Do Mentors Matter in Graduating Pediatrics Residents’ Career Choices? Rachel A. Umoren, MD, MS; Mary Pat Frintner, MSPH From the Department of Clinical Pediatrics and the Department of Pediatrics-Neonatology, Indiana University School of Medicine, and Regenstrief Institute, Inc (Dr Umoren), Indianapolis, Ind; and Department of Research, American Academy of Pediatrics (Ms Frintner), Elk Grove Village, Ill The authors have no conflicts of interest to disclose. Address correspondence to Rachel A. Umoren, MD, MS, 1050 Wishard Blvd, RG 5, Indianapolis, IN 46202 (e-mail: [email protected]). Received for publication October 22, 2013; accepted March 3, 2014.
ABSTRACT OBJECTIVE: Little is known about the association between mentorship and career choice during residency in pediatrics. This study examines graduating residents with mentors who provide career advice and the relationship between having a mentor who is a subspecialist and having a subspecialty practice goal. METHODS: National, random samples of 1000 graduating pediatrics residents were surveyed each year from 2006 to 2012; 4197 (61%) responded. Responses were pooled across years to examine mentor specialty and career goal at time of residency graduation. Multivariable logistic regression was used to examine relationships between mentor specialty and career goal at the time of graduation. RESULTS: Most (87%) residents reported having a mentor who provided career advice during residency; the proportion linearly increased from 83% in 2006 to 87% in 2012; P < .05. Forty-five percent of those with mentors had a mentor who was a subspe-
cialist; 55% had a generalist as a mentor. Overall, 45% of residents had a subspecialty career goal at time of graduation. After controlling for career goal at the start of residency and resident characteristics, residents with a subspecialist mentor were more likely to have a subspecialty career goal at time of graduation (adjusted odds ratio ¼ 5.25; 95% confidence interval, 4.41– 6.25). Residents who were male, without children, without debt, not married, not minority, and from larger residency programs were also more likely to have a subspecialty career goal at the time of graduation from residency. CONCLUSIONS: Almost 9 in 10 pediatric residents have a mentor who provides career advice. Although multiple factors shape decisions about careers, mentor specialty is one factor that might encourage residents to pursue fellowship training.
KEYWORDS: residents; mentors; career choice ACADEMIC PEDIATRICS 2014;14:348–352
WHAT’S NEW
expertise with a resident’s needs or professional interests and are in a position to offer career advice.7 Successful mentoring relationships are characterized by reciprocity, mutual respect, clear expectations, personal connection, and shared values.8 Structured mentoring programs have been described in multiple specialties with outcomes of increased resident satisfaction, knowledge attainment, improved clinical skills, increased research interests, correction of deficiencies, and career advancement.7,9,10 Although previous research indicates the benefit of mentorship during residency, little is known about the degree to which mentoring occurs in pediatrics residency programs and its effect on residents’ career choices. Current shortages of pediatric specialists in many areas and primary care pediatricians in rural and other underserved areas have been reported,11 which highlight the importance of understanding career choices and considering the strategic engagement of residents during training. The focus of this study was to explore the relationship of mentors who provide career advice during pediatric residency and the choices by graduating pediatric residents to pursue generalist or subspecialist careers. The objectives were to determine 1) the proportion of pediatric residents nationally who report having a mentor who provided career advice, and 2) the relationship between having a mentor
Nearly 9 in 10 graduating pediatrics residents reported having a mentor who provided career advice. When controlling for goals at the start of residency, graduating residents with a subspecialist mentor had 5 times the odds of having a subspecialty career goal compared with residents who had a generalist mentor.
MENTORS ARE INCREASINGLY recognized as crucial to success in the career development of young physicians.1–3 Surveys of program directors of graduate medical education programs report the proportion of programs offering mentorship vary from 49% in US Internal Medicine programs4 to 65% in Canadian postgraduate programs.5 When mentoring programs are available, the mentoring activities are often loosely monitored and the outcomes poorly evaluated.4 The Residency Review Committee for Pediatrics mandates that the program must provide a system to assist residents in career planning: “The individualized curriculum must be determined by the learning needs and career plans of each resident and must be developed through the guidance of a faculty mentor.”6 Although faculty advisers might fulfill this role, faculty mentors should match specific ACADEMIC PEDIATRICS Copyright ª 2014 by Academic Pediatric Association
348
Volume 14, Number 4 July–August 2014
ACADEMIC PEDIATRICS
DO MENTORS MATTER IN CAREER CHOICES?
who is a subspecialist and having a subspecialty career goal at the time of graduation from residency.
METHODS We analyzed data from 7 years of the American Academy of Pediatrics (AAP) Annual Survey of Graduating Residents, 2006 through 2012. Data have been collected with this survey each year since 1997, and findings have provided insight into resident characteristics, training, career intentions, and job search experiences.12–14 The survey is administered annually to a national sample of 1000 graduating pediatric residents during and after their last year of training (May through August). Residents are contacted up to 4 times through the mail and up to 4 times by e-mail, for up to 8 contacts. The e-mail includes a link to the online survey and the mailed survey includes a postage-paid return envelope. The surveys include core questions that are repeated each year and thematic questions that vary from year to year. This study focused on questions that were common across all 6 years including 1) resident characteristics, 2) having a mentor who provided career advice, 3) the mentor’s specialty, 4) plan to practice primary care at the start of residency, and 5) career goal at time of graduation. Surveys for this study were approved by the AAP Institutional Review Board. SAMPLE In the years 2006 through 2012, 1000 graduating residents were randomly selected each year from an AAP database that includes residents from all US categorical pediatrics residency programs. Residents from combined training programs, such as Medicine and Pediatrics, were not included in each sample.
