Predictors of Professional and Personal Satisfaction With a Career in Psychiatry

Transcription

1 Original Research Predictors of Professional and Personal Satisfaction With a Career in Psychiatry Paul E Garfinkel, MD, FRCPC 1, R Michael Bagby, PhD, CPsych 2, Deborah R Schuller, MD, FRCPC 3, Susan E Dickens, MA 4, Fiona S Schulte, MA 5 Background: Many factors, including personal experience and personality traits, contribute to the emotional difficulties that psychiatrists experience in their work. The nature of the work itself also plays a significant role. Objective: To determine those personal and characteristics that predict with the practice of psychiatry. Method: We mailed a questionnaire that included items pertaining to aspects of personal and life to the entire population of psychiatrists in Ontario (N = 1574). Results: Of the 1574, 52% (n = 802) responded. We conducted a series of regression analyses to determine factors related to career or regret. A belief in the intrinsic value of psychiatry, a low perceived degree of emotional burden from patients, financial success, and with psychotherapeutic work emerged consistently as significant predictors. A subsequent discriminant function analysis indicated that all 4 of these variables accurately predicted those psychiatrists with extreme or dis with work. Conclusions: These results reveal several variables associated with career in the practice of psychiatry that might be useful to discuss with residents who are beginning their careers. (Can J Psychiatry 2005;50: ) Information on author affiliations appears at the end of the article. Clinical Implications Several variables accurately predicted psychiatrists who experience extreme or dis with work. Limitations The study was limited to psychiatrists in Ontario. The evaluation was cross-sectional. We collected self-report data only. Key Words: career, practice characteristics, emotional vulnerability Psychiatrists are known to be at risk for depression, anxiety, and burnout (1 4). The perceived stress associated tional burden of treating very ill patients, the relative isolation work itself, however, clearly contributes further. The emo- with the practice of psychiatry is a factor in dissuading medical students of different cultural backgrounds from choosing of the work, and an inability to balance personal and lives all play an important role. Recent studies from the psychiatry as a specialty (5). Many factors including experience and personality traits contribute to the emotional difficulties that psychiatrists experience (6). The nature of the tice patterns, reimbursement, managed care, and US have also related career dis to changing prac- increased Can J Psychiatry, Vol 50, No 6, May

2 The Canadian Journal of Psychiatry Original Research competition from nonphysician mental health s (7 9). Nevertheless, psychiatrists also experience tremendous in caring for others. There is now an expanded therapeutic armamentarium that includes evidence-based psychotherapies and newly developed medications, which affords a greater probability of treatment success. To possess the ability and perseverance necessary to successfully treat patients in serious emotional distress is for many psychiatrists intrinsically rewarding. Recently, using data from a survey of over 600 physicians, Sturm found that only 21.2% of psychiatrists in his sample reported being dissatisfied with their careers (10). Similar rates were found among a cohort of Dutch psychiatrists (11). Dis in this cohort was predicted by feeling poorly managed and resourced, feeling devalued, and feeling a lack of intellectual stimulation, as well as by poor communicative responsiveness and social support from colleagues. An earlier report on another cohort of psychiatrists that the first author has been following for more than 20 years (12) found that personal and external measures of career success were related to personality traits and to attributes associated with the psychiatrists personal and lives. These findings, however, were based on a small sample of psychiatrists who graduated from their residency programs in the 1970s and were followed longitudinally, limiting the generalizability of the results. In the current study, we used data from a survey of a large sample of Ontario psychiatrists to examine factors related to psychiatrists perception of and personal with their careers. In contrast to our previous study (12), which was also limited by a small sample size, we were able to test a large number of predictor variables that the previous literature suggested as potential predictors of career. We also used 4 different questions to assess and personal, compared with a single outcome variable, which we believe enhances the generalizability of our results. Method Subjects, Procedure, and Questionnaire Design In the spring of 1999, as part of a study of Ontario psychiatrists practice characteristics, we mailed a 12-page survey to the entire population of licensed