Research during Pediatric Residency: Predictors and Resident-Determined Influences

Ullrich, Nicole MD, PhD; Botelho, Christine A. MPH; Hibberd, Patricia MD, PhD; Bernstein, Henry H. DO

Academic Medicine:
Research Report

Purpose: Participation in research during residency is thought to be a strong predictor of future research activity; however, the proportion of residents who actually engage in research is small. This study examined (1) which factors are associated with research during residency; (2) which factors influence residents’ abilities to conduct research; and (3) the number of residents conducting research in a research-oriented training program.

Method: One hundred fifteen pediatrics residents were asked at a housestaff retreat to complete a questionnaire about their attitudes toward research. Comparisons between those who were or were not conducting research during residency were made using chi-square or Fisher exact tests; stepwise logistic regression was used to identify factors associated with conducting research during residency.

Results: The response rate was 95% (n = 110, or 82% of the residency program). Respondents were representative of postgraduate year, gender, and residency track. Although 92% of respondents reported having conducted research before residency, only 18% were currently involved in research (p < .001). Fifty-five percent reported interest in conducting research. Advanced degrees and future career plans influenced their decisions to do research (p < .05). Respondents were more likely to conduct clinical research than basic science or laboratory-based research (14% versus 3% of all respondents, p = .007). The most commonly identified influences to conducting research were availability of time (97%), personal interest in research (84%), availability of opportunities (76%), and mentors on hand (72%).

Conclusion: Interest in research during residency is high, but participation in research is low. There are several influences to the types and amounts of research conducted during pediatrics residency.

Author Information

Dr. Ullrich is clinical fellow in neurology and Dr. Bernstein is associate chief, general pediatrics, both at Children's Hospital Boston, Massachusetts; Dr. Hibberd is director and Ms. Botehlo is project editor, both at Clinical Research Institute at Tufts-New England Medical Center, Boston, Massachusetts.

Correspondence and requests for reprints should be addressed to Nicole Ullrich, MD, PhD, Department of Neurology, Children's Hospital, 300 Longwood Avenue, Boston, MA 02115; phone: (617) 355-6363; fax: (617) 277-0708; e-mail: 〈〉.

Article Outline

Innovations in genomics and molecular biology will lead to new approaches in disease prevention and treatment, but the pace of future research will depend on the ability to attract young physicians and scientists into the basic science laboratory or into a career in clinical research. Approximately one fifth of all pediatrics physicians are involved in academic careers, yet less than 20% of their time is devoted to teaching and research.1,2 Of the 128 pediatrics departments in the United States, 37.5% have one or more National Institutes of Health (NIH) training grant(s).3 At a national level, children's hospitals have received slightly more than one third of all NIH awards granted for pediatrics research and research training, yet pediatrics fellows and research trainees have received proportionately fewer grants than have fellows in other departments.4 More pediatricians and support are needed to conduct research.

To date, research during residency is suggested but not mandated by the Pediatrics Residency Review Committee. On a national level, 38% of all general residency programs that accept postgraduate first-year (PGY-1) residents require research projects, compared with 68% of subspecialty residency programs.5 Research projects can include presentations in a departmental meeting or submission of an abstract or manuscript for publication.6 Although few pediatrics training programs make research mandatory, some offer residents elective time that may be designated for research.7,8 Because, to our knowledge, no uniform recommendation or definition of research exists for general residency programs, the exact nature of the research programs offered and their implementation is varied and open to interpretation. Research is important for the future growth and development of the field of pediatrics, and research during residency fosters an interest in an academic career. Improved mentoring of research, teaching, and clinical experience is necessary to stimulate and improve research during training.8–10

We conducted this study to determine (1) which factors are associated with research during residency; (2) which influence residents’ abilities to conduct research; and (3) the number of residents conducting research at our research-oriented training program.