MEASURES CAREER GOALS AT THE START OF RESIDENCY AND AT THE TIME OF GRADUATION All residents were asked 2 questions related to their career intentions: 1) career goal at the start of residency: “At the time you first entered residency, did you plan to practice primary care?” and 2) career goal at time of graduation: “Please describe your future clinical practice goal.” Response options for the first question included yes, no, and uncertain, and the second question included the following options: primary care, primary and subspecialty practice, subspecialty practice, hospitalist, and not entering clinical practice. Responses for the second question were categorized as follows: general pediatrics (primary care and hospitalist) and subspecialty (subspecialty and combined primary care and subspecialty) care; those not entering clinical practice were excluded from the analyses. PRESENCE OF A MENTOR To assess the presence of a mentor who provided career advice, residents were asked 1) if during their residency there was a physician who they identified as a mentor
349
and who provided career advice, and 2) the specialty of that individual. Individual mentor specialties were entered as open-ended responses by the resident. Responses on mentor specialty were categorized as follows 1) No mentor, 2) generalist (general pediatrics, family medicine, internal medicine, hospitalist, academic medicine, emergency medicine [not including pediatric emergency medicine]), and 3) subspecialist (allergy, cardiology, critical care, developmental/behavioral pediatrics, adolescent medicine, endocrinology, gastroenterology, hematology/oncology, infectious disease, nephrology, neonatal-perinatal medicine, pediatric emergency medicine, rheumatology, sports medicine, surgery, anesthesiology, dermatology, genetics, neurology, pathology, psychiatry, toxicology, child abuse, other). If respondents identified 2 mentors including a generalist and a subspecialist mentor, they were coded to the subspecialist group. ANALYSES Data on sex and age were available in the AAP database, from which the sample was drawn. We used chi-square and t-tests to compare gender and age of the respondents to those of non-respondents to assess potential response bias. Responses were pooled across years to examine resident characteristics, mentor specialty, and career goal on graduation. We used chi-square tests to examine the relationship between resident characteristics and having a mentor, and linear association chi-square tests to examine whether having a mentor changed across survey years. Logistic regression was used to examine the influence of mentor specialty and resident characteristics on having a career goal of subspecialty practice at time of graduation, while controlling for career goals at the start of residency (primary care, no primary care, or uncertain). Other predictor variables included: survey year, sex (male or female), minority race (yes: Hispanic/black or African American/ Native American, or no: white, non-Hispanic/Asian/ other), marital status (married or unmarried), have children (yes or no), have educational debt (yes or no), medical school (in or outside the United States), and program size (#15 or >15 residents per class). Many of these characteristics have been linked to differences in career intentions.12,13 The number of cases in each statistical analysis varied slightly because of missing values for specific questions. An a level of .05 was used to determine statistical significance.
RESULTS Adjusted response rates to the survey varied from a low of 57% in 2010 to a high of 64% in 2012; the combined response rate was 61% (n ¼ 4197). Most residency programs were represented within the study results by at least 1 respondent. The mean number of programs represented across the survey years (2006–2012) was 176; the total number of pediatrics programs reported in 2011 and 2012 was 198.15,16 There was no significant age difference between the study respondents (mean age ¼ 31.4
350
UMOREN AND FRINTNER
years) and nonrespondents (mean age ¼ 31.3, P ¼ .62). Women were more likely to respond than men (female response rate ¼ 65% and male response rate ¼ 57%; P < .001). Analyses included only residents who responded to the question on mentors (n ¼ 4155; 99% of respondents).
RESPONDENT CHARACTERISTICS In Table 1 the characteristics of respondents are shown. Approximately three-fourths were women, and 1 in 5 graduated from a medical school outside of the United States. Both percentages are similar to American Board of Pediatrics data for third-year residents.17
PRESENCE OF A MENTOR
ACADEMIC PEDIATRICS Table 1. Characteristics of Survey Respondents Overall Percentage (n) (N ¼ 4155)
Characteristic Female sex Minority (Hispanic, Black/African American, and Native American) Medical school outside the United States Married Have children Have educational debt Plan to practice primary care on entering residency Yes No Uncertain Future clinical practice goal General pediatrics Subspecialty Program class size (third-year residents) #15 Residents per class >15 Residents per class
74 (3066) 15 (609) 22 (903) 69 (2877) 32 (1325) 74 (3074)
37 (1484) 38 (1547) 25 (1022) 55 (2165) 45 (1786)
Over the years 2006 to 2012, 87% of pediatric residents reported having a mentor who provided career advice during residency. The proportion of residents with a mentor linearly increased from 83% in 2006 to 87% in 2012 (P < .05). Mentor presence was not related to sex, medical school location, minority race, marital status, parental status, having debt, or career goal at the start of residency to practice primary care. A small difference was found for residency program size; 89% of residents from smaller programs (#15 residents per class) and 86% of residents from larger programs (>15 residents per class) reported having a mentor who provided career advice (P < .01). Among the residents who reported having a mentor who provided career advice, 93% reported the specialty of the mentor, which included 30 different specialties. Fortyfive percent of the responses on mentor specialty were categorized as subspecialist and 55% of the responses were categorized as generalist.
had a subspecialty career goal at the time of graduation (P < .001). In multivariable analyses controlling for career goal at the start of residency, residents with subspecialist mentors were more likely to have a subspecialty career goal at the time of graduation, rather than a general pediatrics career (adjusted odds ratio ¼ 5.25; 95% confidence interval, 4.41–6.25). Other factors that were associated with having a subspecialty career goal at the end of residency included being male, unmarried, without children, without debt, white/Asian/other race, and graduation from a larger residency program (Table 2).
