Canadian Psychiatric Association
Association des psychiatres du Canada
Original Research
Career Interests of Canadian Psychiatry Residents: What Makes Residents Choose a Research Career?
The Canadian Journal of Psychiatry / La Revue Canadienne de Psychiatrie 2016, Vol. 61(2) 86-92 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0706743715625952 TheCJP.ca | LaRCP.ca
Inte´reˆts professionnels des re´sidents en psychiatrie canadiens : qu’est-ce qui pousse les re´sidents a` choisir une carrie`re en recherche?
Vincent Laliberte´, MD, MSc1, Mark J. Rapoport, MD, FRCP(C)2, Melissa Andrew, MD, MEd, FRCP(C)3, Marla Davidson, MD, FRCP(C)4, and Soham Rej, MD, MSc, FRCP(C)5
Abstract Objectives: Training future clinician-researchers remains a challenge faced by Canadian psychiatry departments. Our objectives were to determine the prevalence of residents interested in pursuing research and other career options as part of their practice, and to identify the factors associated with interest in research. Method: Data from a national online survey of 207 Canadian psychiatry residents from a total of 853 (24.3% response rate) were examined. The main outcome was interest in research as part of residents’ future psychiatrist practice. Bivariate and multivariate analyses were performed to identify demographic and vocational variables associated with research interest. Results: Interest in research decreases by 76% between the first and fifth year of psychiatry residency (OR 0.76 per year, 95% CI 0.60 to 0.97). Training in a department with a residency research track did not correlate with increased research interest (w2 ¼ 0.007, df ¼ 1, P ¼ 0.93). Conclusions: Exposing and engaging psychiatry residents in research as early as possible in residency training appears key to promoting future research interest. Psychiatry residency programs and research tracks could consider emphasizing research training initiatives and protected research time early in residency. Abre´ge´ Objectifs : La formation de futurs cliniciens-chercheurs demeure un de´fi pour les de´partements de psychiatrie canadiens. Les objectifs de cette e´tude e´taient de 1) de´terminer la pre´valence des re´sidents inte´resse´s par la recherche et d’autres choix de carrie`re pour leur pratique, et 2) identifier les facteurs associe´s a` l’inte´reˆt pour la recherche. Me´thode : Les donne´es d’une enqueˆte nationale en ligne, mene´e aupre`s de 207 re´sidents en psychiatrie canadiens sur un total de 853 (taux de re´ponse de 24,3%) ont e´te´ examine´es. Le re´sultat principal e´tait l’inte´reˆt pour la recherche comme faisant partie de la future pratique psychiatrique des re´sidents. Des analyses bivarie´es et multivarie´es ont e´te´ effectue´es afin d’identifier les variables de´mographiques et professionnelles associe´es a` l’inte´reˆt pour la recherche.
1 2 3 4 5
Department of Department of Department of Department of Department of
Psychiatry, Psychiatry, Psychiatry, Psychiatry, Psychiatry,
McGill University, Montreal, Quebec Division of Geriatric Psychiatry, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Ontario Division of Geriatric Psychiatry, Queen’s University, Kingston, Ontario Division of Geriatric Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan University of Toronto, Toronto, Ontario
Corresponding Author: Soham Rej, MD, MSc, FRCPC, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Baycrest Avenue, Room FG-08, Toronto, ON M4N 3M5. Email: [email protected]
La Revue Canadienne de Psychiatrie 61(2)
87
Re´sultats : L’inte´reˆt pour la recherche diminue de 76% entre la 1re et la 5e anne´e de re´sidence en psychiatrie (RC 0,76 par anne´e, IC a` 95 0,60 a` 0,97). La formation dans un de´partement dote´ d’un volet de recherche pour les re´sidents ne corre´lait pas avec un inte´reˆt accru pour la recherche (w2 ¼ 0,007, ddl ¼ 1, P ¼ 0,93). Conclusions : Exposer et engager les re´sidents en psychiatrie a` la recherche de`s que possible durant la formation de re´sidence semble eˆtre la cle´ pour favoriser l’inte´reˆt pour la future recherche. Les programmes de re´sidence en psychiatrie et les volets de recherche pourraient envisager mettre l’accent sur les initiatives de formation en recherche et des pe´riodes prote´ge´es pour la recherche au de´but de la re´sidence. Keywords psychiatry residents, research, predictors of interest, clinician-researcher, research track, medical education
Introduction
Recruitment
Research improves our understanding of mental illness, enhances patient care, and guides novel psychiatric health services. Nonetheless, the number of physician-scientists have declined in the United States1-6 and Canada7 despite growing needs for translational research,8,9 and the drop is even more significant in psychiatry.10,11 Research tracks (RTs) were recently created across Canada to help promote residents’ interest in research. However, research training during residency remains challenging due to the wide range of skills that need to be acquired often concurrently with clinical training,12 including study design, data collection and analysis, writing manuscripts and grants, training students, and running a laboratory. To sustain a steady pipeline of future psychiatristresearchers, it will be of major importance to better understand what factors promote residents’ research interest. Previous studies have shown the critical importance of early exposure to research13 in the first years of residency training14 and even before,1,15 as well as good mentoring.8,10,13,16 In contrast, high level of trainee’s debts6,8,13 may impede research interest. In our paper, we aimed to describe psychiatry residents’ interest in research as part of their future psychiatrist practice, among other career options. We then attempted to identify potential correlates of research interest. We hypothesized that research interest would correlate with factors previously identified, particularly mentorship and early exposure to research. We also hypothesized that training in a research-focused university with a residency research track (RT) would be associated with interest in research.
Between September and October 2013, email invitations to participate were sent twice to all Canadian psychiatry residency program directors through the Coordinators of Psychiatric Education (COPE). Program directors then forwarded invitations to residents, who were offered a chance to win a $400 Amazon gift certificate if they participated. Ethics approval was obtained from University of Saskatchewan and McGill University and all participants gave online written consent.
Methods We used data from a cross-sectional Internet survey sent to all Canadian psychiatry residents from postgraduate year PGY-1 to PGY-5 among the 17 medical schools with postgraduate psychiatry training programs. The parent study was promoted as an assessment of residents’ needs and interests in geriatric psychiatry by the Canadian Academy of Geriatric Psychiatry.17 An online survey website was used (https://surveymonkey.com) and the design and implementation of this survey were based on Dillman’s Tailored Design Method.18
Outcomes Residents’ interests in research and 18 other fields of psychiatry and practice settings were assessed using their answer to the question, ‘‘What fields or practice settings are you interested in as part of your future psychiatrist practice?’’ on a 5-point Likert scale (1 ¼ not at all interested, 3 ¼ neutral, and 5 ¼ very interested). To facilitate interpretation, a rating of interested (4 or 5) or very interested (5) were defined as interest in this field of psychiatry or practice setting, while ratings of neutral or below (1 to 3) were considered not interested.