Back to Top | Article Outline


A 20-item questionnaire was developed using both focus groups and informal discussions among residents and faculty at Children's Hospital Boston. The survey was distributed to 115 of 134 current pediatrics residents in the Boston Combined Residency Program in Pediatrics who attended either the mandatory 1999–2000 general medical housestaff or the medicine/pediatrics housestaff retreat on October 5 or November 12, 1999, respectively. Interns were included in the study because, by late fall, we believed that they had enough experience to gauge the importance and role of research in their professional training. The principal investigator explained the purpose of the study to the residents and invited them to participate. The questionnaire was self-administered, confidentiality of respondents was protected, and no identifying information was requested. The institutional review board of Children's Hospital Boston approved this study.

For the purposes of this study, we asked all residents to identify themselves as being in the primary care or categorical track. Questions were organized into four sections: demographic data, perceptions about research, previous research experience, and current research experience. Analyses were directed at identifying which demographic features might predict research activity. To determine motivations for involvement in research, we asked participants to describe the reasons they became involved in the research project(s) before residency. Residents used five-point Likert scales to rate the following influences on their ability to conduct research: time availability, interest in research topic, opportunities available, mentor availability, faculty interest in research topic, space, hardware/software/supplies, and funding. We asked residents who were currently involved in research projects the type of research, responsibilities they had, and importance of the aforementioned influences to their conducting research.

Back to Top | Article Outline
Statistical Analysis

We performed analyses on all respondents rather than the entire cohort of housestaff in our training program. We made this conservative decision to avoid introducing systematic bias by using imputation for the 24 residents who did not complete the survey (i.e., nonrespondents and nonattendees). In fact, we expect the percentages we report would be even lower if the entire cohort was used, further emphasizing the need to entice physicians into research careers. Responses were entered into a database, and data were then compared item by item using SPSS version 10.0. Discrepancies were resolved by referring to the original questionnaire. Data entry was repeated on 10% of the responses selected by random-number generation, and the accuracy of data entry was found to be greater than 98%. Cleaned data were converted to SPSS format using DBMS Copy Version 7. We prepared descriptive statistics (proportions and confidence intervals for categorical data and means and standard deviations for normally distributed data) for all questions. Chi-square or Fisher exact tests were used to compare the responses of residents who were or were not conducting research during residency. Comparisons between doing research before or during residency were made using the McNemar test. We used stepwise logistic regression to determine which factors were associated with participation in research during residency. A .05 two-sided significance level was used for all analyses.

Back to Top | Article Outline


General Characteristics of the Respondents

A total of 110 of 115 residents responded (95% response rate, representing 82% of our training program). The percentages of respondents in each postgraduate year, track, and gender are shown in Table 1. Respondents were almost equally divided by year of residency; most were women (66%), and most were in the categorical track (59%).

Of the 24 residents who did not complete the questionnaire, 19 did not attend the retreat and five attended but did not respond. Overall, nonresponders were two postgraduate first-year residents, 12 postgraduate second-year residents, and ten residents from third year and beyond. There were 17 women and seven men, which reflects the overall gender distribution of the program.

Back to Top | Article Outline
Predictors of Research Activity

A total of 101 residents (92% of respondents) reported participation in a research project before they entered residency, but only 20 (18%) were currently involved in research (p < .001). Involvement in research was individually defined and not verified. A majority of respondents (60%) documented their interest in participating in research during residency. Participation in research increased by year of residency (7% of PGY-1 residents, 21% of postgraduate second-year residents, 30% of postgraduate third-year residents; p = .038). Residents in the categorical track were more likely to be involved in research than were residents in the primary care track (27% versus 7%, p = .009). In addition, those residents with graduate degrees other than doctor of medicine (e.g., doctor of philosophy, master of public health, or master of science) were more likely to participate in research during residency (35% versus 13%, p = .015). Thirty-one percent of men and 13% of women were currently doing research (p = .023). Using a stepwise logistic regression, we determined the ability of each characteristic to independently predict the level of research ability. Only level of postgraduate training and future career plans remained simultaneously and independently significant. Year of residency, current residency track, and gender were not significant.