MENTOR SPECIALTY AND CAREER INTENTIONS
DISCUSSION
Forty-five percent of residents reported having a career goal of subspecialty care (38% subspecialty and 7% combined primary care and subspecialty practice) and 55% reported a general pediatrics career goal (43% primary care and 12% hospitalist) at the time of graduation. Among the 45% of residents who had a subspecialty career goal, 61% had a subspecialist mentor, 26% a generalist mentor, and 13% had no mentor. Among all residents, those with a subspecialist mentor were more likely to have a subspecialist career goal at the time of graduation (70%) than residents with a generalist mentor (25%) or no mentor (46%); P < .001. More than a third of the residents (37%) reported a career goal at the start of residency to practice primary care (Table 1). Residents with a subspecialist mentor were more likely than those with a generalist mentor to have a subspecialist career goal at the time of graduation (P < .001), regardless of their career goal at the start of residency to practice primary care (Figure). Of the residents who had a primary care goal at the start of residency, 41% of those with a subspecialist mentor, 19% of those with no mentor, and 11% of those with a generalist mentor
Using data from the AAP Annual Survey of Graduating Residents from 2006 to 2012, we found that nearly 9 in 10 residents reported having a mentor who provided career advice. Residents with subspecialist mentors were more likely than those with a generalist or no mentor to have a subspecialty rather than a general pediatrics career goal at the time of graduation. The proportion of pediatrics residents with a mentor is greater than published reports of other US graduate medical education specialties and international postgraduate medical programs in which the reported number of programs with mentors ranges from 49% to 65%.2,4,18 The recommendation by the Pediatrics Residency Review Committee to have faculty assist residents with learning needs and career plans might contribute to the number of programs that provide mentors. Pediatrics residents who complete a senior project as part of their residency requirements often need faculty guidance, which might evolve into a deeper mentoring relationship. When controlling for career goals at the start of residency, residents with a subspecialist mentor had 5 times the odds of having a subspecialty rather than a general
46 (1884) 54 (2211)
ACADEMIC PEDIATRICS
DO MENTORS MATTER IN CAREER CHOICES?
351
Figure. Mentor specialty and the proportion of residents with a subspecialty career goal at the time of graduation from residency.
pediatrics practice goal at the time of graduation. This compares with previous reports in radiology and emergency medicine in which physicians in academic practice, but not private practice, were influenced by mentors.19–22 Despite that interest in pediatric subspecialties is at an all-time high (33% in 2010, up from 25% in 2004),23 the AAP 2013 Pediatrician Workforce Policy Statement describes a current shortage of pediatric specialists in many areas.11 These findings underscore the need for residency programs to strategically engage subspecialists in general pediatrics residency training to encourage the flow of residents into subspecialties in which pediatric specialists are lacking such as neurology, developmental-behavioral pediatrics, pulmonology, and rheumatology.11,24 In underserved and rural areas, generalists and specialists are lacking25 and residency programs can play an active role in linking residents to appropriate mentors, whether generalist or specialist. Mentoring programs in pediatrics residency need to be well structured and monitored to be able to evaluate their effect and outcomes.8,26 One such model for the optimal Table 2. Predictors of Subspecialty Career Goal at Time of Graduation Predictor Mentor Generalist No mentor Subspecialist Primary care career goal at start of residency Yes No Uncertain Male sex No debt No children Not married/partnered White/Asian/other Race Program class size >15
Adjusted Odds Ratio; 95% CI Reference 2.14; 1.69–2.71 5.25; 4.41–6.25
Reference 6.21; 5.11–7.56 2.75; 2.24–3.38 1.66; 1.38–1.99 1.41; 1.13–1.76 1.48; 1.22–1.79 1.32; 1.09–1.60 1.30; 1.03–1.64 1.26; 1.07–1.49
CI ¼ confidence interval. Also in the model but not significant: survey year; medical school outside of the United States.
mentoring relationship developed by Davis and Nakamura includes 6 elements: “emotional safety, support, protegecenteredness, informality, responsiveness, and respect.”27 With these 6 mentoring practices, mentees can develop independence, reflection, extrapolation, and synthesis. Pediatric programs might also adapt tool kits developed in other specialties such as internal medicine, such as a program described by Levy et al that provides special recognition to faculty for contributing to the career development of residents.28 Other predictors of having a subspecialty career goal are gender, marital and parental status, debt, program size, and race. These findings are similar to other studies that examined the role of these factors in student planning of academic careers.12,20,25,29–31 In our model, international medical graduate status did not significantly influence residents’ subspecialty career goal. There are limitations to this study. The data were obtained by resident self-report and could be subject to recall bias. Overall, women were more likely to respond to the survey. As a result, interest in subspecialty training might be underestimated. Analyses were based on career goals, but it is not known if the residents actually followed through with these plans. There is also a time-order limitation in that even though we controlled for the residents’ plans to practice primary care at the start of residency, it is still unknown if the mentors changed the residents’ attitudes or if the residents’ attitudes changed and then they sought the mentors. We did not distinguish whether the trainee’s choice of subspecialty was the same as, or different from, the mentor’s specialty, and this could be a subject for future study. Although it is relevant to determine if the trainees decide to specialize in the same area as their mentor, it might be equally important for programs that assign mentors to recognize that subspecialty and generalist mentors might provide different guidance to trainees, and that when subspecialty mentors are paired with trainees, this might influence the trainees’ career choices. Whether mentors were assigned by the program or selected by the resident was also unclear. Half of the group of pediatric residents who graduated residency in
352
UMOREN AND FRINTNER
2011 and reported having a mentor who provided career advice reported that the mentor was assigned by their program.32 Although program assignments might be based on resident career intentions, it is possible that other factors were involved. Further investigation is needed to examine the nature of career advice provided and to determine whether resident-selected versus program-assigned mentors influence residents’ career choices.
