Continuing medical education and primary ... - Wiley Online Library

Research papers

Continuing medical education and primary physicians' job stress, burnout and dissatisfaction Talma Kushnir,1 Avner Herman Cohen2,3 & Eli Kitai3

Objectives The association between opportunities for continuing medical education (CME) and primary physicians' job stress, burnout and job dissatisfaction has not been investigated. It was hypothesized that participation in CME activities and perceived opportunities at work for keeping up-to-date with medical and professional developments would be correlated negatively with job stress and burnout, and positively with job satisfaction.

faction, among family physicians. Among paediatricians, CME was associated negatively with burnout. Perceived opportunities for professional updating were associated negatively with burnout and (marginally) with stress among family physicians, and negatively with stress and positively with satisfaction among paediatricians. A third of the paediatricians and a quarter of the family physicians wanted to increase their involvement in CME.

Method 309 primary care physicians (183 family physicians and 126 paediatricians) employed in health maintenance organizations in Israel responded to a mailed questionnaire. The independent variables were the extent of engagement in CME activities and perceived opportunities at work for professional updating. The dependent variables were job stress, burnout and job satisfaction. Multiple regression analyses were employed.

Discussion In this cross-sectional study, causality cannot be established and the CME measures should be re®ned. However, the results are consistent with the study's hypotheses and suggest that opportunities for CME and professional updating may reduce physicians' job distress and dissatisfaction.

Results After controlling for age, sex and professional status, participation in CME activities was associated negatively with job stress and positively with job satis-

Introduction Continuing medical education (CME) has been broadly de®ned as àll ways by which physician learning and clinical practice may be altered by educational or persuasive means'.1 Systematic reviews of the CME literature1,2 have revealed studies focusing either on physician performance, or on patient or health care. There is a paucity of studies concerning CME and physicians' job stress and burnout. Stress, burnout and job dissatisfaction are important issues for health care professionals because they are considered signi®cant risks to health and well-being of 1

Occupational Health and Rehabilitation Institute, Ra'anana, Israel Pediatric Ambulatory Community Clinic, Petach Tikva, Israel 3 Department of Family Medicine, Tel Aviv University, Medical School, Tel Aviv, Israel 2

Correspondence: Dr Talma Kushnir, Occupational Health & Rehabilitation Institute, P.O.Box 3, Ra'anana, Israel 43100

430

Keywords *Burnout, professional; cross-sectional studies; *education, medical, continuing; Israel, epidemiology; job satisfaction; physicians; physicians, family; regression analysis; stress. Medical Education 2000;34:430±436

physicians,3,4 and are also associated with reduced quality of health care5 attrition6 and reduced commitment to the practice.7 These studies, as with most research on stress sources in health care, have been conducted among family physicians, a medical specialty considered to be intrinsically more stressful than others.8 It has been argued that much of medical training consists of learning to cope with limits of medical knowledge and that making serious clinical decisions based on con¯icting, incomplete and untimely data is routine.9,10 Indeed, insuf®cient knowledge base, inadequate training for fundamental tasks and having to face an increasingly informed public have been suggested as causal factors in physicians' stress.9±11 Concerns about maintaining personal or professional competence were also found to be a signi®cant stressor among consultant doctors.12 A hypothesis that learning opportunities may help prevent or reduce stress and dissatisfaction can be based speci®cally on such ®nd-

Ó Blackwell Science Ltd ME D I C A L ED U C A T I ON 2000;34:430±436

CME, stress and burnout

·

T Kushnir et al.