Among all respondents, 39 (36%) planned to enter clinical practice after residency; interestingly, only two (5%) residents in this subgroup reported current involvement in research. In contrast, 53% of all respondents planned to begin fellowship or assume other academic positions; 14 in this subgroup (24%) reported current involvement in research. Of the residents who were undecided about their future plans (11% overall), none reported current involvement in research. Even though only a minority of our residents were involved in research, 70% included research in their future plans.

Back to Top | Article Outline
Perceived Influences on Research during Residency

Ninety-seven percent of the respondents strongly agreed or agreed that time availability was an influence to conducting research during residency. Personal interest in research topic (84%), opportunities available (76%), mentors available (73%), funding (63%), faculty interest in research topic (59%), equipment (44%), and space (40%) were also important influences (see Figure 1).

Of the subgroup of residents who reported involvement in a research project before entering residency programs, two thirds conducted research as undergraduates and/or during graduate or medical school and 32% did so as part of a job. Table 2 depicts the overall percentages of residents currently conducting research, and those who had done research in the past. More residents performed basic science or laboratory-based research rather than clinical research before residency (67% versus 42% of all respondents, p = .01). During residency, respondents were more likely to conduct clinical research rather than basic science or laboratory-based research (14% versus 3% of all respondents, p = .007). Residents also reported significantly less involvement in literature reviews (p = .007), case series/case reports (p = .004), and chart reviews (p = .004) during residency than before. Residents’ greatest research responsibilities related to data collection, whereas the fewest responsibilities were related to data management. There was no significant difference in research focus between residents with MD degrees only and those with additional graduate degree(s).

Back to Top | Article Outline


The 1998 NIH task force recognized that physician–investigators are becoming “endangered” and made several recommendations to reverse this trend.3 The residents’ disparate exposure to research by residency programs was of particular concern, largely because research during residency is indicative of research later in one's career.11,3 The Federation of Pediatric Organizations has set goals for fellowship education in pediatrics that include proficiency in research; commitment to attainment of research training and a career in academic pediatrics; and training in experimental design, statistical analysis, scientific writing, biomedical ethics, and preparation of grants.12 These goals reflect the Federation's high regard for prior research (i.e., during residency), attainment of advanced degree, and publication of research manuscripts.

This is the first study to document the attitudes of current residents, and the results are strengthened by a high response rate. In addition, we gathered prospective information regarding residents’ future plans that can be directly compared with their actual career plans upon completion of training. Our study is unique because we used a prospective, self-report, cross-sectional design to collect data about residents’ involvement in research during their training at a research-oriented pediatrics program in a freestanding children's hospital. In contrast, previous studies of research during pediatrics residencies and during other disciplines have been retrospective, either as a survey of residency program directors,6 physicians in practice,2,13 or through identification of resident publications and ultimate career choice.4,6,11,14,15

We found the strongest predictor of research in residency was having one or more additional degree(s), such as a PhD, MPH, or MS. Residents with an interest in research may be more likely to have completed an additional degree to further their research experience. Although a majority of our study's cohort expressed interest in research and included it in future career plans, only a minority of these residents are currently conducting research. Self-reporting is more likely to result in an overestimation of residents’ genuine interest in performing or reporting of research. Therefore, there is an apparent discrepancy between the desire to perform research and in its actual execution.

Ongoing experiences are likely to influence residents’ ultimate career goals and future plans. One retrospective study revealed that 77% of graduates from a pediatrics residency program reported that their past career goals resemble their current positions, suggesting that residents’ career choices reflect their past experiences.16 In the last three years, two-thirds of graduating residents from our program have chosen academic career paths, irrespective of academic track. These results confirm that participation in research during residency is a strong predictor of both career choice and future research activity. This finding is substantiated by an earlier study of academically trained pediatricians, which showed that having a publication before completing residency was the most important correlate with subsequent research productivity.14 For some residents, participation in research prepares them for an academic career, as is true within other medical disciplines.17–22 Residents who defer research training tend to do so until fellowship training.