CONCLUSION Almost 9 in 10 graduating pediatric residents had a mentor who provided career advice. Controlling for career plans at the start of residency, pediatrics residents with subspecialist mentors were more likely to plan a subspecialty career at the time of graduation, rather than a general pediatrics career. Mentor specialty is one of many factors including gender, marital and parental status, debt, program size, and race associated with pediatric residents’ career choices. ACKNOWLEDGMENTS This study was funded by the AAP. We thank all the pediatricians who completed the AAP Annual Survey of Graduating Residents.
REFERENCES 1. Steele MM, Fisman S, Davidson B. Mentoring and role models in recruitment and retention: a study of junior medical faculty perceptions. Med Teach. 2012;35:e1–e9. 2. Stamm M, Buddeberg–Fischer B. The impact of mentoring during postgraduate training on doctors’ career success. Med Educ. 2011; 45:488–496. 3. Harnisch BA, Stolzmann KL, Lerner LB. Mentoring, fellowship training and academic careers of women urologists. BJU Int. 2010; 105:1629–1631. 4. Castiglioni A, Bellini LM, Shea JA. Program directors’ views of the importance and prevalence of mentoring in internal medicine residencies. J Gen Int Med. 2004;19:779–782. 5. Donovan A, Donovan J. Mentorship in postgraduate training programmes: views of Canadian programme directors. Med Educ. 2009;43:155–158. 6. Accreditation Council for Graduate Medical Education. ACGME Program Requirements for Graduate Medical Education in Pediatrics. Available at: http://www.acgme.org/acgmeweb/Portals/0/PFAssets/ 2013-PR-FAQ-PIF/320_pediatrics_07012013.pdf. Accessed October 16, 2013. 7. Woods SK, Burgess L, Kaminetzky C, et al. Defining the roles of advisors and mentors in postgraduate medical education: faculty perceptions, roles, responsibilities, and resource needs. J Grad Med Educ. 2010;2:195–200. 8. Straus SE, Johnson MO, Marquez C, Feldman MD. Characteristics of successful and failed mentoring relationships: a qualitative study across two academic health centers. Acad Med. 2013;88:82–89. 9. Ogunyemi D, Solnik MJ, Alexander C, et al. Promoting residents’ professional development and academic productivity using a structured faculty mentoring program. Teach Learn Med. 2010;22:93–96. 10. Curtis JA, Adam H, Shelov SP. A formal mentoring program in a pediatric residency. Acad Med. 1995;70:453–454.
ACADEMIC PEDIATRICS 11. Basco WT, Rimsza ME. Committee on Pediatric Workforce; American Academy of Pediatrics. Pediatrician workforce policy statement. Pediatrics. 2013;132:390–397. 12. Frintner MP, Mulvey HJ, Pletcher BA, Olson LM. Pediatric resident debt and career intentions. Pediatrics. 2013;131:312–318. 13. Frintner MP, Cull WL. Pediatric training and career intentions, 20032009. Pediatrics. 2012;129:522–528. 14. Cull WL, Caspary GL, Olson LM. Many pediatric residents seek and obtain part-time positions. Pediatrics. 2008;121:276–281. 15. Brotherton SE, Etzel SI. Graduate medical education, 2011-2012. JAMA. 2012;308:2264–2279. 16. Brotherton SE, Etzel SI. Graduate medical education, 2010-2011. JAMA. 2011;306:1015–1030. 17. American Board of Pediatrics. Workforce Data 2012-2013. Available at: https://www.abp.org/abpwebsite/stats/wrkfrc/workforcebook.pdf. Accessed January 9, 2013. 18. Ramanan RA, Taylor WC, Davis RB, Phillips RS. Mentoring matters. J Gen Int Med. 2006;21:340–345. 19. Rubeck RF, Donnelly MB, Jarecky RM, et al. Demographic, educational, and psychosocial factors influencing the choices of primary care and academic medical careers. Acad Med. 1995;70:318–320. 20. Wood PS, Altmaier EM, Franken EA, et al. Factors influencing choice of academic or practice careers in radiology. Invest Radiol. 1990;25: 675–677. 21. Sanders AB, Fulginiti JV, Witzke DB, Bangs KA. Characteristics influencing career decisions of academic and nonacademic emergency physicians. Ann Emerg Med. 1994;23:81–87. 22. Bilbey J, Fache J, Burhenne H. Are there predictors for future academic radiologists? A Canadian survey. Can Assoc Radiol J. 1992; 43:369–373. 23. Althouse LA, Stockman JA. The pediatric workforce: an update on general pediatrics and pediatric subspecialties workforce data from the American Board of Pediatrics. J Pediatr. 2011;159:1036–1040.e3. 24. Children’s Hospital Association. Pediatric specialist physician shortages affect access to care. Available at: http://www.childrenshospitals.net/ AM/Template.cfm?Template¼/CM/ContentDisplay.cfm&ContentID¼ 63293. Accessed October 16, 2013. 25. 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;92:436–441. 26. Cameron C, Collie CL, Baldwin CD, et al. The development of scientific communication skills: a qualitative study of the perceptions of trainees and their mentors. Acad Med. 2013;88:1499–1506. 27. Davis OC, Nakamura J. A proposed model for an optimal mentoring environment for medical residents: a literature review. Acad Med. 2010;85:1060–1066. 28. Levy BD, Katz JT, Wolf MA, et al. An initiative in mentoring to promote residents’ and faculty members’ careers. Acad Med. 2004;79: 845–850. 29. Cain JM, Schulkin J, Parisi V, et al. Effects of perceptions and mentorship on pursuing a career in academic medicine in obstetrics and gynecology. Acad Med. 2001;76:628–634. 30. O’Sullivan PS, Niehaus B, Lockspeiser TM, Irby DM. Becoming an academic doctor: perceptions of scholarly careers. Med Educ. 2009; 43:335–341. 31. Straus SE, Straus C, Tzanetos K. Career choice in academic medicine: systematic review. J Gen Int Med. 2006;21:1222–1229. 32. American Academy of Pediatrics. Annual survey of graduating residents. Available at: http://www.aap.org/research/graduatingressurvey. htm. Accessed October 16, 2013.