psychiatrists in Ontario (N = 1547). The questionnaires were accompanied by a cover letter explaining the nature of the study and an anonymous form for responding. Four months later, we sent a follow-up mailing to those who did not respond to the initial mailing. Of the potential respondents, 802 (52%) completed and returned their questionnaire in response to these 2 mailings. This return rate 334 is comparable to similar studies that have been conducted with physicians in other specialties (13,14). The mean age of the overall sample was 50.1 years, SD 10.4; 552 (68.8%) were men; and 250 (31.2%) were women. The modal year of graduation from psychiatric residency was 1986, while the median year was The earliest year of graduation from psychiatric residency was 1951, and the latest was The questionnaire included items designed to assess a psychiatrist s personal and life and the interface between each (see Appendix). As well, the following questions were specifically designed to address in different ways the respondents sense of and personal with their career as psychiatrists: To what degree do you have regrets about choosing psychiatry as a profession? To what degree have the expectations that led you to choose psychiatry as a specialty at the beginning of your career been met? If you were able to choose again, would you choose psychiatry as a specialty? Overall, how satisfied are you with your career? Responses were recorded on a Likert scale ranging from 1 to 5, where 1 indicated the more negative response and 5 reflected a highly positive response. The questions were generated by a working group that included 2 psychiatrists with extensive knowledge and experience of psychiatrists and academic responsibilities within the private and public sectors. This questionnaire was used in our earlier study (12) and received approval for use by the Research Ethics Board of the University of Toronto and the Centre for Addiction and Mental Health. Statistical Analysis We performed 4 regression analyses, 1 for each of the 4 questions assessing and personal (which served as criterion variables) together with the 23 questions listed in the Appendix (which served as predictor variables). All predictor variables were entered according to a stepwise method, with entry level set at P < Despite the large number of predictor variables, the subject-to-variable ratio was 33 to 1, providing more than adequate statistical power. We selected the items that formed the criterion and predictor variables from both an extensive review of the literature and the findings of our previous longitudinal study (12) together with the hypotheses emanating from it. For each regression analysis, we report unadjusted R 2 values (see the Results section) as well as the standardized beta coefficients and partial and part correlations (reported in Tables 1 to 4). These statistics provide an estimation of the association between identified predictor and criterion variables directly and also control for the association between the other predictor variables and the criterion variable. Can J Psychiatry, Vol 50, No 6, May 2005

3 Predictors of Professional and Personal Satisfaction With a Career in Psychiatry Table 1 Regression analysis for variables found to predict regrets with choosing psychiatry as a career choice Final model Step Predictors a R 2 F df R 2 chg F chg df B SE Partial r Part r 1 Intrinsic value of work 2 Degree of emotional burden 3 Academic 4 Financial success 5 Availability of contact 6 Psychotherapy 7 Community ** 1, ** 1, ** 1, ** 1, ** 1, * 1, * 1, Overall model ** 7, 678 a Only variables that emerged as significant predictors of regret with choosing psychiatry as a career choice are included. **P < 0.01; *P < 0.05; SE = standard error Results Table 5 displays the means and SDs for the survey questions (the predictor and criterion variables used to evaluate overall level of career in the regression analyses). Tables 1 to 4 display the results from the regression analyses. Overall, 15 of the 23 predictor variables that were entered into the model emerged as significant predictors for at least 1 of the 4 criterion variables assessing different aspects of and personal (see Table 6). On average, the significantly predicting variables accounted for 35% of the variance. Four predictor variables belief in the intrinsic value of psychiatric care, perceived degree of emotional burden, perceived financial success, and derived from psychotherapeutic work emerged consistently as the most reliable predictors. For 3 of the 4 criterion variables, the perceived intrinsic value of psychiatry proved to be the most powerful predictor. To determine which of these significantly predicting variables distinguished those individual psychiatrists least and most satisfied with their careers, we divided the sample into 2 groups and performed a discriminant function analysis. To create these groups, we first computed a composite score by summing together the 4 criterion variables ( regrets about choosing psychiatry, expectations not been met, choose psychiatry again, and satisfied with career ) used in the above-described regression analyses. These scores could range from 4 to 20. The highly dissatisfied group of psychiatrists comprised those individuals who scored in the lower quartile of this composite variable and who endorsed all 4 of the criterion questions on the lower portion of the Likert scale. The highly satisfied group of psychiatrists comprised those individuals who scored in the upper quartile and who endorsed all 4 criterion variables items only on the upper portion of the Likert scale. The highly dissatisfied group of psychiatrists consisted of 61 respondents (7%) and the highly satisfied group consisted of 232 respondents (28%). The mean score on the composite scale for the highly dissatisfied group was 8.80, SD 2.30; for the highly satisfied group, it was 19.57, SD These means were significantly different from one another (t 291 = 64.28, P < 0.001, Cohen s d [measure of effect size] = 6.42). For the discriminant function analysis (DFA), these 2 groups of psychiatrists served as the criterion variable, while the 15 significantly predicting variables from the 4 previous regression equations served as the predictor variables. We entered these variables as a block, and all proved to be significant predictors (Wilks = 0.42; 2 = , df 15; P < 0.001) that correctly classified 94% of the psychiatrists highly dissatisfied with their career and 91% of those highly satisfied with their careers. Table 7 presents the standardized canonical discriminant function coefficients for these 15 variables in descending order of their predictive magnitude. As with the regression analyses, the perceived intrinsic value of psychiatric care proved to be the most powerful predictor, followed by Can J Psychiatry, Vol 50, No 6, May

4 The Canadian Journal of Psychiatry Original Research Table 2 Regression analysis for variables found to predict expectations of career in psychiatry Final model Step Predictors a R 2 F df R 2 chg F chg df B SE Partial r Part r 1 Intrinsic value of work 2 Degree of emotional burden 3 Professional peer 4 Financial success 5 Quality of contact 6 Other medical profession respect 7 Balance between and private life 8 Psychotherapy 9 Time for career development 10 Community involvement 11 Degree of isolation ** 1, ** 1, ** 1, ** 1, ** 1, ** 1, ** 1, ** 1, * 1, * 1, * 1, Overall model ** 11, 673 a Only variables that emerged as significant predictors of expectations of a career in psychiatry are included. **P < 0.01; *P < 0.05; SE = standard error quality of contact, with psychotherapy, and degree of isolation. Discussion In this study we used 4 different, but related, questions to assess job and level of regret in the choice of psychiatry. Four predictor variables perceived intrinsic value of psychiatric care, degree of perceived emotional burden, perceived financial success, and derived from psychotherapeutic work emerged consistently as the most reliable predictors. In the 4 regression analyses performed, the perceived intrinsic value of psychiatry proved to be the most powerful predictor in 3 of the 4 equations. When we divided 336 the sample into high and low or regret groups, the variables that emerged in a discriminant function analysis correctly classified 94% of the psychiatrists highly dissatisfied with their career and 91% of those highly satisfied with their careers. Again, the perceived intrinsic value of psychiatry proved to be the most powerful of the significant predictors for this analysis. In contrast to previous studies conducted primarily in the US, Ontario psychiatrists have far less regret about their career selection than do their US colleagues, allied health s in general, and other medical specialists in Canada. A recent study of psychiatrists in Texas, for example, found that Can J Psychiatry, Vol 50, No 6, May 2005

5 Predictors of Professional and Personal Satisfaction With a Career in Psychiatry Table 3 Regression analysis for variables found to predict choose psychiatry as a specialty again Final model Step Predictors a R 2 F df R 2 chg F chg df B SE Partial r Part r 1 Intrinsic value of work 2 Degree of emotional burden , ** 1, Financial success ** 1, Availability of contact 5 Professional peer 6 Psychotherapy 7 Time for career development Overall model ** 7, 676 a Only variables that emerged as significant predictors of choose psychiatry as a specialty again are included. **P < 0.01; *P < 0.05; SE = standard error ** 1, ** 1, * 1, * 1, Table 4 Regression analysis for variables found to predict overall with career Final model Step Predictors a R 2 F df R 2 chg F chg df B SE Partial r Part r 1 Professional peer 2 Balance between and private life 3 Appreciation from patients 4 Degree of emotional burden 5 Intrinsic value of work ** 1, ** 1, ** 1, ** 1, ** 1, Financial success ** 1, Time for career development 8 Degree of isolation 9 Psychotherapy ** 1, ** 1, ** 1, Overall model ** 9, 677 a Only variables that emerged as significant predictors of overall with career are included. **P < 0.01; SE = standard error Can J Psychiatry, Vol 50, No 6, May