Our study was limited to a single academic training institution, which did not mirror the national population's primary care focus (41% versus 69% nationally).15 Selection bias might be a concern, therefore, because all residents were not included in the sample; anonymity precluded tracking nonrespondents or residents who did not attend the retreat.

The demands of graduate medical education have an impact on residents’ abilities and potential to participate in research.8,9,23–29 In addition to various environmental factors, residents who were interested in research identified a lack of mentors and funding as the primary reasons for not being currently involved. Their perceptions about a lack of mentors and funding were somewhat surprising, and likely do not reflect the true availability and proximity of these resources. The total value of research grants received at Children's Hospital Boston in the fiscal year 2001 was nearly $80 million. One would expect that, given this high level of research funding and a reasonably supportive research environment, there would be a greater level of participation in research by pediatrics residents compared with institutions with less of a research focus. Our finding suggests that improved access to mentors and other resources for residents is necessary to conduct research.

One study of pediatrics residency programs concluded that residents have not taken full advantage of the research opportunities available.6 The study identified influences consistent with those that had been identified elsewhere: guidance with research, teaching, and clinical experience by mentors.6,8–10,16,22–24,30 Analyses of other successful techniques encouraging research productivity have been described in fields outside of pediatrics.9,23,27–29,31,32 These studies found that the most important factors leading to the success of research projects were protected time, staff support, and the development of clear goals and a curriculum. These findings are consistent with our own, which found time, availability of mentors, and opportunities available to be important influences on research participation.

The NIH budget for FY2003 is projected to double, increasing opportunities for clinical research. Historically, basic science projects are favored by funding agencies and academic institutions.25 The Child Health Act of 2000 has called attention to the value in clinical investigation and an increase in funding for pediatrics. The renewed push to increase interest, involvement, and funding in pediatrics research has renewed recognition of the importance of research during clinical training. In addition, there is a shift in attitude toward dedicated time for clinical and basic science research during residency and fellowship training. The American Board of Pediatrics has plans to allow residents to pursue research interests during residency.

Given the demands of residency education, resources promoting research must be widely publicized and easily attained. For example, establishing a Web-based central resource of available mentors, funding, and projects would make the information more accessible. Establishing journal clubs to discuss basic and clinical research and forums for presenting residents’ research will increase the visibility and, ideally, emphasize the importance of research. In addition, increasing funding opportunities and resources for attending academic conferences would allow residents to increase their exposure to research on a national level. Lastly, formal training in research may encourage and facilitate participation in research during residency. Research during residency may not only foster interest in an academic career but may also make residents more competitive for future fellowships, other academic positions, and office-based research.

Back to Top | Article Outline


We found a high level of interest in research in future research plans by the current pediatrics residents in our research-oriented training program; however, their current participation in research activities is low. The residents perceived time, interest, opportunities available, and mentors on hand to be necessary for successful research. This study can be the impetus for other residency programs to identify those trainees most likely to conduct research and address influences on participation in research during residency.

The authors thank Drs. Frederick Lovejoy, David Greenes, and Edward Rothstein for their thought-ful reviews of the manuscript, the housestaff at the Boston Combined Residency Program in Pediat-rics for their participation, and Victoria Floriani, Sarah Rieber, Lori Rutman, and Rebecca Stoltz for their meticulous attention to detail and edito-rial assistance.