ABSTRACT OBJECTIVE: Little is known about the association between mentorship and career choice during residency in pediatrics. This study examines graduating residents with mentors who provide career advice and the relationship between having a mentor who is a subspecialist and having a subspecialty practice goal. METHODS: National, random samples of 1000 graduating pediatrics residents were surveyed each year from 2006 to 2012; 4197 (61%) responded. Responses were pooled across years to examine mentor specialty and career goal at time of residency graduation. Multivariable logistic regression was used to examine relationships between mentor specialty and career goal at the time of graduation. RESULTS: Most (87%) residents reported having a mentor who provided career advice during residency; the proportion linearly increased from 83% in 2006 to 87% in 2012; P < .05. Forty-five percent of those with mentors had a mentor who was a subspe-
cialist; 55% had a generalist as a mentor. Overall, 45% of residents had a subspecialty career goal at time of graduation. After controlling for career goal at the start of residency and resident characteristics, residents with a subspecialist mentor were more likely to have a subspecialty career goal at time of graduation (adjusted odds ratio ¼ 5.25; 95% confidence interval, 4.41– 6.25). Residents who were male, without children, without debt, not married, not minority, and from larger residency programs were also more likely to have a subspecialty career goal at the time of graduation from residency. CONCLUSIONS: Almost 9 in 10 pediatric residents have a mentor who provides career advice. Although multiple factors shape decisions about careers, mentor specialty is one factor that might encourage residents to pursue fellowship training.
KEYWORDS: residents; mentors; career choice ACADEMIC PEDIATRICS 2014;14:348–352
WHAT’S NEW
expertise with a resident’s needs or professional interests and are in a position to offer career advice.7 Successful mentoring relationships are characterized by reciprocity, mutual respect, clear expectations, personal connection, and shared values.8 Structured mentoring programs have been described in multiple specialties with outcomes of increased resident satisfaction, knowledge attainment, improved clinical skills, increased research interests, correction of deficiencies, and career advancement.7,9,10 Although previous research indicates the benefit of mentorship during residency, little is known about the degree to which mentoring occurs in pediatrics residency programs and its effect on residents’ career choices. Current shortages of pediatric specialists in many areas and primary care pediatricians in rural and other underserved areas have been reported,11 which highlight the importance of understanding career choices and considering the strategic engagement of residents during training. The focus of this study was to explore the relationship of mentors who provide career advice during pediatric residency and the choices by graduating pediatric residents to pursue generalist or subspecialist careers. The objectives were to determine 1) the proportion of pediatric residents nationally who report having a mentor who provided career advice, and 2) the relationship between having a mentor
Nearly 9 in 10 graduating pediatrics residents reported having a mentor who provided career advice. When controlling for goals at the start of residency, graduating residents with a subspecialist mentor had 5 times the odds of having a subspecialty career goal compared with residents who had a generalist mentor.
MENTORS ARE INCREASINGLY recognized as crucial to success in the career development of young physicians.1–3 Surveys of program directors of graduate medical education programs report the proportion of programs offering mentorship vary from 49% in US Internal Medicine programs4 to 65% in Canadian postgraduate programs.5 When mentoring programs are available, the mentoring activities are often loosely monitored and the outcomes poorly evaluated.4 The Residency Review Committee for Pediatrics mandates that the program must provide a system to assist residents in career planning: “The individualized curriculum must be determined by the learning needs and career plans of each resident and must be developed through the guidance of a faculty mentor.”6 Although faculty advisers might fulfill this role, faculty mentors should match specific ACADEMIC PEDIATRICS Copyright ª 2014 by Academic Pediatric Association
348
Volume 14, Number 4 July–August 2014
ACADEMIC PEDIATRICS
DO MENTORS MATTER IN CAREER CHOICES?
who is a subspecialist and having a subspecialty career goal at the time of graduation from residency.
METHODS We analyzed data from 7 years of the American Academy of Pediatrics (AAP) Annual Survey of Graduating Residents, 2006 through 2012. Data have been collected with this survey each year since 1997, and findings have provided insight into resident characteristics, training, career intentions, and job search experiences.12–14 The survey is administered annually to a national sample of 1000 graduating pediatric residents during and after their last year of training (May through August). Residents are contacted up to 4 times through the mail and up to 4 times by e-mail, for up to 8 contacts. The e-mail includes a link to the online survey and the mailed survey includes a postage-paid return envelope. The surveys include core questions that are repeated each year and thematic questions that vary from year to year. This study focused on questions that were common across all 6 years including 1) resident characteristics, 2) having a mentor who provided career advice, 3) the mentor’s specialty, 4) plan to practice primary care at the start of residency, and 5) career goal at time of graduation. Surveys for this study were approved by the AAP Institutional Review Board. SAMPLE In the years 2006 through 2012, 1000 graduating residents were randomly selected each year from an AAP database that includes residents from all US categorical pediatrics residency programs. Residents from combined training programs, such as Medicine and Pediatrics, were not included in each sample.