ings and on several theoretical models from the general literature on occupational stress. First, learning and training may improve professional self-ef®cacy, i.e. professionals' belief in their abilities to perform well in their work roles. Because stress and burnout were found to be linked to inability to achieve a sense of competence and success in one's work,13 self-ef®cacy is considered one of the most important anti-stress resources. A meta-analytic study14 has shown that self ef®cacy is also a signi®cant contributor to work motivation and job satisfaction. Family physicians' sense of clinical competence was found to be a consistent area of satisfaction.15 Therefore, factors in the individual or the workplace which improve capability, mastery and skill, such as opportunities for learning and further training, should reduce the potential for stress and burnout and increase job satisfaction. Second, according to the Job Demands±Control model,16 occupational stress is negatively associated with positive job features, including the requirement to learn new things on the job, to use intellectual skills and having opportunities to develop new skills. Such job characteristics are considered to be important coping resources which create challenge, rather than stress, and are therefore conditions of growth and regeneration.17 A third relevant approach is the `person-environment ®t model' (P-E ®t) which emphasizes the imbalance that occurs between individuals' needs and their job environments as a source for stress.18 A related theory of career choice suggests that people tend to choose a profession which ®ts their personality type.19 Those who choose medical sciences and medical services tend to be mostly of the investigative type, that is to possess scienti®c ability, to value science, and to be inclined to be critical, curious and precise. When the environment fails to match the individual's personal needs, an incongruence occurs which is a signi®cant source for stress and dissatisfaction. It can be argued therefore that many physicians would value keeping up-to-date with progress and developments in general medicine and speci®cally in their ®eld of specialization. This requires constant formal and informal learning. When such opportunities are lacking, stress and dissatisfaction can be expected. Based on the above considerations it was hypothesized that job stress, burnout and dissatisfaction would be inversely related to the extent of perceived opportunities at work for continuing medical education and for keeping up-dated with medical and professional developments.

431

Method Population

The study population included 385 primary care physicians employed in Israeli Health Maintenance Organizations. These included a random sample of 225 members of the Israeli Society of Family Physicians, and 160 members of the Israeli Society of Ambulatory Paediatrics, whose names and addresses were included in the mailing lists of these Societies. The two Societies involve about 2500 and 1600 members, respectively. Thus the sample represents an estimated 10% of the entire population, taking into account the fact that some of the members are not physicians. The two societies represent approximately 80% of primary family physicians and ambulatory paediatricians in Israel. A questionnaire and a selfaddressed and stamped envelope were sent to each of these physicians. Measures

The independent and dependent variables were assessed via a self-report questionnaire. 1 Independent variables CME activities One item assessed the extent to which physicians participated at work in formal CME activities, i.e. those opportunities afforded by the employer to physicians during work hours, at the expense of the employer: `To what extent do you attend scienti®c courses, seminars, conferences, and formal consultations?' Response scale ranged from 1 (a little) to 3 (a lot). Opportunities for professional updating The perceived amount of opportunities at work for updating was assessed by two questions: `Does your job provide you with opportunities for keeping up-to-date with developments in medicine?'; and `Does your job provide you with opportunities for keeping up-to-date in your medical specialty?' The 3-point response scale for both questions ranged from 1 (Much less than I would like) to 3 (As much as I like). Since the responses for the two questions were highly correlated (r 0á75, P < 0á001, for family physicians; and r 0á65, P < 0á001, for paediatricians), the scores for the two items were combined into one by averaging the separate responses. Demographic variables Sex, age and professional status. The latter included three categories for the family physicians (specialist, resident and general practitioner); and two categories for the paediatricians

Ó Blackwell Science Ltd ME D I C AL ED U C AT I ON 2000;34:430±436

432


·

T Kushnir et al.

