reported are independent of age. Descriptive statistics were computed for frequency of professional activities and hours worked per week.
Three groups were defined according to level of concern for each of the 6 career pressures items: 1) a “low-impact” group composed of those whose ratings were 0 (not at all) or 1; 2) a “moderate-impact” group (ratings of 2 through 4); and 3) a “high-impact” group whose ratings were 5 or 6 (greatly). The distribution of respondents for each of the 18 groups is presented in Table 1.
To examine whether overall career satisfaction varied according to the extent of exposure to liability insurance concerns, 2 groups were defined based on reported practice type. The practice types were combined according to whether they involved high or low exposure to legal liability. Health maintenance organizations, military settings, and “other” (primarily Indian Health Service) were combined into the low-exposure group (n = 99, 12.1%). Solo, multispecialty, university, and group practices were combined into the high-exposure group (n = 719; 86.8%). Analysis of variance was used to compare the liability exposure groups on satisfaction scores.
Finally, analysis of variance was used to examine differences between the career pressures groups on overall career satisfaction scores and on total hours worked per week. Post hoc comparisons between the 3 career pressures groups were conducted with t tests in cases for which a significant effect was identified. Percentages were compared with χ2 tests. Significance was evaluated at alpha < .05 and confidence intervals of 95%.
A total of 842 clinicians returned questionnaires of 1,500 mailed, yielding a response rate of 56%. This response rate is within the range of similar published studies of questionnaire mailings to physicians.15,16 Eleven respondents were omitted because they were retired from active practice, and 3 respondents were omitted because of substantial missing data (ie, none had provided ratings of satisfaction or social variables), resulting in a final sample of 828 clinicians. The mean age of the sample was 46.52 (standard deviation 10.32) years, which is comparable with the mean age of ACOG Fellows (49.74 years). The sample consisted of more men (57.6%) than women (42.4%), which is again consistent with ACOG membership of 40% women and 60% men. The mean years in practice, as computed by years since residency was completed, were 14.78 (standard deviation 10.37). The majority belonged to a group practice (46.2%) or a solo practice (21.3%). The majority of respondents practiced both obstetrics and gynecology (78.6%). Nearly all respondents attended grand rounds (99.5%), performed surgery (95.9%), performed office visits (98.6%), or performed hospital rounds (93.5%); 83.2% of respondents performed deliveries, whether planned cesarean deliveries or vaginal deliveries, and 88.9% performed on-call/in-hospital hours. Table 1 presents demographic information for each of the impact groups.
The career pressure with which clinicians were most concerned was the extent to which the cost of liability insurance “will affect the duration of my career in OB/GYN.” As can be seen in Table 1, the majority of male and female respondents are in the high-impact group for this item. The mean rating for this item was significantly higher than the ratings for items involving managed care and continuing medical education (all at P < .001) and was the only item for which the majority of respondents were in the high-impact group. In contrast to clinicians’ concerns about the cost of liability insurance affecting the duration of their careers in obstetrics and gynecology, they appear to be less concerned about the impact of this career pressure with regard to its effect on “the amount of time that I can spend with patients.” More clinicians were in the low-impact group for this item than the moderate- or high-impact groups. It should be noted that 10 respondents had missing data on the item used to define exposure to legal liability.
Clinicians were likely to have some concerns that managed care “limits my diagnostic and treatment options,” with the majority of respondents in the moderate-impact group (P < .001 for low-versus-moderate and high-versus-moderate comparisons). Nearly half of clinicians were in the moderate-impact group for the item that asked the degree to which managed care “affects medical decision-making” (P < .001 for low-versus-moderate and high-versus moderate-comparisons).
Clinicians were slightly-to-moderately concerned that obtaining CME was a financial burden (P < .001 for each comparison between groups) and that keeping up with educational requirements was a time burden (P < .001 for each comparison).
The satisfaction ratings for the 10 aspects of professional life were submitted to confirmatory principle components factor analysis.17 One factor, “satisfaction with features of professional life,” was extracted, with all aspects having loadings of 0.613 or higher. This factor accounted for 48.56% of the variance in this set of variables. In light of the evidence from this factor analysis that all 10 aspects were highly correlated (by using Pearson correlations), a central construct representing overall career satisfaction was formed that was indicative of physicians’ satisfaction with their professional lives overall.
The overall career satisfaction rating had a range of -50 to 50, and a mean of 15.15 (standard deviation 17.72). A mean of 0 would have indicated a neutral rating across all aspects of professional life; therefore, it can be concluded that, as a group, respondents are generally satisfied with their professional lives overall.
The relationship between level of concern about each of the career pressures and career satisfaction is especially evident in these analyses (Table 2). For each career pressure item, satisfaction scores differed significantly among each of the 3 impact groups (P < .001 or better). For example, for the statement that liability insurance costs “will affect the duration of my career in OB/GYN,” respondents in the high-impact group reported a satisfaction score of 10.45, the moderate-impact group reported a score of 18.85, and the low-impact group reported a score of 22.44 (on a scale of -50 to 50). Indeed, there is an increase in satisfaction as the level of impact career pressures has on professional life decreases, regardless of pressure type. Clinicians who were extremely pressured by the cost of CME provided the lowest satisfaction ratings (5.46, at least P < .01 for each group), and all of the mean satisfaction ratings of clinicians who reported being greatly impacted by career pressures were well below the overall group mean on satisfaction (all at P < .001).