MEASURES CAREER GOALS AT THE START OF RESIDENCY AND AT THE TIME OF GRADUATION All residents were asked 2 questions related to their career intentions: 1) career goal at the start of residency: “At the time you first entered residency, did you plan to practice primary care?” and 2) career goal at time of graduation: “Please describe your future clinical practice goal.” Response options for the first question included yes, no, and uncertain, and the second question included the following options: primary care, primary and subspecialty practice, subspecialty practice, hospitalist, and not entering clinical practice. Responses for the second question were categorized as follows: general pediatrics (primary care and hospitalist) and subspecialty (subspecialty and combined primary care and subspecialty) care; those not entering clinical practice were excluded from the analyses. PRESENCE OF A MENTOR To assess the presence of a mentor who provided career advice, residents were asked 1) if during their residency there was a physician who they identified as a mentor
349
and who provided career advice, and 2) the specialty of that individual. Individual mentor specialties were entered as open-ended responses by the resident. Responses on mentor specialty were categorized as follows 1) No mentor, 2) generalist (general pediatrics, family medicine, internal medicine, hospitalist, academic medicine, emergency medicine [not including pediatric emergency medicine]), and 3) subspecialist (allergy, cardiology, critical care, developmental/behavioral pediatrics, adolescent medicine, endocrinology, gastroenterology, hematology/oncology, infectious disease, nephrology, neonatal-perinatal medicine, pediatric emergency medicine, rheumatology, sports medicine, surgery, anesthesiology, dermatology, genetics, neurology, pathology, psychiatry, toxicology, child abuse, other). If respondents identified 2 mentors including a generalist and a subspecialist mentor, they were coded to the subspecialist group. ANALYSES Data on sex and age were available in the AAP database, from which the sample was drawn. We used chi-square and t-tests to compare gender and age of the respondents to those of non-respondents to assess potential response bias. Responses were pooled across years to examine resident characteristics, mentor specialty, and career goal on graduation. We used chi-square tests to examine the relationship between resident characteristics and having a mentor, and linear association chi-square tests to examine whether having a mentor changed across survey years. Logistic regression was used to examine the influence of mentor specialty and resident characteristics on having a career goal of subspecialty practice at time of graduation, while controlling for career goals at the start of residency (primary care, no primary care, or uncertain). Other predictor variables included: survey year, sex (male or female), minority race (yes: Hispanic/black or African American/ Native American, or no: white, non-Hispanic/Asian/ other), marital status (married or unmarried), have children (yes or no), have educational debt (yes or no), medical school (in or outside the United States), and program size (#15 or >15 residents per class). Many of these characteristics have been linked to differences in career intentions.12,13 The number of cases in each statistical analysis varied slightly because of missing values for specific questions. An a level of .05 was used to determine statistical significance.
RESULTS Adjusted response rates to the survey varied from a low of 57% in 2010 to a high of 64% in 2012; the combined response rate was 61% (n ¼ 4197). Most residency programs were represented within the study results by at least 1 respondent. The mean number of programs represented across the survey years (2006–2012) was 176; the total number of pediatrics programs reported in 2011 and 2012 was 198.15,16 There was no significant age difference between the study respondents (mean age ¼ 31.4
350
UMOREN AND FRINTNER
years) and nonrespondents (mean age ¼ 31.3, P ¼ .62). Women were more likely to respond than men (female response rate ¼ 65% and male response rate ¼ 57%; P < .001). Analyses included only residents who responded to the question on mentors (n ¼ 4155; 99% of respondents).
RESPONDENT CHARACTERISTICS In Table 1 the characteristics of respondents are shown. Approximately three-fourths were women, and 1 in 5 graduated from a medical school outside of the United States. Both percentages are similar to American Board of Pediatrics data for third-year residents.17
PRESENCE OF A MENTOR
ACADEMIC PEDIATRICS Table 1. Characteristics of Survey Respondents Overall Percentage (n) (N ¼ 4155)
Characteristic Female sex Minority (Hispanic, Black/African American, and Native American) Medical school outside the United States Married Have children Have educational debt Plan to practice primary care on entering residency Yes No Uncertain Future clinical practice goal General pediatrics Subspecialty Program class size (third-year residents) #15 Residents per class >15 Residents per class
74 (3066) 15 (609) 22 (903) 69 (2877) 32 (1325) 74 (3074)
37 (1484) 38 (1547) 25 (1022) 55 (2165) 45 (1786)
Over the years 2006 to 2012, 87% of pediatric residents reported having a mentor who provided career advice during residency. The proportion of residents with a mentor linearly increased from 83% in 2006 to 87% in 2012 (P < .05). Mentor presence was not related to sex, medical school location, minority race, marital status, parental status, having debt, or career goal at the start of residency to practice primary care. A small difference was found for residency program size; 89% of residents from smaller programs (#15 residents per class) and 86% of residents from larger programs (>15 residents per class) reported having a mentor who provided career advice (P < .01). Among the residents who reported having a mentor who provided career advice, 93% reported the specialty of the mentor, which included 30 different specialties. Fortyfive percent of the responses on mentor specialty were categorized as subspecialist and 55% of the responses were categorized as generalist.
had a subspecialty career goal at the time of graduation (P < .001). In multivariable analyses controlling for career goal at the start of residency, residents with subspecialist mentors were more likely to have a subspecialty career goal at the time of graduation, rather than a general pediatrics career (adjusted odds ratio ¼ 5.25; 95% confidence interval, 4.41–6.25). Other factors that were associated with having a subspecialty career goal at the end of residency included being male, unmarried, without children, without debt, white/Asian/other race, and graduation from a larger residency program (Table 2).