Variables Included in the Survey Demographic information were gathered: age,1,13 sex,19 ethnicity,15 first language, marital status13; as well as information on their residency training: the year of residency,19 geographic location or university, and the presence of an RT in the university.6,13 Residents also rated the importance of the following vocational variables: mentors,6,8,10 annual conferences in a field of interest, intellectual stimulation,13 subjective selfrated availability of research opportunities or job opportunities,13 the importance of incomes,13,20 lifestyle,8,20 prestige, and of the length of training. Using a similar 5-point Likert scale ranging from not important at all (1 point) to very important (5 points), the score corresponds to residents’ perception or opinion of the importance of those factors.
Statistical Analysis The prevalence of interest in research and other practice fields or settings were assessed. Our sample’s demographics were compared with all Canadian psychiatry residents using
88
The Canadian Journal of Psychiatry 61(2)
Table 1. Association of residents’ characteristics with interest in research as part of your future psychiatrist practice, n ¼ 205.
Variable
All residents, n ¼ 205
Interested in research, n ¼ 62
Not interested in research, n ¼ 143
w2 or Mann–Whitney U, df, P
Demographic, n (%) Age category, years 20–25 26–30 31–35 36–40 41–45 >45 Caucasian Speaks French as a first language Currently married or in a long-term relationship Male sex
w2 ¼ 7.427, df ¼ 5, P ¼ 0.19 24 (11.7) 105 (51.2) 53 (25.9) 20 (9.8) 2 (1.0) 1 (0.5) 142 (69.3) 34 (16.6) 142 (69.3) 62 (30.2)
11 29 12 9 1 0 39 10 40 23
(17.7) (46.8) (19.4) (14.5) (1.6) (0) (62.9) (16.1) (64.5) (37.1)
13 (9.1) 76 (53.1) 41 (28.7) 11 (7.8) 1 (0.7) 1 (0.7) 103 (72.0) 24 (16.8) 102 (71.3) 39 (27.2)
w2 ¼ w2 ¼ w2 ¼ w2 ¼
2.48 (1.37) 19 (30.6) 15 (24.2)
3.06 (1.33) 43 (30.0) 32 (22.4)
U ¼ 3319, P ¼ 0.006 w2 ¼ 0.007, df ¼ 1, P ¼ 0.93 w2 ¼ 0.081, df ¼ 1, P ¼ 0.78
129 53 85 132 17 113 87 44 113
w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼
1.69, df ¼ 1, P ¼ 0.19 0.038, df ¼ 1, P ¼ 0.98 0.94, df ¼ 1, P ¼ 0.33 1.97, df ¼ 1, P ¼ 0.16
Residency training Year of residency, mean (SD) Attending a university with a resident research track, n (%) Attending a Quebec university, n (%)
2.88 (1.37) 62 (30.2) 47 (23.0)
Vocational variables (rated as important or very important by residents), n (%) Intellectual stimulation Short length of training Financial compensation Lifestyle (working hours) Prestige Research opportunities Mentors in my field of interest Annual conferences in my field of interest Job opportunities
189 (92.2) 73 (35.6) 123 (60.0) 182 (88.8) 28 (13.7) 164 (80.0) 135 (65.9) 74 (36.1) 164 (80.0)
aggregated data from the Canadian Post-MD Education Registry (CAPER). Potential bivariate correlates of interest in research as a field of psychiatry were ascertained using chi-square, Student t test, and Mann–Whitney U tests, as appropriate. Then, logistic regression analyses, including all variables with significant bivariate correlations (2-tailed P < 0.05), were conducted to assess independent correlates of research interest. Interaction terms (for example, year of residency lifestyle ) had been assessed as part of the logistic regression modelling.
Results Respondent Characteristics The survey was completed by 207 Canadian psychiatry residents from a total of 853 (24.0%), with 205 responding to items about career interests. Respondents were mostly aged 26 to 30 years (51.2%), female (70.0%), Caucasian (68.6%), and in long-term relationships (66.7%) (Table 1). There were 45 residents who reported to be in PGY-1, 36 in PGY-2, 53 in PGY-3, 38 in PGY-4, and 33 in PGY-5. Compared with all Canadian residents (n ¼ 853), survey
60 20 38 50 11 51 48 30 51
(96.8) (32.3) (61.3) (80.6) (17.7) (82.2) (77.4) (48.4) (82.3)
(90.2) (37.1) (59.4) (92.3) (11.9) (79.0) (60.8) (30.8) (79.0)
2.59, df ¼ 1, P ¼ 0.11 0.44, df ¼ 1, P ¼ 0.51 0.062, df ¼ 1, P ¼ 0.80 5.91, df ¼ 1, P ¼ 0.015 1.26, df ¼ 1, P ¼ 0.26 0.28, df ¼ 1, P ¼ 0.60 5.29, df ¼ 1, P ¼ 0.021 5.82, df ¼ 1, P ¼ 0.016 0.28, df ¼ 1, P ¼ 0.60
respondents did not differ significantly in age, sex, year of residency, or geographic location. The prevalence of residents’ interest in a number of fields or settings of practice is described (Figure 1). Relatively high rates of residents were interested in adults aged 18 to 64 years (89.8%), outpatient (80%), psychosis (69.3%), and large urban settings (69.3%), while less residents were interested in personality disorders (42.0%) and rural settings (28.8%).
Prevalence and Bivariate Correlates of Interest in Research as a Part of Future Psychiatry Practice Residents (30.2% and 9.3%) were interested and very interested in research as part of their psychiatric practice, respectively. Potential correlates of this interest were then assessed using bivariate analyses (Table 1). Year of residency was inversely correlated with interest in research (Mann–Whitney U ¼ 3319.0, P ¼ 0.006). PGY-5 residents were 76% less likely to be interested in research than those in PGY-1 (OR 0.24, 95% CI 0.08 to 0.70), with interest decreasing from 48.9% (n ¼ 22/45) to 18.8% (n ¼ 6/32). Other demographic variables were not significantly correlated.