The clinicians who were extremely concerned about the impact of professional liability costs on the duration of their career represent approximately half of the sample, although this group does not differ from those who reported less concern on age, sex distribution, years since residency completed, or marital status. The single background variable on which this group differed was in the type of practice in which they worked (P < .001). Clinicians who were greatly concerned about the effect professional liability might have on the duration of their careers were far more likely to work in solo practices, university settings, and partnerships and less likely to work in health maintenance organizations and the military.
This effect can be seen in Table 3, which divides the sample into high- and low-exposure to liability groups based on practice type. These groups differ significantly (P < .01) in their ratings of each of the managed care and CME career pressures items, except the cost burden of CME.
There were several significant differences in the activities the respondents engaged in according to their responses to the career pressures items. For liability insurance costs, the high-impact group for the statement that liability insurance cost “will affect the duration of my career in OB/GYN” were more likely to perform surgery (P < .001), planned cesarean deliveries (P < .005), and hospital rounds (P < .004) than those in the low-impact group. A similar pattern was found for the statement that liability insurance cost “decreases the amount of time that I can spend with patients.” Those who perceived this factor as having a great impact on their careers were more likely to be doing surgery than those who felt it had a low impact (P < .036).
There were no significant differences between the high-impact group and the low-impact group defined by the statements that managed care “affects my medical decision making” and “limits medical decision options.” Clinicians in the high-impact group were equally likely to engage in the 7 professional activities as those in the low-impact group.
There were 2 significant effects for the high-impact group corresponding to the statement, “keeping up with educational requirements is a time burden.” Those who thought that the time burden had a great effect were more likely to do on-call/in-hospital hours (P < .01) and rounds (P < .027) than those who perceived the time commitment as having little effect.
There were no differences in impact group membership according to whether the respondents perform deliveries. Too few respondents did not perform the remaining activities to justify comparisons in these analyses.
The percentage of respondents at each satisfaction level for each of the 7 professional activities is listed in Table 4. Respondents were most satisfied with surgery and deliveries (planned Cesarean and vaginal) and least satisfied with on-call/in-hospital time. The difference in satisfaction between the most and least satisfying activities was significant (P < .01).
The overall career satisfaction score was significantly correlated (P < .001 in all cases) with all 7 of the ratings of job-specific satisfaction with individual professional activities (eg, surgery). These correlations ranged from r = 0.22 for planned cesarean deliveries to r = 0.38 for on-call/in-hospital hours and suggest that clinicians who have a greater sense of job-specific satisfaction are generally less concerned about the impact of liability insurance, managed care, and CME requirements and have greater satisfaction with their overall careers.
The mean number of hours respondents reported working per week was 67.05 (standard deviation 22.46). Results across impact groups are presented in Table 1. There were significant differences among the impact groups on the total number of hours worked. For liability concerns, the difference between the low-impact and high-impact groups (P < .001) and the moderate-impact and high-impact groups (P < .027) on number of hours worked was significant. However, the difference between the low-impact and moderate-impact groups was not significant. Again, the effect is linear, with number of hours worked per week increasing as the level of impact increases. In two instances, the groups did not differ significantly on total number of hours worked, those defined by “managed care affects my medical decision-making” and “obtaining CME is a financial burden.” For each of the other groups, those who felt that the career pressure had a great impact worked significantly more total hours than those who felt the pressure had low impact.
The overall career satisfaction scores, as derived from the satisfaction ratings of 10 aspects of professional life, were also correlated with total hours worked (r = −0.15; P < .001). The more hours clinicians in this sample work, the lower their scores on career satisfaction.
It should be noted that there is potential for error in any self-report task. However, because of the incorporation of several measures of time worked and the high intercorrelation of corresponding items with the work hours measures, there is some confidence that the measure of hours worked is a fair representation of the relative distribution of the hours worked by clinicians in this sample.
Most obstetrician-gynecologists in this study are greatly impacted by the pressures associated with professional liability and the cost of liability insurance. Specifically, there was a strong belief that liability concerns will shorten the duration of their careers. Clinicians who were most highly pressured by this specific issue were working a greater number of hours than those who considered liability a low-pressure concern. This group was also more likely to perform surgery, planned cesarean deliveries, and hospital rounds. Additionally, results suggest that obstetrician-gynecologists who report high levels of professional pressure are more likely to work in solo practices, university settings, and partnerships, where there is higher exposure to liability insurance. Managed care is a moderate pressure for the majority of obstetrician-gynecologists, and the decreased level of impact associated with this pressure corresponds with fewer hours spent working.
Continuing medical education interest and obstetrician-gynecologist satisfaction level were found to be positively correlated. Respondents who had a greater interest in CME (ie, low-impact group) had a significantly higher satisfaction rating than those who consider it a burden (ie, high-impact group). An important finding was that among obstetrician-gynecologists who consider CME a burden, overall career satisfaction was lowest compared with all other groups, suggesting that less interest in CME is strongly associated with career dissatisfaction.