MENTOR SPECIALTY AND CAREER INTENTIONS
DISCUSSION
Forty-five percent of residents reported having a career goal of subspecialty care (38% subspecialty and 7% combined primary care and subspecialty practice) and 55% reported a general pediatrics career goal (43% primary care and 12% hospitalist) at the time of graduation. Among the 45% of residents who had a subspecialty career goal, 61% had a subspecialist mentor, 26% a generalist mentor, and 13% had no mentor. Among all residents, those with a subspecialist mentor were more likely to have a subspecialist career goal at the time of graduation (70%) than residents with a generalist mentor (25%) or no mentor (46%); P < .001. More than a third of the residents (37%) reported a career goal at the start of residency to practice primary care (Table 1). Residents with a subspecialist mentor were more likely than those with a generalist mentor to have a subspecialist career goal at the time of graduation (P < .001), regardless of their career goal at the start of residency to practice primary care (Figure). Of the residents who had a primary care goal at the start of residency, 41% of those with a subspecialist mentor, 19% of those with no mentor, and 11% of those with a generalist mentor
Using data from the AAP Annual Survey of Graduating Residents from 2006 to 2012, we found that nearly 9 in 10 residents reported having a mentor who provided career advice. Residents with subspecialist mentors were more likely than those with a generalist or no mentor to have a subspecialty rather than a general pediatrics career goal at the time of graduation. The proportion of pediatrics residents with a mentor is greater than published reports of other US graduate medical education specialties and international postgraduate medical programs in which the reported number of programs with mentors ranges from 49% to 65%.2,4,18 The recommendation by the Pediatrics Residency Review Committee to have faculty assist residents with learning needs and career plans might contribute to the number of programs that provide mentors. Pediatrics residents who complete a senior project as part of their residency requirements often need faculty guidance, which might evolve into a deeper mentoring relationship. When controlling for career goals at the start of residency, residents with a subspecialist mentor had 5 times the odds of having a subspecialty rather than a general
46 (1884) 54 (2211)
ACADEMIC PEDIATRICS
DO MENTORS MATTER IN CAREER CHOICES?
351
Figure. Mentor specialty and the proportion of residents with a subspecialty career goal at the time of graduation from residency.
pediatrics practice goal at the time of graduation. This compares with previous reports in radiology and emergency medicine in which physicians in academic practice, but not private practice, were influenced by mentors.19–22 Despite that interest in pediatric subspecialties is at an all-time high (33% in 2010, up from 25% in 2004),23 the AAP 2013 Pediatrician Workforce Policy Statement describes a current shortage of pediatric specialists in many areas.11 These findings underscore the need for residency programs to strategically engage subspecialists in general pediatrics residency training to encourage the flow of residents into subspecialties in which pediatric specialists are lacking such as neurology, developmental-behavioral pediatrics, pulmonology, and rheumatology.11,24 In underserved and rural areas, generalists and specialists are lacking25 and residency programs can play an active role in linking residents to appropriate mentors, whether generalist or specialist. Mentoring programs in pediatrics residency need to be well structured and monitored to be able to evaluate their effect and outcomes.8,26 One such model for the optimal Table 2. Predictors of Subspecialty Career Goal at Time of Graduation Predictor Mentor Generalist No mentor Subspecialist Primary care career goal at start of residency Yes No Uncertain Male sex No debt No children Not married/partnered White/Asian/other Race Program class size >15
Adjusted Odds Ratio; 95% CI Reference 2.14; 1.69–2.71 5.25; 4.41–6.25
Reference 6.21; 5.11–7.56 2.75; 2.24–3.38 1.66; 1.38–1.99 1.41; 1.13–1.76 1.48; 1.22–1.79 1.32; 1.09–1.60 1.30; 1.03–1.64 1.26; 1.07–1.49
CI ¼ confidence interval. Also in the model but not significant: survey year; medical school outside of the United States.
mentoring relationship developed by Davis and Nakamura includes 6 elements: “emotional safety, support, protegecenteredness, informality, responsiveness, and respect.”27 With these 6 mentoring practices, mentees can develop independence, reflection, extrapolation, and synthesis. Pediatric programs might also adapt tool kits developed in other specialties such as internal medicine, such as a program described by Levy et al that provides special recognition to faculty for contributing to the career development of residents.28 Other predictors of having a subspecialty career goal are gender, marital and parental status, debt, program size, and race. These findings are similar to other studies that examined the role of these factors in student planning of academic careers.12,20,25,29–31 In our model, international medical graduate status did not significantly influence residents’ subspecialty career goal. There are limitations to this study. The data were obtained by resident self-report and could be subject to recall bias. Overall, women were more likely to respond to the survey. As a result, interest in subspecialty training might be underestimated. Analyses were based on career goals, but it is not known if the residents actually followed through with these plans. There is also a time-order limitation in that even though we controlled for the residents’ plans to practice primary care at the start of residency, it is still unknown if the mentors changed the residents’ attitudes or if the residents’ attitudes changed and then they sought the mentors. We did not distinguish whether the trainee’s choice of subspecialty was the same as, or different from, the mentor’s specialty, and this could be a subject for future study. Although it is relevant to determine if the trainees decide to specialize in the same area as their mentor, it might be equally important for programs that assign mentors to recognize that subspecialty and generalist mentors might provide different guidance to trainees, and that when subspecialty mentors are paired with trainees, this might influence the trainees’ career choices. Whether mentors were assigned by the program or selected by the resident was also unclear. Half of the group of pediatric residents who graduated residency in
352
UMOREN AND FRINTNER
2011 and reported having a mentor who provided career advice reported that the mentor was assigned by their program.32 Although program assignments might be based on resident career intentions, it is possible that other factors were involved. Further investigation is needed to examine the nature of career advice provided and to determine whether resident-selected versus program-assigned mentors influence residents’ career choices.
CONCLUSION Almost 9 in 10 graduating pediatric residents had a mentor who provided career advice. Controlling for career plans at the start of residency, pediatrics residents with subspecialist mentors were more likely to plan a subspecialty career at the time of graduation, rather than a general pediatrics career. Mentor specialty is one of many factors including gender, marital and parental status, debt, program size, and race associated with pediatric residents’ career choices. ACKNOWLEDGMENTS This study was funded by the AAP. We thank all the pediatricians who completed the AAP Annual Survey of Graduating Residents.