La Revue Canadienne de Psychiatrie 61(2)
89
Figure 1. Residents’ responses to the question: ‘‘In what fields or settings of psychiatry are you interested in practicing?’’ (n ¼ 205). Note, a 5-point Likert scale was used. Interested was defined by a rating of 4 or 5. Very interested corresponded to a rating of 5.
Compared with people aged 25 to 35 years, extreme age groups (aged 20 to 25 or >35 years) were more interested in research (44.7%, compared with 25.9%, w2 ¼ 6.02, df ¼ 1, P ¼ 0.014); however, this was an exploratory post hoc finding. Additionally, a few vocational factors were associated with research interest: mentors in my field of interest (w2 ¼ 5.29, df ¼ 1, P ¼ 0.02) and annual conferences in my field of interest (w2 ¼ 5.82, df ¼ 1, P ¼ 0.02). Meanwhile lifestyle (w2 ¼ 5.91, df ¼ 1, P ¼ 0.02) showed a negative association: residents interested in research were less likely to find lifestyle (shorter working hours) important. Lastly, compared with programs without an RT, research interest was no different in the research-oriented psychiatry programs with RTs12 (the University of Toronto, the University of British Columbia, and McGill University) (w2 ¼ 0.007, df ¼ 1, P ¼ 0.93). Training programs with an RT had somewhat lower response rates 18.2% (62/340), compared with 27.9% (143/513) (w2 ¼ 10.4, df ¼ 1, P ¼ 0.001). In non-RT universities, 41.2% of residents interested in an academic career were also interested in research, compared with 6.5% of those striving for nonacademic careers (w2 ¼ 17.9, df ¼ 1, P < 0.001). This was similar to findings in RT universities (41.3%, compared with 0%, w2 ¼ 9.53, df ¼ 1, P ¼ 0.002). Respondents from a program with an RT were not different from non-RT program respondents in terms of their year of residency.
Independent Correlates of Interest in Research as a Part of Future Psychiatry Practice In logistic regression using the 4 variables found to be correlates of interest in research with a P value of less than 0.05, 2 variables were independently and inversely associated with research interest: year of residency (OR 0.76 per additional year of residency, 95% CI 0.60 to 0.97) and considering lifestyle (shorter working hours) to be important (OR 0.46, 95% CI 0.29 to 0.73). In a second logistic regression including only these 2 variables, year of residency (OR 0.76 per year, 95% CI 0.60 to 0.95) and considering lifestyle to be important (OR 0.35, 95% CI 0.14 to 0.86) remained independent correlates of research interest.
Discussion Our study was the first nation-wide survey of Canadian residents assessing interest in various fields or settings of psychiatry, including research. We also examined a broad range of potential predictors of research interest. In our survey, 30.2% of residents were interested in research as part of their future psychiatric practice, which is comparable with a survey in 33 large American university programs where 29% of psychiatry residents had a moderate or high interest in research.13
90
The Canadian Journal of Psychiatry 61(2)
Figure 2. Proportion of psychiatry residents interested in research, by year of residency.
In keeping with the literature, we found that increasing age and marital status were not predictably associated with research interest.13 Also, we found that lifestyle (shorter working hours) are not as important among residents with research interest, which may suggest that a balanced work–life might not be as important for research-interested residents as previously considered.8 Also in contrast with previous studies,13,19,21 sex was not associated with research interest. Meanwhile mentoring and annual conferences may be contributory to research interest,9,17,22 even if they may not independently predict interest. These findings support current initiatives put in place by the Canadian Psychiatric Association (CPA), such as the CPA mentoring program23 and making conferences accessible to trainees. The most striking finding was that rates of research interest decreased markedly during residency. Compared with PGY-1 residents, those in PGY-5 were 76% less likely to be interested in research (OR 0.24, 95% CI 0.08 to 0.70) (Figure 2). Further, increased year of residency was found to independently predict lower research interest (OR 0.76 per year, 95% CI 0.60 to 0.97). Previous authors found that early developmental factors were critical in research,1,13 for example showing that over 86% of residents interested in a research career considered research prior to residency.13 Ours is the first paper to suggest that residents’ research interest may decline as their training progresses. This could be due to the various challenges faced by residents who wish to pursue a research career, such as the lack of continuity in research time during potentially productive years, competing for grants with graduating PhD students from other disciplines, as well as a relative lack of research mentoring.1 Also striking is the discrepancy between the percentage of residents interested in working in an academic setting (69.8%) and the number of residents actually interested in doing research (30.2%). In both RT and non-RT universities
alike, only 41.2% of residents with academic aspirations were interested in research. One interpretation may be that residents underestimate research as intrinsically meaningful and as an often necessary component of career promotion in academic settings. Interestingly, completing a residency program in a psychiatry department with an RT did not predict higher research interest (w2 ¼ 0.007, df ¼ 1, P ¼ 0.93). This was notable as the residency programs with an RT were situated in the most productive psychiatry research departments in Canada (University of Toronto, McGill University, and the University of British Columbia). Perhaps the most motivated residents will pursue research interests in spite of difficulties obtaining support locally: that is, that researchers can potentially bloom from any garden.
Limitations A limitation of this secondary-analysis study was that the parent study17 was promoted as a survey of interest in becoming a geriatric psychiatrist (although interest in other areas had also been assessed), which probably skewed the results toward a higher prevalence of geriatric psychiatry interest. Nonetheless, this did not appear to impact the correlation between research and potential predictors, as including geriatric psychiatry interest in the multivariate analysis did not modify our main findings. Certain potential predictor variables include the level of trainees’ debts,6,8,13 being a parent,13 exposure to clinical research,8 or the presence of a mandatory research experience in the curriculum.24 Academic productivity tended to increase in internal medicine or surgery programs that included obligatory scholarly activity,25-27 but this information had not been collected in the parent study.17 Further, information about research experience at any stage of the training including prior to residency
La Revue Canadienne de Psychiatrie 61(2)
was not part of the study, even though it may correlate with research interest in psychiatry residents13 and other recently graduated physicians.28 As well, causality is difficult to infer from a cross-sectional study, for example, to confirm whether research interest decreases during residency, one could follow a cohort of residents. Also, multiple comparisons may have contributed to alpha inflation. However, our most robust findings still approach significance even if a Bonferroni correction is applied. Also, our relatively large survey was able to test a broad range of hypotheses previously identified in smaller quantitative and qualitative studies. Future qualitative studies could help tease out why residents lose interest in research over the course of training. Finally, there was only a 24.3% response rate, although this was still quite reasonable for a demographicallyrepresentative web survey of residents,29 which often have dismal 10% response rates.18
Conclusion Research interest diminishes markedly with additional years of psychiatry residency training: a 76% decrease over 5-years of residency. Additionally, attending a residency program in a research-intensive university with an RT does not seem to predict increased interest in research. These findings have important implications for the recruitment and training of future clinician-researchers in psychiatry. Training programs, even those with an RT,12 have tended to place almost exclusive importance on clinical training in the first 2 years of residency, with little dedicated time available for research activities until PGY-4.12 Training future clinician-researchers remains a priority for Canadian psychiatry departments and universities have invested extensive resources into RTs.12 Residency programs and their RTs could consider emphasizing research training initiatives and protected research time as early as possible in residency (for example, in PGY-1 and PGY-2), and perhaps even in medical school.13 Acknowledgements We thank the CAPER, and the COPE for their generous help in implementing this survey. We also thank Dr Dallas Seitz for his support and advice with study methodology.
Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article. This work was supported by the Canadian Academy of Geriatric Psychiatry (CAGP). Dr Rej has received Master’s Training Awards from the Canadian Institutes of Health Research and the Fonds de Recherche en Sante´ du Quebec. Dr Rapoport is the president of the CAGP, while Dr Rej and Dr Davidson are CAGP board
91
members. Both Dr Rapoport and Dr Rej are on the CAGP research committee.
References 1. Gordon R. The vanishing physician scientist: a critical review and analysis. Account Res. 2012;19(2):89-113. 2. Rosenberg L. Physician-scientists—endangered and essential. Science. 1999;283(5400):331-332. 3. Ley TJ, Rosenberg LE. The physician-scientist career pipeline in 2005: build it, and they will come. JAMA. 2005;294(11): 1343-1351. 4. Zemlo TR, Garrison HH, Partridge NC, Ley TJ. The physicianscientist: career issues and challenges at the year 2000. FASEB J. 2000;14(2):221-230. 5. Hauser SL, McArthur JC. Saving the clinician-scientist: report of the ANA long range planning committee. Ann Neurol. 2006; 60(3):278-285. 6. Kupfer DJ, Hyman SE, Schatzberg AF, et al. Recruiting and retaining future generations of physician scientists in mental health. Arch Gen Psychiatry. 2002;59(7):657-660. 7. Honer WG, Linseman MA. The physician-scientist in Canadian psychiatry. J Psychiatry Neurosci. 2004;29(1): 49-56. 8. Arbuckle MR, Gordon JA, Pincus HA, et al. Bridging the gap: supporting translational research careers through an integrated research track within residency training. Acad Med. 2013; 88(6):759-765. 9. Dickler HB, Korn D, Gabbe SG. Promoting translational and clinical science: the critical role of medical schools and teaching hospitals. PLoS Med. 2006;3(9):e378. 10. Posporelis S, Sawa A, Smith GS, et al. Promoting careers in academic research to psychiatry residents. Acad Psychiatry. 2014;38(2):185-190. 11. Fenton W, James R, Insel T. Psychiatry residency training, the physician-scientist, and the future of psychiatry. Acad Psychiatry. 2004;28(4):263-266. 12. Bhat V, Leong K, Lee J, et al. Research-track programs for residents in psychiatry: a review of literature and a report of 3 Canadian experiences. Can J Psychiatry. 2014;59(5): 268-275. 13. Silberman EK, Belitsky R, Bernstein CA, et al. Recruiting researchers in psychiatry: the influence of residency vs. early motivation. Acad Psychiatry. 2012;36(2):85-90. 14. Gilbert AR, Tew JD Jr, Reynolds CF 3rd, et al. A developmental model for enhancing research training during psychiatry residency. Acad Psychiatry. 2006;30(1):55-62. 15. Back SE, Book SW, Santos AB, et al. Training physicianscientists: a model for integrating research into psychiatric residency. Acad Psychiatry. 2011;35(1):40-45. 16. Lambert MT, Garver DL. Mentoring psychiatric trainees’ first paper for publication. Acad Psychiatry. 1998;22(1):47-55. 17. Rej S, Laliberte´ V, Rapoport MJ, et al. What makes residents interested in geriatric psychiatry? A pan-Canadian online survey of psychiatry residents. Am J Geriatr Psychiatry. 2015; 23(7):735-743.
92
18. Dillman DA, Smyth JD, Christian LM. Internet, mail, and mixed-mode surveys: the tailored design method. New York (NY): John Wiley & Sons; 2008. 19. Andrews NC. The other physician-scientist problem: where have all the young girls gone? Nat Med. 2002;8(5):439-441. 20. Institute of Medicine. Research training in psychiatry residency: strategies for reform. Washington (DC): The National Academies Press; 2003. 21. Leibenluft E, Dial TH, Haviland MG, et al. Sex differences in rank attainment and research activities among academic psychiatrists. Arch Gen Psychiatry. 1993;50(11):896-904. 22. Abrams M. Research training in psychiatry residency: strategies for reform. Washington (DC): National Academies Press; 2003. 23. Canadian Psychiatric Association (CPA). The CPA mentoring program [Internet]. Ottawa (ON): CPA; 2015. Available from: http://www.cpa-apc.org. 24. Kupfer DJ. A research agenda for DSM-V. Washington (DC): American Psychiatric Association; 2002.
The Canadian Journal of Psychiatry 61(2)
25. Kanna B, Deng C, Erickson SN, et al. The research rotation: competency-based structured and novel approach to research training of internal medicine residents. BMC Med Educ. 2006;6:52. 26. Rothberg MB, Kleppel R, Friderici JL, et al. Implementing a resident research program to overcome barriers to resident research. Acad Med. 2014;89(8):1133-1139. 27. Papasavas P, Filippa D, Reilly P, et al. Effect of a mandatory research requirement on categorical resident academic productivity in a university-based general surgery residency. J Surg Educ. 2013;70(6):715-719. 28. Ledley FD, Lovejoy FH Jr. Factors influencing the interests, career paths, and research activities of recent graduates from an academic, pediatric residency program. Pediatrics. 1993; 92(3):436-441. 29. Grava-Gubins I, Scott S. Effects of various methodologic strategies: survey response rates among Canadian physicians and physicians-in-training. Can Fam Physician. 2008;54(10): 1424-1430.
Association des psychiatres du Canada
Original Research
Career Interests of Canadian Psychiatry Residents: What Makes Residents Choose a Research Career?