(specialist and general practitioner). Israeli general practitioners are physicians who are not board certi®ed specialists, and may, or may not have had formal training after medical school. 2 Dependent variables Job stress The assessment of job stress in this study followed the strategy used by researchers of the Job Demands±Control (JDC) model of occupational stress.16 Job stress is de®ned as the ratio of job demands and resources available to the individual for coping with these demands. Because there were no previous empirical studies conducted in Israel which assessed speci®c job demands, we ®rst interviewed several family physicians and paediatricians, men and women, specialists, general practitioners and residents, and gathered information for compiling scales of negative job features, representing job demands, or stressors. The ®rst draft of the questionnaire was pre-tested on another sample of physicians, until two reliable scales of negative job features were achieved, one for paediatricians (Cronbach alpha 0á85, 25 items); and one for family physicians (Cronbach alpha 0á80, 23 items). Some of the items were identical for the two groups, such as `My relationship with my director is unpleasant'; `The job is routine and repetitive' and `My salary is too low'. A sample item for family physicians: `The status of family medicine is lower than the status of other medical specialties'. A sample item for paediatricians: Ì ®nd it dif®cult to concentrate when parents bring all their children and demand that they all be examined'. An additional scale assessing positive job features as an index of resources was also compiled. This 10 item scale included general items that are not speci®c to any medical specialty. A sample item: `My job enables me to do things I am good at'. The coef®cient of reliability (Cronbach alpha) was 0á77 for family physicians and 0á71 for paediatricians. (Scales can be obtained from the authors). As explained above, job stress was assessed by dividing the score on the negative features scale by the positive scale score. Professional burnout To assess the level of burnout we used a scale measuring ®ve aspects of energetic exhaustion20: emotional and physical exhaustion, tension, listlessness and cognitive weariness. The scale included 22 items with a 7-point response scale ranging from 1 (almost never) to 7 (almost always). The validity and reliability of this scale was discussed elsewhere.21 Sample items: Ì feel physically exhausted'; Ì feel restless'; Ì feel disorganized lately'. The alpha Cronbach for the present sample was 0á89.

Job satisfaction Using the global approach to the assessment of this variable, one question assessed job satisfaction: `Taking into account all possible aspects of your work, how satis®ed are you with your job?' The 20-point response scale ranged from 1 (maximum dissatisfaction) to 20 (maximum satisfaction). Such an item with a 9-point scale was found to correlate 0á56 with the Job Description Index (JDI) and with others measures of well-being, such as anxiety, somatic complaints and self-esteem.22 Further needs

An open question asked participants to indicate which activities at work they would like to increase. Procedures and data analysis

About a month after mailing the questionnaires we contacted all the physicians who had not returned them individually by telephone, emphasizing the importance of this study. Each was sent an additional questionnaire. About a month later we started analysing the data using the SAS package. Since the negative job features scales differed slightly between the family physicians and the paediatricians, the data was analysed separately for the two groups. Associations between variables were assessed by Pearson correlations. The independent contributions of the independent variables (extent of CME activities, opportunities for professional updating, professional status, age and sex) to the dependent variables (job stress, burnout and job satisfaction) were determined by a series of multiple regression analyses.

Results 183 of the family physicians (81á3%) and 126 (77á8%) of the paediatricians returned the questionnaires. Thus the ®nal sample included 309 primary physicians employed in health maintenance organizations. Table 1 presents the demographic characteristics of the study population. Educational needs

In response to an open-ended question about job activities they would like to increase, 47 (25á8%) family physicians and 38 (30á2%) paediatricians responded that they would like to participate more in CME activities. Nineteen (10á4%) of the family physicians and 8 (6á3%) of the paediatricians needed more professional consultations.



·

T Kushnir et al.

433

Table 1 Characteristics of the sample Variable

Paediatricians

Family physicians

Sample size Age (years) Sex (% females) Marital status (% married) Professional status (% specialists)

126 47á1(+8á3) 54á1 82á5 66á9

183 43á1(+6á9) 32á8 93á4 78á5

Table 2 presents the Pearson correlations between the dependent variables (job stress, burnout and dissatisfaction); and the independent variables (demographic variables and opportunities for CME and for professional updating). The CME and professional updating variables were associated signi®cantly, although modestly (r 0á23, P < 0á001) among the family physicians but not among the paediatricians. The two variables were related signi®cantly and in the expected direction to most dependent variables, in both groups, as follows: CME activities were associated signi®cantly with job stress, burnout and job satisfaction among the family physicians, but only with burnout among the paediatricians. Perceived opportunities for updating were associated with all dependent variables (marginally with job satisfaction) among both groups of physicians. The