6 The Canadian Journal of Psychiatry Original Research almost one-quarter were considering leaving the field (9). Similarly, 70% of respondents in a survey of Manhattan psychiatrists indicated that they would not recommend a career in private practice (8). Earlier surveys of physicians and nurses found comparable rates: between 20% and 50% indicated that they would not enter the field again if given the choice (15). In a recent survey of more than 2300 Ontario physicians, almost 41% reported being either dissatisfied or very dissatisfied with current medical practice (16). Almost one-half (46%) reported that they find their current medical practice to be very or extremely stressful. Sturm (10) described data from the Community Tracking Study physician survey, which involved over 600 psychiatrists, compared with over 4000 other medical and surgical specialists and over 7000 primary care doctors. Over 21% of psychiatrists reported being very or somewhat dissatisfied with their careers, a higher figure than for other specialists and general practitioners, owing to an age-cohort effect. An earlier study of Canadian physicians found that both men and women had high levels of both stress and job (17). Although only about 10% of the Ontario psychiatrists sampled in this study expressed extreme dis with having selected psychiatry as their profession, it is worth exploring what may be contributing to this small group s dis. It is also worth exploring why most in this sample find their careers satisfying. Such exploration may offer some explanation for the higher overall rates of dis found in the studies cited above. In the current sample, 2% of the psychiatrists rated themselves as moderately or strongly not believing in the intrinsic value of psychiatry. Presumably, people who enter medicine and any of the helping professions have a strong desire to care for and offer treatment to others. This particular prosocial attitude may explain why the perceived intrinsic value of psychiatric care so strongly predicts with a career as a psychiatrist. Satisfaction in the psychiatric role without a belief in the value of such work seems counterintuitive: it is difficult to consider being satisfied with the psychiatrist s role without believing in its intrinsic value. The question arises of whether this small group of psychiatrists has any positive effect on clinical care, teaching, and resident supervision. The perception of a high degree of emotional burden from patients was a second important predictor for both the regression and discriminant function analyses. Interestingly, this burden does not necessarily result in dis but can be mediated by other factors, including financial and diversification of activities. For example, the faculty of the Columbia University Public Psychiatry Fellowship surveyed its alumni regarding their roles in the organizations within which they work. The survey revealed that respondents who are medical directors perform a greater 338 Table 5 Descriptive statistics for the overall sample as derived from responses to a Likert scale Item Mean SD 1. Degree of emotional burden Availability of contact Quality of contact Degree of isolation Quality of contact Medical specialists respect psychiatry 7. Nonmedical health s respect psychiatry 8. Able to devote time to career Perception of financial success Perception of peer 11. Perception of academic 12. Perception of community Appreciation from patients Satisfaction from psychotherapy Satisfaction from university teaching Satisfaction from research Satisfaction from community involvement 18. Good balance between and personal life 19. Struggle to achieve balance between and personal life 20. Non commitments impinge on life 21. Professional commitments impinge on personal life 22. Non aspects sustain life Work has intrinsic value Regrets about choosing psychiatry a 25. Expectations not been met a Choose psychiatry again a Satisfied with career a a Criterion variables (all other variables are predictor variables). Can J Psychiatry, Vol 50, No 6, May 2005

7 Predictors of Professional and Personal Satisfaction With a Career in Psychiatry Table 6 Variables that emerged as significant predictors for specific criterion variables Predictors Regrets with choosing psychiatry as a career choice Expectations of Choose psychiatry as career in psychiatry a specialty again Overall with career Intrinsic value of work Time for career development Degree of emotional burden Other medical profession respect Degree of isolation Financial success Psychotherapy Community involvement Quality of contact Academic Availability of contact Community Professional peer Balance between and private life Appreciation from patients variety of tasks and report higher job than those who are staff psychiatrists. Respondents reported