The Canadian Journal of Psychiatry / La Revue Canadienne de Psychiatrie 2016, Vol. 61(2) 86-92 ª The Author(s) 2015 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0706743715625952 TheCJP.ca | LaRCP.ca
Inte´reˆts professionnels des re´sidents en psychiatrie canadiens : qu’est-ce qui pousse les re´sidents a` choisir une carrie`re en recherche?
Vincent Laliberte´, MD, MSc1, Mark J. Rapoport, MD, FRCP(C)2, Melissa Andrew, MD, MEd, FRCP(C)3, Marla Davidson, MD, FRCP(C)4, and Soham Rej, MD, MSc, FRCP(C)5
Abstract Objectives: Training future clinician-researchers remains a challenge faced by Canadian psychiatry departments. Our objectives were to determine the prevalence of residents interested in pursuing research and other career options as part of their practice, and to identify the factors associated with interest in research. Method: Data from a national online survey of 207 Canadian psychiatry residents from a total of 853 (24.3% response rate) were examined. The main outcome was interest in research as part of residents’ future psychiatrist practice. Bivariate and multivariate analyses were performed to identify demographic and vocational variables associated with research interest. Results: Interest in research decreases by 76% between the first and fifth year of psychiatry residency (OR 0.76 per year, 95% CI 0.60 to 0.97). Training in a department with a residency research track did not correlate with increased research interest (w2 ¼ 0.007, df ¼ 1, P ¼ 0.93). Conclusions: Exposing and engaging psychiatry residents in research as early as possible in residency training appears key to promoting future research interest. Psychiatry residency programs and research tracks could consider emphasizing research training initiatives and protected research time early in residency. Abre´ge´ Objectifs : La formation de futurs cliniciens-chercheurs demeure un de´fi pour les de´partements de psychiatrie canadiens. Les objectifs de cette e´tude e´taient de 1) de´terminer la pre´valence des re´sidents inte´resse´s par la recherche et d’autres choix de carrie`re pour leur pratique, et 2) identifier les facteurs associe´s a` l’inte´reˆt pour la recherche. Me´thode : Les donne´es d’une enqueˆte nationale en ligne, mene´e aupre`s de 207 re´sidents en psychiatrie canadiens sur un total de 853 (taux de re´ponse de 24,3%) ont e´te´ examine´es. Le re´sultat principal e´tait l’inte´reˆt pour la recherche comme faisant partie de la future pratique psychiatrique des re´sidents. Des analyses bivarie´es et multivarie´es ont e´te´ effectue´es afin d’identifier les variables de´mographiques et professionnelles associe´es a` l’inte´reˆt pour la recherche.
1 2 3 4 5
Department of Department of Department of Department of Department of
Psychiatry, Psychiatry, Psychiatry, Psychiatry, Psychiatry,
McGill University, Montreal, Quebec Division of Geriatric Psychiatry, Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Ontario Division of Geriatric Psychiatry, Queen’s University, Kingston, Ontario Division of Geriatric Psychiatry, University of Saskatchewan, Saskatoon, Saskatchewan University of Toronto, Toronto, Ontario
Corresponding Author: Soham Rej, MD, MSc, FRCPC, Sunnybrook Health Sciences Centre, University of Toronto, 2075 Baycrest Avenue, Room FG-08, Toronto, ON M4N 3M5. Email: [email protected]
La Revue Canadienne de Psychiatrie 61(2)
87
Re´sultats : L’inte´reˆt pour la recherche diminue de 76% entre la 1re et la 5e anne´e de re´sidence en psychiatrie (RC 0,76 par anne´e, IC a` 95 0,60 a` 0,97). La formation dans un de´partement dote´ d’un volet de recherche pour les re´sidents ne corre´lait pas avec un inte´reˆt accru pour la recherche (w2 ¼ 0,007, ddl ¼ 1, P ¼ 0,93). Conclusions : Exposer et engager les re´sidents en psychiatrie a` la recherche de`s que possible durant la formation de re´sidence semble eˆtre la cle´ pour favoriser l’inte´reˆt pour la future recherche. Les programmes de re´sidence en psychiatrie et les volets de recherche pourraient envisager mettre l’accent sur les initiatives de formation en recherche et des pe´riodes prote´ge´es pour la recherche au de´but de la re´sidence. Keywords psychiatry residents, research, predictors of interest, clinician-researcher, research track, medical education
Introduction
Recruitment
Research improves our understanding of mental illness, enhances patient care, and guides novel psychiatric health services. Nonetheless, the number of physician-scientists have declined in the United States1-6 and Canada7 despite growing needs for translational research,8,9 and the drop is even more significant in psychiatry.10,11 Research tracks (RTs) were recently created across Canada to help promote residents’ interest in research. However, research training during residency remains challenging due to the wide range of skills that need to be acquired often concurrently with clinical training,12 including study design, data collection and analysis, writing manuscripts and grants, training students, and running a laboratory. To sustain a steady pipeline of future psychiatristresearchers, it will be of major importance to better understand what factors promote residents’ research interest. Previous studies have shown the critical importance of early exposure to research13 in the first years of residency training14 and even before,1,15 as well as good mentoring.8,10,13,16 In contrast, high level of trainee’s debts6,8,13 may impede research interest. In our paper, we aimed to describe psychiatry residents’ interest in research as part of their future psychiatrist practice, among other career options. We then attempted to identify potential correlates of research interest. We hypothesized that research interest would correlate with factors previously identified, particularly mentorship and early exposure to research. We also hypothesized that training in a research-focused university with a residency research track (RT) would be associated with interest in research.
Between September and October 2013, email invitations to participate were sent twice to all Canadian psychiatry residency program directors through the Coordinators of Psychiatric Education (COPE). Program directors then forwarded invitations to residents, who were offered a chance to win a $400 Amazon gift certificate if they participated. Ethics approval was obtained from University of Saskatchewan and McGill University and all participants gave online written consent.