REFERENCES 1. Steele MM, Fisman S, Davidson B. Mentoring and role models in recruitment and retention: a study of junior medical faculty perceptions. Med Teach. 2012;35:e1–e9. 2. Stamm M, Buddeberg–Fischer B. The impact of mentoring during postgraduate training on doctors’ career success. Med Educ. 2011; 45:488–496. 3. Harnisch BA, Stolzmann KL, Lerner LB. Mentoring, fellowship training and academic careers of women urologists. BJU Int. 2010; 105:1629–1631. 4. Castiglioni A, Bellini LM, Shea JA. Program directors’ views of the importance and prevalence of mentoring in internal medicine residencies. J Gen Int Med. 2004;19:779–782. 5. Donovan A, Donovan J. Mentorship in postgraduate training programmes: views of Canadian programme directors. Med Educ. 2009;43:155–158. 6. Accreditation Council for Graduate Medical Education. ACGME Program Requirements for Graduate Medical Education in Pediatrics. Available at: http://www.acgme.org/acgmeweb/Portals/0/PFAssets/ 2013-PR-FAQ-PIF/320_pediatrics_07012013.pdf. Accessed October 16, 2013. 7. Woods SK, Burgess L, Kaminetzky C, et al. Defining the roles of advisors and mentors in postgraduate medical education: faculty perceptions, roles, responsibilities, and resource needs. J Grad Med Educ. 2010;2:195–200. 8. Straus SE, Johnson MO, Marquez C, Feldman MD. Characteristics of successful and failed mentoring relationships: a qualitative study across two academic health centers. Acad Med. 2013;88:82–89. 9. Ogunyemi D, Solnik MJ, Alexander C, et al. Promoting residents’ professional development and academic productivity using a structured faculty mentoring program. Teach Learn Med. 2010;22:93–96. 10. Curtis JA, Adam H, Shelov SP. A formal mentoring program in a pediatric residency. Acad Med. 1995;70:453–454.
ACADEMIC PEDIATRICS 11. Basco WT, Rimsza ME. Committee on Pediatric Workforce; American Academy of Pediatrics. Pediatrician workforce policy statement. Pediatrics. 2013;132:390–397. 12. Frintner MP, Mulvey HJ, Pletcher BA, Olson LM. Pediatric resident debt and career intentions. Pediatrics. 2013;131:312–318. 13. Frintner MP, Cull WL. Pediatric training and career intentions, 20032009. Pediatrics. 2012;129:522–528. 14. Cull WL, Caspary GL, Olson LM. Many pediatric residents seek and obtain part-time positions. Pediatrics. 2008;121:276–281. 15. Brotherton SE, Etzel SI. Graduate medical education, 2011-2012. JAMA. 2012;308:2264–2279. 16. Brotherton SE, Etzel SI. Graduate medical education, 2010-2011. JAMA. 2011;306:1015–1030. 17. American Board of Pediatrics. Workforce Data 2012-2013. Available at: https://www.abp.org/abpwebsite/stats/wrkfrc/workforcebook.pdf. Accessed January 9, 2013. 18. Ramanan RA, Taylor WC, Davis RB, Phillips RS. Mentoring matters. J Gen Int Med. 2006;21:340–345. 19. Rubeck RF, Donnelly MB, Jarecky RM, et al. Demographic, educational, and psychosocial factors influencing the choices of primary care and academic medical careers. Acad Med. 1995;70:318–320. 20. Wood PS, Altmaier EM, Franken EA, et al. Factors influencing choice of academic or practice careers in radiology. Invest Radiol. 1990;25: 675–677. 21. Sanders AB, Fulginiti JV, Witzke DB, Bangs KA. Characteristics influencing career decisions of academic and nonacademic emergency physicians. Ann Emerg Med. 1994;23:81–87. 22. Bilbey J, Fache J, Burhenne H. Are there predictors for future academic radiologists? A Canadian survey. Can Assoc Radiol J. 1992; 43:369–373. 23. Althouse LA, Stockman JA. The pediatric workforce: an update on general pediatrics and pediatric subspecialties workforce data from the American Board of Pediatrics. J Pediatr. 2011;159:1036–1040.e3. 24. Children’s Hospital Association. Pediatric specialist physician shortages affect access to care. Available at: http://www.childrenshospitals.net/ AM/Template.cfm?Template¼/CM/ContentDisplay.cfm&ContentID¼ 63293. Accessed October 16, 2013. 25. 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;92:436–441. 26. Cameron C, Collie CL, Baldwin CD, et al. The development of scientific communication skills: a qualitative study of the perceptions of trainees and their mentors. Acad Med. 2013;88:1499–1506. 27. Davis OC, Nakamura J. A proposed model for an optimal mentoring environment for medical residents: a literature review. Acad Med. 2010;85:1060–1066. 28. Levy BD, Katz JT, Wolf MA, et al. An initiative in mentoring to promote residents’ and faculty members’ careers. Acad Med. 2004;79: 845–850. 29. Cain JM, Schulkin J, Parisi V, et al. Effects of perceptions and mentorship on pursuing a career in academic medicine in obstetrics and gynecology. Acad Med. 2001;76:628–634. 30. O’Sullivan PS, Niehaus B, Lockspeiser TM, Irby DM. Becoming an academic doctor: perceptions of scholarly careers. Med Educ. 2009; 43:335–341. 31. Straus SE, Straus C, Tzanetos K. Career choice in academic medicine: systematic review. J Gen Int Med. 2006;21:1222–1229. 32. American Academy of Pediatrics. Annual survey of graduating residents. Available at: http://www.aap.org/research/graduatingressurvey. htm. Accessed October 16, 2013.