pattern of associations was somewhat different in the two groups. Among the family physicians, CME had stronger associations with the dependent variables; among the paediatricians opportunities for updating had stronger associations. CME activities were associated signi®cantly with professional status among family physicians, with general practitioners participating less than specialists and residents. Opportunities for updating were associated signi®cantly with sex among paediatricians, with males reporting more opportunities. Sex was associated with job stress in both groups but in different ways. Among paediatricians, women were less stressed than men. Among family physicians, women had higher levels of burnout and dissatisfaction than male doctors. Age was signi®cantly and negatively associated with stress and burnout among family physicians, with older physicians reporting lower levels of stress and burnout. In order to assess the contribution of each of the independent variables to job stress, burnout and job satisfaction, above and beyond the effect of demographic variables (sex, age, and professional status), a series of multiple regression analyses were performed. Table 3 shows the results of these analyses. After controlling for the effects of demographic variables on the dependent variables, participation in CME activities was associated signi®cantly with job stress (negatively) and with job satisfaction (positively)

Table 2 Pearson correlations between independent and dependent variables in the two groups CME

Updating

Stress

Burnout

Paediatricians CME Updating Stress Burnout Job satisfaction Status Sex Age

0á16 )0á09 )0á23* 0á14 0á05 0á00 0á06

)0á36*** )0á21* 0á33*** 0á04 0á21* 0á00

0á50*** )0á60*** )0á20* )0á35*** )0á02

)0á44*** 0á08 )0á13 )0á13

0á08 0á11 0á09

Family physicians CME Updating Stress Burnout Job satisfaction Status Sex Age

0á23** )0á27*** )0á18* 0á27*** )0á16* 0á03 0á10

)0á19* )0á19** 0á14+ 0á08 )0á11 )0á07

0á44*** )0á62*** 0á14+ 0á29*** )0á17*

)0á43*** )0á06 0á17* )0á28***

)0á07 )0á24** 0á10

*P < 0á05; **P < 0á01; ***P < 0á001; + = 0á05 < P < 0á10. Ó Blackwell Science Ltd ME D I C AL ED U C AT I ON 2000;34:430±436

Job satisfaction

Status

Sex

0á45*** 0á05

)0á06

0á03 0á06

)0á22**

434


·

T Kushnir et al.

Job stress Predictor variables

Ba

Paediatricians CME Updating Status Sex Age

)0á05 )0á32** )0á03 )0á26* )0á02

Family physicians CME Updating Status Sex Age

)0á19* )0á14+ )0á01 )0á24** )0á19*

Burnout R2

Ba

0á18***

)0á23* )0á11 0á21+ )0á21+ )0á18+

0á18***

)0á10 )0á16* 0á10 )0á10 )0á23**

Job satisfaction R2

Ba

0á11**

0á07 0á30** 0á01 0á08 0á11

0á11***

0á24*** 0á06 0á01 )0á24*** 0á03

R2

Table 3 Multiple regression analyses of the contribution of CME and opportunities for professional updating to stress, burnout and dissatisfaction

0á08*

0á11***

a = Standardized beta. *P < 0á05; **P < 0á01; ***P < 0á001; + = 0á05 < P < 0á10.

among family physicians, and with burnout (negatively) among paediatricians. Perceived opportunities for professional updating were associated signi®cantly with job stress (negatively) and with job satisfaction (positively) among paediatricians; and with job stress (marginally) and burnout (negatively) among family physicians.

Discussion In the present study, perceived lack of opportunities for CME and for keeping up with professional developments were signi®cantly associated, albeit modestly, with job stress, burnout and job dissatisfaction, among primary care physicians. These associations were independent of the effects of several important demographic variables (age, sex and professional status) which were controlled for in this study. A sizable number of family physicians and paediatricians have speci®cally expressed a need to engage more in CME activities. One way of interpreting these cross-sectional ®ndings is to assume that there is a causal association between perceived opportunities for participation in various CME activities, and reduction of stress and burnout. Such an interpretation is consistent with the study's hypotheses. However, the reverse should also be considered, i.e. that high levels of physicians' occupational distress and dissatisfaction may reduce interest and participation in such activities, and/or lead to perceptions of limited availability of CME opportunities. Furthermore, it is possible that the CME±stress associations found here are spurious and that a third (or more) variable is associated with both CME participa-