Methods We used data from a cross-sectional Internet survey sent to all Canadian psychiatry residents from postgraduate year PGY-1 to PGY-5 among the 17 medical schools with postgraduate psychiatry training programs. The parent study was promoted as an assessment of residents’ needs and interests in geriatric psychiatry by the Canadian Academy of Geriatric Psychiatry.17 An online survey website was used (https://surveymonkey.com) and the design and implementation of this survey were based on Dillman’s Tailored Design Method.18
Outcomes Residents’ interests in research and 18 other fields of psychiatry and practice settings were assessed using their answer to the question, ‘‘What fields or practice settings are you interested in as part of your future psychiatrist practice?’’ on a 5-point Likert scale (1 ¼ not at all interested, 3 ¼ neutral, and 5 ¼ very interested). To facilitate interpretation, a rating of interested (4 or 5) or very interested (5) were defined as interest in this field of psychiatry or practice setting, while ratings of neutral or below (1 to 3) were considered not interested.
Variables Included in the Survey Demographic information were gathered: age,1,13 sex,19 ethnicity,15 first language, marital status13; as well as information on their residency training: the year of residency,19 geographic location or university, and the presence of an RT in the university.6,13 Residents also rated the importance of the following vocational variables: mentors,6,8,10 annual conferences in a field of interest, intellectual stimulation,13 subjective selfrated availability of research opportunities or job opportunities,13 the importance of incomes,13,20 lifestyle,8,20 prestige, and of the length of training. Using a similar 5-point Likert scale ranging from not important at all (1 point) to very important (5 points), the score corresponds to residents’ perception or opinion of the importance of those factors.
Statistical Analysis The prevalence of interest in research and other practice fields or settings were assessed. Our sample’s demographics were compared with all Canadian psychiatry residents using
88
The Canadian Journal of Psychiatry 61(2)
Table 1. Association of residents’ characteristics with interest in research as part of your future psychiatrist practice, n ¼ 205.
Variable
All residents, n ¼ 205
Interested in research, n ¼ 62
Not interested in research, n ¼ 143
w2 or Mann–Whitney U, df, P
Demographic, n (%) Age category, years 20–25 26–30 31–35 36–40 41–45 >45 Caucasian Speaks French as a first language Currently married or in a long-term relationship Male sex
w2 ¼ 7.427, df ¼ 5, P ¼ 0.19 24 (11.7) 105 (51.2) 53 (25.9) 20 (9.8) 2 (1.0) 1 (0.5) 142 (69.3) 34 (16.6) 142 (69.3) 62 (30.2)
11 29 12 9 1 0 39 10 40 23
(17.7) (46.8) (19.4) (14.5) (1.6) (0) (62.9) (16.1) (64.5) (37.1)
13 (9.1) 76 (53.1) 41 (28.7) 11 (7.8) 1 (0.7) 1 (0.7) 103 (72.0) 24 (16.8) 102 (71.3) 39 (27.2)
w2 ¼ w2 ¼ w2 ¼ w2 ¼
2.48 (1.37) 19 (30.6) 15 (24.2)
3.06 (1.33) 43 (30.0) 32 (22.4)
U ¼ 3319, P ¼ 0.006 w2 ¼ 0.007, df ¼ 1, P ¼ 0.93 w2 ¼ 0.081, df ¼ 1, P ¼ 0.78
129 53 85 132 17 113 87 44 113
w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼ w2 ¼
1.69, df ¼ 1, P ¼ 0.19 0.038, df ¼ 1, P ¼ 0.98 0.94, df ¼ 1, P ¼ 0.33 1.97, df ¼ 1, P ¼ 0.16
Residency training Year of residency, mean (SD) Attending a university with a resident research track, n (%) Attending a Quebec university, n (%)
2.88 (1.37) 62 (30.2) 47 (23.0)
Vocational variables (rated as important or very important by residents), n (%) Intellectual stimulation Short length of training Financial compensation Lifestyle (working hours) Prestige Research opportunities Mentors in my field of interest Annual conferences in my field of interest Job opportunities
189 (92.2) 73 (35.6) 123 (60.0) 182 (88.8) 28 (13.7) 164 (80.0) 135 (65.9) 74 (36.1) 164 (80.0)
aggregated data from the Canadian Post-MD Education Registry (CAPER). Potential bivariate correlates of interest in research as a field of psychiatry were ascertained using chi-square, Student t test, and Mann–Whitney U tests, as appropriate. Then, logistic regression analyses, including all variables with significant bivariate correlations (2-tailed P < 0.05), were conducted to assess independent correlates of research interest. Interaction terms (for example, year of residency lifestyle ) had been assessed as part of the logistic regression modelling.
Results Respondent Characteristics The survey was completed by 207 Canadian psychiatry residents from a total of 853 (24.0%), with 205 responding to items about career interests. Respondents were mostly aged 26 to 30 years (51.2%), female (70.0%), Caucasian (68.6%), and in long-term relationships (66.7%) (Table 1). There were 45 residents who reported to be in PGY-1, 36 in PGY-2, 53 in PGY-3, 38 in PGY-4, and 33 in PGY-5. Compared with all Canadian residents (n ¼ 853), survey
60 20 38 50 11 51 48 30 51
(96.8) (32.3) (61.3) (80.6) (17.7) (82.2) (77.4) (48.4) (82.3)
(90.2) (37.1) (59.4) (92.3) (11.9) (79.0) (60.8) (30.8) (79.0)
2.59, df ¼ 1, P ¼ 0.11 0.44, df ¼ 1, P ¼ 0.51 0.062, df ¼ 1, P ¼ 0.80 5.91, df ¼ 1, P ¼ 0.015 1.26, df ¼ 1, P ¼ 0.26 0.28, df ¼ 1, P ¼ 0.60 5.29, df ¼ 1, P ¼ 0.021 5.82, df ¼ 1, P ¼ 0.016 0.28, df ¼ 1, P ¼ 0.60
respondents did not differ significantly in age, sex, year of residency, or geographic location. The prevalence of residents’ interest in a number of fields or settings of practice is described (Figure 1). Relatively high rates of residents were interested in adults aged 18 to 64 years (89.8%), outpatient (80%), psychosis (69.3%), and large urban settings (69.3%), while less residents were interested in personality disorders (42.0%) and rural settings (28.8%).
Prevalence and Bivariate Correlates of Interest in Research as a Part of Future Psychiatry Practice Residents (30.2% and 9.3%) were interested and very interested in research as part of their psychiatric practice, respectively. Potential correlates of this interest were then assessed using bivariate analyses (Table 1). Year of residency was inversely correlated with interest in research (Mann–Whitney U ¼ 3319.0, P ¼ 0.006). PGY-5 residents were 76% less likely to be interested in research than those in PGY-1 (OR 0.24, 95% CI 0.08 to 0.70), with interest decreasing from 48.9% (n ¼ 22/45) to 18.8% (n ¼ 6/32). Other demographic variables were not significantly correlated.