Level of professional pressure has an inverse relationship with clinicians’ satisfaction with various professional activities and their overall satisfaction with their careers. As obstetrician-gynecologists’ satisfaction ratings increase, their level of concern with career pressures decreases. Job-specific satisfaction and overall career satisfaction were found to be highly correlated, indicating that satisfaction in the particular activities chosen to represent professional activities is associated with overall career satisfaction.
Although the results of this study indicate several important trends, there are a few limitations of this study that should be noted. First, it is unclear whether the results of this study are applicable to medical professionals other than obstetrician-gynecologists. Second, only 3 areas of career pressures were examined in this study and may not completely estimate the aggregate career pressure experienced by obstetrician-gynecologists. Third, social desirability is a concern with any self-report measure, although steps were taken to decrease this bias by informing respondents of their anonymity.
In conclusion, the implications of these findings are that obstetrician-gynecologists who report working significantly more hours are greatly pressured by liability concerns and are at a higher risk of career dissatisfaction and professional burnout. The importance of increasing physician satisfaction level has been stated in recent medical literature (Bettes BA, et al. Am J Obstet Gynecol. In press)3,5 and is reiterated here. Identifying ways to increase career satisfaction levels for our obstetrician-gynecologists is imperative not only to the physicians, but also to the patients who are in their care. The current study adds to our knowledge base about the detrimental roles professional liability and the high cost of liability insurance play in determining the career satisfaction of obstetrician-gynecologists, and underscores liability insurance concerns as potentially shortening the duration of their careers. In addition, interest in CME is identified as a protective factor for obstetrician-gynecologist career satisfaction, and accentuating the importance of life-long learning may be helpful in guarding against dissatisfaction.
1. Williams ES, Skinner AC. Outcomes of physician job satisfaction: a narrative review, implications, and directions for future research [review]. Health Care Manage Rev 2003;28:119–39.
2. Williams ES, Konrad TR, Linzer M, McMurray J, Pathman DE, Gerrity M, et al. Physician, practice, and patient characteristics related to primary care physician physical and mental health: results from the Physician Worklife Study. Health Serv Res 2002;37:121–43.
3. Kravitz RL, Leigh JP, Samuels SJ, Schembri M, Gilbert WM. Tracking career satisfaction and perceptions of quality among US obstetricians and gynecologists. Obstet Gynecol 2003;102:463–70.
4. Mello MM, Studdert DM, Brennan TA. The new medical malpractice crisis. N Engl J Med 2003;348:2281–4.
5. Leigh JP, Kravitz RL, Schembri M, Samuels SJ, Mobley S. Physician career satisfaction across specialties. Arch Intern Med 2002;162:1577–84.
6. Williams ES, Konrad TR, Scheckler WE, Pathman DE, Linzer M, McMurray JE, et al. Understanding physicians’ intentions to withdraw from practice: the role of job satisfaction, job stress, mental and physical health. Health Care Manage Rev 2001;26:7–19.
7. Landon BE, Reschovsky J, Blumenthal D. Changes in career satisfaction among primary care and specialist physicians, 1997–2001. JAMA 2003;289:442–9.
8. Stoddard JJ, Hargraves JL, Reed M, Vratil A. Managed care, professional autonomy and income. J Gen Intern Med 2001;16:675–84.
9. Landon BE, Aseltine R, Shaul JA, Miller Y, Auerbach BA, Cleary PD. Evolving dissatisfaction among primary care physicians. Am J Manag Care 2002;8:890–901.
10. Coping with the stress of malpractice litigation. ACOG Committee opinion 236. Washington, DC: American College of Obstetricians and Gynecologists; 2000.
11. Benbassat J, Pilpel D, Schor R. Physicians’ attitudes toward litigation and defensive practice: development of a scale. Behav Med 2001;27:52–60.
12. Kushnir T, Cohen AH, Kitai E. Continuing medical education and primary physicians’ job stress, burnout and dissatisfaction. Med Educ 2000;34:430–6.
13. Warr P. Well-being and the workplace. In: Kahneman D, Diener E, Schwarz N, editors. Well-being: the foundations of hedonic psychology. New York (NY): Russell Sage Foundation; 1999. p. 392–412.
14. Kahneman D, Diener E, Schwarz N, editors. Well-being: the foundations of hedonic psychology. New York (NY): Russell Sage Foundation; 1999.
15. Farquhar CM, Kofa E, Power ML, Zinberg S, Schulkin J. Clinical practice guidelines as educational tools for obstetrician–gynecologists. J Reprod Med 2002;47:897–902.
16. Gerrity MS, Williams JW, Dietrick AJ, Olson AL. Identifying physicians likely to benefit from depression education: a challenge for health care organizations. Med Care 2001;39:856–66.
17. Thurstone LL. Multiple-factor analysis: a development and expansion of the vectors of the mind. Chicago (IL): The University of Chicago Press; 1947.