tion and the stress variables. For example, depression may be associated with a perception of reduced CME opportunities (or disinterest in CME) on the one hand, and physician distress on the other hand. Since the present study was of a cross-sectional nature, it is impossible to reject any of these possibilities. At this stage it can be stated, however, that the ®ndings are consistent with the study's hypotheses, and that further longitudinal studies could clarify the nature of the associations found here. Although statistically signi®cant, the associations between CME participation and job stress factors were modest and much of the variance remained unexplained. Since participation was assessed with one global question and perceived opportunities with two speci®c questions, it could be that these measures did not tap a broader range of variability and that the true associations are higher. Further studies may re®ne these measures and improve their validity. CME and updating opportunities were associated signi®cantly among the family physicians. A possible interpretation is that participating in CME activities may increase the perception of having more opportunities for professional updating. Whether or not the association is causal can only be determined by further longitudinal studies. Since we have not assessed the contents of the CME activities and opportunities, it is impossible at this stage to speculate about the origins of the differences between the two groups. These can be related to differences in the medical specialties and/or to personal variables. Further studies may clarify such differences. The modest associations between the two independent variables suggest that the kind of CME



·

T Kushnir et al.

activities in which these physicians engage may not be highly effective in creating a sense of having suf®cient opportunities to develop professionally. Assuming a causal association between learning opportunities and reduction in job-related distress, several possible linking mechanisms can be suggested on the basis of prominent occupational stress models. First, learning and training may increase professional self-ef®cacy, i.e. the belief in one's ability to perform job demands skilfully (a well-known anti-stress resource). Second, CME may introduce new elements to the physician's job performance through new knowledge, improved skills and attitude change. These should combat the effects of boredom created by routinization and repetitiveness of the job ± well known stress-inducing and satisfaction-reducing features. Learning and training opportunities are important resources for selfrenewal. By refreshing and adding variety, complexity and interest to the job, it may be possible to reduce stressful boredom and professional stagnation. A signi®cant bonus is improved quality of care which is dependent on lifelong learning.23 It would be premature to suggest, on the basis of the ®ndings, ways of reducing distress and dissatisfaction through CME activities since our measures were nonspeci®c perceptions of CME opportunities at work. We have not assessed these variables directly and objectively and can therefore only speculate about their mediating role in the learning/stress-reduction relationships. Further studies are warranted in which such mediators are speci®cally and objectively operationalized to elucidate their role. Moreover, further studies may distinguish between types and contents of CME activities, and investigate environmental pressures promoting or impeding participation. Finally, another negative job aspect associated in the stress literature with lack of educational opportunities is professional isolation. It is often mentioned in conjunction with lack of professional interactions and lack of stimulation, especially among physicians in remote areas or solo practice arrangements.24 Professional isolation and lack of stimulation were found to be major sources of job stress, dissatisfaction and demoralization among physicians.15,25 Thus, the role of CME in preventing stressful professional isolation appears to be an important subject for further studies. In conclusion, physicians' job stress has been attributed to a variety of causes. In the present study we have focused on perceived lack of opportunities for learning and professional updating. We have attempted to highlight the potential role of continuing education in preventing stress and burnout and increasing satisfaction. The results are consistent with

435

the suggestion that among primary care physicians, perceived opportunities on the job for learning and for keeping up with developments in medicine and in one's speci®c ®eld of specialization, serve as resources for reducing job stress and burnout, and for increasing their job satisfaction. If such ®ndings are replicated and causality is established, one could recommend developing organizational measures to promote initiation, and facilitating participation in a variety of CME activities. Moreover, it has been suggested recently that there is consensus among health care professionals, that continuing professional development is essential in achieving high quality care for all patients.23 The present paper suggests that one of the ways in which continuing staff development could improve quality of care is by promoting physicians' well-being, through reduction of occupational stress and dissatisfaction.