La Revue Canadienne de Psychiatrie 61(2)
89
Figure 1. Residents’ responses to the question: ‘‘In what fields or settings of psychiatry are you interested in practicing?’’ (n ¼ 205). Note, a 5-point Likert scale was used. Interested was defined by a rating of 4 or 5. Very interested corresponded to a rating of 5.
Compared with people aged 25 to 35 years, extreme age groups (aged 20 to 25 or >35 years) were more interested in research (44.7%, compared with 25.9%, w2 ¼ 6.02, df ¼ 1, P ¼ 0.014); however, this was an exploratory post hoc finding. Additionally, a few vocational factors were associated with research interest: mentors in my field of interest (w2 ¼ 5.29, df ¼ 1, P ¼ 0.02) and annual conferences in my field of interest (w2 ¼ 5.82, df ¼ 1, P ¼ 0.02). Meanwhile lifestyle (w2 ¼ 5.91, df ¼ 1, P ¼ 0.02) showed a negative association: residents interested in research were less likely to find lifestyle (shorter working hours) important. Lastly, compared with programs without an RT, research interest was no different in the research-oriented psychiatry programs with RTs12 (the University of Toronto, the University of British Columbia, and McGill University) (w2 ¼ 0.007, df ¼ 1, P ¼ 0.93). Training programs with an RT had somewhat lower response rates 18.2% (62/340), compared with 27.9% (143/513) (w2 ¼ 10.4, df ¼ 1, P ¼ 0.001). In non-RT universities, 41.2% of residents interested in an academic career were also interested in research, compared with 6.5% of those striving for nonacademic careers (w2 ¼ 17.9, df ¼ 1, P < 0.001). This was similar to findings in RT universities (41.3%, compared with 0%, w2 ¼ 9.53, df ¼ 1, P ¼ 0.002). Respondents from a program with an RT were not different from non-RT program respondents in terms of their year of residency.
Independent Correlates of Interest in Research as a Part of Future Psychiatry Practice In logistic regression using the 4 variables found to be correlates of interest in research with a P value of less than 0.05, 2 variables were independently and inversely associated with research interest: year of residency (OR 0.76 per additional year of residency, 95% CI 0.60 to 0.97) and considering lifestyle (shorter working hours) to be important (OR 0.46, 95% CI 0.29 to 0.73). In a second logistic regression including only these 2 variables, year of residency (OR 0.76 per year, 95% CI 0.60 to 0.95) and considering lifestyle to be important (OR 0.35, 95% CI 0.14 to 0.86) remained independent correlates of research interest.
Discussion Our study was the first nation-wide survey of Canadian residents assessing interest in various fields or settings of psychiatry, including research. We also examined a broad range of potential predictors of research interest. In our survey, 30.2% of residents were interested in research as part of their future psychiatric practice, which is comparable with a survey in 33 large American university programs where 29% of psychiatry residents had a moderate or high interest in research.13
90
The Canadian Journal of Psychiatry 61(2)
Figure 2. Proportion of psychiatry residents interested in research, by year of residency.
In keeping with the literature, we found that increasing age and marital status were not predictably associated with research interest.13 Also, we found that lifestyle (shorter working hours) are not as important among residents with research interest, which may suggest that a balanced work–life might not be as important for research-interested residents as previously considered.8 Also in contrast with previous studies,13,19,21 sex was not associated with research interest. Meanwhile mentoring and annual conferences may be contributory to research interest,9,17,22 even if they may not independently predict interest. These findings support current initiatives put in place by the Canadian Psychiatric Association (CPA), such as the CPA mentoring program23 and making conferences accessible to trainees. The most striking finding was that rates of research interest decreased markedly during residency. Compared with PGY-1 residents, those in PGY-5 were 76% less likely to be interested in research (OR 0.24, 95% CI 0.08 to 0.70) (Figure 2). Further, increased year of residency was found to independently predict lower research interest (OR 0.76 per year, 95% CI 0.60 to 0.97). Previous authors found that early developmental factors were critical in research,1,13 for example showing that over 86% of residents interested in a research career considered research prior to residency.13 Ours is the first paper to suggest that residents’ research interest may decline as their training progresses. This could be due to the various challenges faced by residents who wish to pursue a research career, such as the lack of continuity in research time during potentially productive years, competing for grants with graduating PhD students from other disciplines, as well as a relative lack of research mentoring.1 Also striking is the discrepancy between the percentage of residents interested in working in an academic setting (69.8%) and the number of residents actually interested in doing research (30.2%). In both RT and non-RT universities
alike, only 41.2% of residents with academic aspirations were interested in research. One interpretation may be that residents underestimate research as intrinsically meaningful and as an often necessary component of career promotion in academic settings. Interestingly, completing a residency program in a psychiatry department with an RT did not predict higher research interest (w2 ¼ 0.007, df ¼ 1, P ¼ 0.93). This was notable as the residency programs with an RT were situated in the most productive psychiatry research departments in Canada (University of Toronto, McGill University, and the University of British Columbia). Perhaps the most motivated residents will pursue research interests in spite of difficulties obtaining support locally: that is, that researchers can potentially bloom from any garden.
Limitations A limitation of this secondary-analysis study was that the parent study17 was promoted as a survey of interest in becoming a geriatric psychiatrist (although interest in other areas had also been assessed), which probably skewed the results toward a higher prevalence of geriatric psychiatry interest. Nonetheless, this did not appear to impact the correlation between research and potential predictors, as including geriatric psychiatry interest in the multivariate analysis did not modify our main findings. Certain potential predictor variables include the level of trainees’ debts,6,8,13 being a parent,13 exposure to clinical research,8 or the presence of a mandatory research experience in the curriculum.24 Academic productivity tended to increase in internal medicine or surgery programs that included obligatory scholarly activity,25-27 but this information had not been collected in the parent study.17 Further, information about research experience at any stage of the training including prior to residency
La Revue Canadienne de Psychiatrie 61(2)
was not part of the study, even though it may correlate with research interest in psychiatry residents13 and other recently graduated physicians.28 As well, causality is difficult to infer from a cross-sectional study, for example, to confirm whether research interest decreases during residency, one could follow a cohort of residents. Also, multiple comparisons may have contributed to alpha inflation. However, our most robust findings still approach significance even if a Bonferroni correction is applied. Also, our relatively large survey was able to test a broad range of hypotheses previously identified in smaller quantitative and qualitative studies. Future qualitative studies could help tease out why residents lose interest in research over the course of training. Finally, there was only a 24.3% response rate, although this was still quite reasonable for a demographicallyrepresentative web survey of residents,29 which often have dismal 10% response rates.18
Conclusion Research interest diminishes markedly with additional years of psychiatry residency training: a 76% decrease over 5-years of residency. Additionally, attending a residency program in a research-intensive university with an RT does not seem to predict increased interest in research. These findings have important implications for the recruitment and training of future clinician-researchers in psychiatry. Training programs, even those with an RT,12 have tended to place almost exclusive importance on clinical training in the first 2 years of residency, with little dedicated time available for research activities until PGY-4.12 Training future clinician-researchers remains a priority for Canadian psychiatry departments and universities have invested extensive resources into RTs.12 Residency programs and their RTs could consider emphasizing research training initiatives and protected research time as early as possible in residency (for example, in PGY-1 and PGY-2), and perhaps even in medical school.13 Acknowledgements We thank the CAPER, and the COPE for their generous help in implementing this survey. We also thank Dr Dallas Seitz for his support and advice with study methodology.
Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article. This work was supported by the Canadian Academy of Geriatric Psychiatry (CAGP). Dr Rej has received Master’s Training Awards from the Canadian Institutes of Health Research and the Fonds de Recherche en Sante´ du Quebec. Dr Rapoport is the president of the CAGP, while Dr Rej and Dr Davidson are CAGP board
91
members. Both Dr Rapoport and Dr Rej are on the CAGP research committee.
References 1. Gordon R. The vanishing physician scientist: a critical review and analysis. Account Res. 2012;19(2):89-113. 2. Rosenberg L. Physician-scientists—endangered and essential. Science. 1999;283(5400):331-332. 3. Ley TJ, Rosenberg LE. The physician-scientist career pipeline in 2005: build it, and they will come. JAMA. 2005;294(11): 1343-1351. 4. Zemlo TR, Garrison HH, Partridge NC, Ley TJ. The physicianscientist: career issues and challenges at the year 2000. FASEB J. 2000;14(2):221-230. 5. Hauser SL, McArthur JC. Saving the clinician-scientist: report of the ANA long range planning committee. Ann Neurol. 2006; 60(3):278-285. 6. Kupfer DJ, Hyman SE, Schatzberg AF, et al. Recruiting and retaining future generations of physician scientists in mental health. Arch Gen Psychiatry. 2002;59(7):657-660. 7. Honer WG, Linseman MA. The physician-scientist in Canadian psychiatry. J Psychiatry Neurosci. 2004;29(1): 49-56. 8. Arbuckle MR, Gordon JA, Pincus HA, et al. Bridging the gap: supporting translational research careers through an integrated research track within residency training. Acad Med. 2013; 88(6):759-765. 9. Dickler HB, Korn D, Gabbe SG. Promoting translational and clinical science: the critical role of medical schools and teaching hospitals. PLoS Med. 2006;3(9):e378. 10. Posporelis S, Sawa A, Smith GS, et al. Promoting careers in academic research to psychiatry residents. Acad Psychiatry. 2014;38(2):185-190. 11. Fenton W, James R, Insel T. Psychiatry residency training, the physician-scientist, and the future of psychiatry. Acad Psychiatry. 2004;28(4):263-266. 12. Bhat V, Leong K, Lee J, et al. Research-track programs for residents in psychiatry: a review of literature and a report of 3 Canadian experiences. Can J Psychiatry. 2014;59(5): 268-275. 13. Silberman EK, Belitsky R, Bernstein CA, et al. Recruiting researchers in psychiatry: the influence of residency vs. early motivation. Acad Psychiatry. 2012;36(2):85-90. 14. Gilbert AR, Tew JD Jr, Reynolds CF 3rd, et al. A developmental model for enhancing research training during psychiatry residency. Acad Psychiatry. 2006;30(1):55-62. 15. Back SE, Book SW, Santos AB, et al. Training physicianscientists: a model for integrating research into psychiatric residency. Acad Psychiatry. 2011;35(1):40-45. 16. Lambert MT, Garver DL. Mentoring psychiatric trainees’ first paper for publication. Acad Psychiatry. 1998;22(1):47-55. 17. Rej S, Laliberte´ V, Rapoport MJ, et al. What makes residents interested in geriatric psychiatry? A pan-Canadian online survey of psychiatry residents. Am J Geriatr Psychiatry. 2015; 23(7):735-743.
92
18. Dillman DA, Smyth JD, Christian LM. Internet, mail, and mixed-mode surveys: the tailored design method. New York (NY): John Wiley & Sons; 2008. 19. Andrews NC. The other physician-scientist problem: where have all the young girls gone? Nat Med. 2002;8(5):439-441. 20. Institute of Medicine. Research training in psychiatry residency: strategies for reform. Washington (DC): The National Academies Press; 2003. 21. Leibenluft E, Dial TH, Haviland MG, et al. Sex differences in rank attainment and research activities among academic psychiatrists. Arch Gen Psychiatry. 1993;50(11):896-904. 22. Abrams M. Research training in psychiatry residency: strategies for reform. Washington (DC): National Academies Press; 2003. 23. Canadian Psychiatric Association (CPA). The CPA mentoring program [Internet]. Ottawa (ON): CPA; 2015. Available from: http://www.cpa-apc.org. 24. Kupfer DJ. A research agenda for DSM-V. Washington (DC): American Psychiatric Association; 2002.
The Canadian Journal of Psychiatry 61(2)
25. Kanna B, Deng C, Erickson SN, et al. The research rotation: competency-based structured and novel approach to research training of internal medicine residents. BMC Med Educ. 2006;6:52. 26. Rothberg MB, Kleppel R, Friderici JL, et al. Implementing a resident research program to overcome barriers to resident research. Acad Med. 2014;89(8):1133-1139. 27. Papasavas P, Filippa D, Reilly P, et al. Effect of a mandatory research requirement on categorical resident academic productivity in a university-based general surgery residency. J Surg Educ. 2013;70(6):715-719. 28. Ledley FD, Lovejoy FH Jr. Factors influencing the interests, career paths, and research activities of recent graduates from an academic, pediatric residency program. Pediatrics. 1993; 92(3):436-441. 29. Grava-Gubins I, Scott S. Effects of various methodologic strategies: survey response rates among Canadian physicians and physicians-in-training. Can Fam Physician. 2008;54(10): 1424-1430.