that clinical collaboration was the domain that most contributed to job (18). However, despite their belief that clinical collaboration activities most contribute to job, performance of administrative tasks most highly correlated with overall job. Various tasks that provide a respite from demanding patients likely decrease the overall emotional burden of the job. The results of our survey similarly revealed that financial success is strongly associated with job, a finding not replicated by the same Columbia group (19). The Columbia group found that frequent performance of administrative tasks was significantly associated with higher job and that salary was not significantly associated with job although control over the budget was a predictor. This difference may be attributable to differences in the work setting and the manner of reimbursement. One way to understand the link between financial success and career in our sample is to conceptualize the financial aspects as the counterbalance to the perceived emotional burden. If perceived emotional burden predicts dis with practice, it is perhaps self-evident that financial compensation for the work would be associated with higher. Satisfaction with psychotherapeutic work also emerged as a consistent predictor of a positively appraised career in psychiatry. A strong interest in psychotherapy is often what initially attracts medical students to psychiatry. Continued interest in psychotherapy would explain why this emerged as a predictor of. Recent trends suggest that psychiatric residents are increasingly trained as diagnosticians and psychopharmacologists and receive less training in psychotherapy. This study sounds a cautionary note in regard to so narrowing the psychiatrist s role. In summary, these data from a large group of Canadian psychiatrists reveal not only those factors associated with in having chosen this specialty as a career; they also highlight a not insignificant number of psychiatrists who expressed marked overall dis with their career choice. Discussing the issues related to predictors of as reported in this study may benefit the training and career planning of medical students, especially those considering a career in psychiatry. Although the relatively large sample and multifaceted assessment of career make this study s results reasonably compelling, there are some potential limitations worth noting. First, this study was conducted among a Canadian sample of psychiatrists, which compromises the generalizability of the results to psychiatrists in other countries. We believe, however, that the findings of our study can be generalized to US psychiatrists, and probably also to psychiatrists in Europe, because the theoretical models and methods used Can J Psychiatry, Vol 50, No 6, May

8 The Canadian Journal of Psychiatry Original Research Table 7 Standardized canonical discriminant function coefficients from the discriminant analysis identifying psychiatrists with very little dis vs psychiatrists with a large amount of dis Item Standardized coefficients Intrinsic value of work Quality of contact Psychotherapy Degree of Isolation Availability of contact Appreciation from patients Financial success Degree of emotional burden Balance between and private life Time for career development Professional peer Other medical profession respect Community Academic Community involvement by US psychiatrists and in most countries in the European Community are more alike than different. Moreover, most psychiatrists, regardless of the country in which they practise, face the same major challenges, which include questions about the value of psychiatry, the emotional burden imposed by patients, and personal feelings of isolation. At the same time, models of funding and health service delivery do vary across different countries and may very well contribute to our finding that Canadian psychiatrists experience less dis than their US colleagues. Future studies might address this issue by undertaking a comparative analysis of the practice characteristics of psychiatrists in Canada, the US, and Europe and attempting to link this with career. While we have already noted that a 52% return rate is comparable to the return rates for other studies conducted with physicians (13,14), it is important to recognize that the results from this investigation can only be generalized to those who responded. It remains unknown whether these individuals differ systematically from the nonrespondents. Finally, we did not explore directly contextual variables that might influence with a career in psychiatry because we were most interested in the general perception of individual psychiatrists across a wide range of subspecialties and practice settings. Future studies would do well to examine specific practice characteristics, such as work setting, workload, and patient characteristics, and examine how these may influence overall career. References 1. Looney JG, Harding RK, Blotcky MJ, Barnhart FD. Psychiatrists