References 1 2

3 4

5

6

7

8

9 10 11

12

13

Davis DA, Thomson MA, Oxman AD, Haynes RB. Evidence for the effectiveness of CME. JAMA 1992;268 (9):1111±7. Davis DA, Thomson MA, Oxman AD, Haynes RB. Changing physician performance. JAMA 1995;274 (9):700±5. Simpson LA, Grant L. Sources and magnitude of job stress among physicians. J Behav Med 1991;14:27±42. Sutherland VJ, Cooper CL. Job stress, satisfaction and mental health among general practitioners before and after introduction of new contract. BMJ 1992;304:1545±8. Groenwegen PP, Hutten JBF. Workload and job satisfaction among general practitioners: a review of the literature. Social Sci Med 1991;10:1111±9. Wine®eld HR, Anstey TJ. Job stress in general practice: Practitioner age, sex and attitudes as predictors. Fam Pract 1991;8 (2):140±4. Ben Sira Z. The plight of primary medical care: the problematics of committedness to the practice. Social Sci Med 1986;22:269±712. Swanson V, Power K, Simpson R. A comparison of stress and job satisfaction in male and female GPs and consultants. Stress Med 1996;12:17±26. Coles C. Medicine and stress. Med Educ 1994;28:3±4. Knox AB. Strengthening Adult and Continuing Education. San Fransisco: Jossey-Bass Publishers, 1993. Rout U, Cooper CL, Rout JK. Job stress among British general practitioners: Predictors of job dissatisfaction and mental ill-health. Stress Med 1996;12:144±66. Agius RM, Blenkin H, Deary IJ, Zealley HE, Wood RA. 1996 Survey of perceived stress and work demands of consultant doctors. Occup Environ Med 1996;53:217±24. Cherniss C. Role of professional self-ef®cacy in the etiology and amelioration of burnout. In: Schaufelli WB, Maslach C, Marek T. eds. Professional Burnout: Recent Developments in

Ó Blackwell Science Ltd ME D I C AL ED U C AT I ON 2000;34:430±436

436


14 1 15

16

17

18 19

·

T Kushnir et al.

Theory and Research, p 135±50. New York: Francis & Taylor, 1993. Sadri G, Robertson IT. Self-ef®cacy and work related behavior: A review and meta-analysis. Applied Psychology: An International Review 1993;139±52. Skolnik NS, Smith DR, Diamond J. Professional satisfaction and dissatisfaction of family physicians. J Family Practice 1993;37 (3):257±63. Karasek RA. Job demands, job decision latitude and mental strain: Implications for job redesign. Administrative Sci Quarterly 1979;24:285±308. Johnson JV. Control, collectivity and psychosocial work environment. In: Sauter S, Hurrell J, Cooper C. eds. Job Control and Worker Health. Chichester, England: John Wiley & Sons, 1989. Kahn RL. Work and Health. New York. John Wiley, 1981. Holland JL. Making vocational choices: A theory of vocational personalities and work environments. Englewood-Cliffs, NJ: Prentice Hall, 1985.

20 Shirom A. Burnout in work organizations. In: Cooper CL, Robertson I. eds. International Review of Industrial and Organizational Psychology. New York: John Wiley, 1989. 21 Kushnir T, Melamed S. The Gulf War and its impact on burnout and well-being of working civilians. Psychol Med 1992;22:987±95. 22 Meir E, Melamed S, Abu-Freha A. Vocational, avocational, and skill utilization congruences, and their relationship with well-being in two cultures. J Vocational Behavior 1990;36:153±65. 23 Field S. Continuing professional development in primary care. Med Educ 1998;32:564±6. 24 Lee MC, Chou MC. Job and life satisfaction among remote physicians in Taiwan. J Formos Med Assoc 1991;90 (7):681±7. 25 Handysides S. Building morale through personal development. BMJ 1994;308 (6921):114±6. Received 28 September 1998; editorial comments to authors 3 November 1998; accepted for publication 17 June 1999