transition from training to career: stress and mastery. Am J Psychiatry 1980;137: Naisberg-Fenning S, Fenning S, Keinan G, Elizur A. Personality characteristics and proneness to burnout: a study among psychiatrists. Stress Med 1991;7: Deary IJ, Agius RM, Sadler A. Personality and stress in consultant psychiatrists. Int J Soc Psychiatry 1996;42: Prosser D, Johnson S, Kuipers E, Szmukler G, Bebbington P, Thornicroft G. Mental health, burnout and job among hospital and community-based mental health staff. Br J Psychiatry 1996;169: Ney PG, Tam WW, Maurice WL. Factors that determine medical student interest in psychiatry. Aust NZJPsychiatry 1990;24: Garfinkel PE, Waring EM. Personality, interests, and emotional disturbance in psychiatric residents. Am J Psychiatry 1981;138: Clark GH Jr, Vaccaro JV. Burnout among CMHC psychiatrists and the struggle to survive. Hosp Community Psychiatry 1987;38: Kalman TP, Goldstein MA. Satisfaction of Manhattan psychiatrists with private practice: assessing the impact of managed care. J Psychother Prac Res 1988;7: McKelvey RS, Webb JA. Career among psychiatrists in Texas. South Med J 1995;88: Sturm R. Datapoints: are psychiatrists more dissatisfied with their careers than other physicians? Psychiatr Serv 2001;52: Visser MR, Smets EM, Oort FJ, De Haes HC. Stress, and burnout among Dutch medical specialists. CMAJ 2003;168: Garfinkel PE, Bagby RM, Schuller DR, Williams CC, Dickens SE, Dorian B. Predictors of success and in the practice of psychiatry: a preliminary follow-up study. Can J Psychiatry 2001;46: Baker JG, Baker DF. Perceived ideological differences, job and organizational commitment among psychiatrists in a community mental health center. Community Ment Health J 1999;35: Burke RJ, Richardsen AM. Sources of and stress among Canadian physicians. Psychol Rep 1990;67: Lemkau J, Rafferty J, Gordon R Jr. Burnout and career-choice regret among family practice physicians in early practice. Fam Pract Res J 1994;14: OMA Human Resources Committee. OMA position paper on physician workforce policy and planning. Ont Med Rev 2002;4:17B26. Available: Accessed 2005 Feb Richardsen AM, Burke RJ. Occupational stress and job among physicians: sex differences. Soc Sci Med 1991;33: Ranz J, Eilenberg J, Rosenheck S. The psychiatrist s role as medical director: task distributions and job. Psychiatr Serv 1997;48: Ranz J, Stueve A, McQuistion HL. The role of the psychiatrist: job of medical directors and staff psychiatrists. Community Ment Health J 2001;37: Manuscript received May 2004, revised, and accepted July President and CEO, Centre for Addiction and Mental Health, Toronto, Ontario; Professor of Psychiatry, University of Toronto, Toronto, Ontario. 2 Professor of Psychiatry, University of Toronto, Toronto, Ontario; Director, Clinical Research, Centre for Addiction and Mental Health, Toronto, Ontario. 3 Staff Psychiatrist, Mood Disorders Program, Centre for Addiction and Mental Health, Toronto, Ontario; Assistant Professor of Psychiatry, University of Toronto, Toronto, Ontario. 4 Manager, Clinical Research, Centre for Addiction and Mental Health, Toronto, Ontario. 5 Research Analyst, Mood Disorders Program, Centre for Addiction and Mental Health, Toronto, Ontario. Address for correspondence: Dr M Bagby, Clinical Research Centre for Addiction and Mental Health, 250 College Street, Toronto, ON M5T 1R8 michael_bagby@camh.net 340 Can J Psychiatry, Vol 50, No 6, May 2005

9 Predictors of Professional and Personal Satisfaction With a Career in Psychiatry Résumé : Les prédicteurs de la professionnelle et personnelle d une carrière en psychiatrie Contexte : De nombreux facteurs, dont l expérience personnelle et les traits de personnalité, contribuent aux difficultés émotionnelles que connaissent les psychiatres dans leur travail. La nature même du travail joue aussi un rôle important. Objectif : Déterminer les caractéristiques personnelles et professionnelles qui prédisent la dans l exercice de la psychiatrie. Méthode : Nous avons posté un questionnaire qui comprenait des questions sur les aspects de la vie personnelle et professionnelle de toute la population des psychiatres de l Ontario (N = 1 574). Résultats : Sur les 1 574, 52 % (n = 802) ont répondu. Nous avons mené une série d analyses de régression pour déterminer les facteurs liés à la ou au regret quant à la carrière. Une croyance dans la valeur intrinsèque de la psychiatrie, un faible degré perçu de fardeau émotionnel provenant des patients, la réussite financière et la quant au travail psychothérapeutique ressortaient constamment comme prédicteurs significatifs. Une analyse discriminante subséquente a indiqué que toutes ces 4 variables prédisaient exactement les psychiatres ayant une extrême ou in quant à leur travail. Conclusions : Ces résultats révèlent plusieurs variables associées à la professionnelle dans l exercice de la psychiatrie, qui peuvent être utiles pour discuter avec les résidents en début de carrière. Can J Psychiatry, Vol 50, No 6, May