Back to Top | Article Outline


1.Brotherton SE, Tang SS, O'Connor KG. Trends in practice characteristics: analyses of 19 periodic surveys (1987–1992) of fellows of the American Academy of Pediatrics. Pediatrics. 1997;100:8–18.
2.Walker WA. A subspecialist's view of training and pediatric practice in the next millennium. Pediatrics. 1998;102:636–43.
3.Hershberg RI, Harper D, Leichner P. Research in psychiatry: residents’ attitudes and use. J Med Educ. 1980;55:540–1.
4.Kelch RP, Novello AC. Training pediatric scientists. Pediatr Res. 1989;25:1–5.
5.Brotherton SE, Simon FA, Etzel SI. US graduate medical education, 2000–2001. JAMA. 2001;286:1056–60.
6.Alguire PC, Anderson WA, Albrecht RR, Poland GA. Resident research in internal medicine training programs. Ann Intern Med. 1996;124:321–8.
7.Brouhard BH, Doyle W, Aceves J, McHugh MJ. Research in pediatric residency programs. Pediatrics. 1996;97:71–3.
8.Winter RJ. Research during residency. Am J Dis Child. 1989;143:521.
9.Neale AV, West P, French L. Surviving your resident research requirement. JAMA. 1998;280:1802.
10.Taniguchi MH, Johnson PD. Rehabilitation resident academic productivity: report on 1993 graduates. Am J Phys Med Rehabil. 1994;73:240–4.
11.Ledley FD, Lovejoy FJ. Factors influencing the interests, career paths, and research activities of recent graduates from an academic, pediatric residency program. Pediatrics. 1993;92:436–41.
12.Federation of Pediatric Organizations. Statement on fellowship training. Pediatrics. 1991;87:265.
13.Liebelt EL, Daniels SR, Farrell MK, Myers MG. Evaluation of pediatric training by the alumni of a residency program. Pediatrics. 1993;91:360–4.
14.Ledley FD, Lovejoy FJ. Prospects for academically trained pediatricians in academic medicine. Clin Invest Med. 1992;15:518–26.
15.Pan RJ, Cull WL, Brotherton SE. Pediatric residents’ career intentions: data from the leading edge of the pediatrician workforce. Pediatrics. 2002;109:182–8.
16.DeAngelis C, Duggan A, Oski F. Twenty-five years of residents: what and where are they now. Pediatrics. 1990;85:10–6.
17.Sanders AB, Fulginiti JV, Witzke DB, Bangs KA. Characteristics influencing career decisions of academic and nonacademic emergency physicians. Ann Emerg Med. 1994;23:81–7.
18.Hirthler MA, Glick PA, Hassett JM, et al. Comparative analysis of successful and unsuccessful candidates for the pediatric surgical matching program. J Pediatr Surg. 1992;27:142–8.
19.Souba WW, Gamelli RL, Lorber MI, et al. Strategies for success in academic surgery. Surgery. 1995;117:90–5.
20.Morrison JC, Meeks JR, Martin JN, Cowan BD, Whitworth NS, Wiser WL. Resident research projects: frequency of presentation and publication in a national forum. Am J Obstet Gynecol. 1994;170:777–81.
21.Hillman BJ, Fajardo LL, Witzke DB, Cardenas D, Irion M, Fulginiti JB. Factors influencing radiologists to choose research careers. Invest Radiol. 1989;24:842–8.
22.Jones J, Dougherty J, Cannon L, Scheible D. Teaching research in the emergency medicine residency curriculum. Ann Emerg Med. 1987;16:247–353.
23.DeHaven MJ, Wilson GR, O'Connor- Kettlestrings P. Creating a research culture: what we can learn from residencies that are successful in research. Fam Med. 1998;30:501–7.
24.Gill S, Levin A, Djurdjev O, Yoshida EM. Obstacles to residents conducting research and predictors of publication. Acad Med. 2001;76:477.
25.Chan JC, Roth KS, Salusky IB. Clinical research and training; an overview. J Pediatr. 2002;140:293–8.
26.Webb JM, Rye B, Fox L. State of dermatology training. The residents’ perspective. J Am Acad Dermatol. 1996;34:1067–71.
27.Kirsner RS, Kerdel FA, Falanga V, Trent J, Eaglstein WH. The role of mandated research during dermatology residency training. J Invest Dermatol. 1993;101:255–6.
28.Blake DJ, Lezotte MC, Yablon S, Rondinelli RD. Structure research training in residency training programs: the impact on the level of resident research activity. Am J Phys Med Rehabil. 1994;73:245–50.
29.Schultz JH. Research during internal medicine residency training: meeting the challenge of the residency review committee. Ann Intern Med. 1996;124:340–2.
30.Temte JL, Hunter PH, Beasley JW. Factors associated with research interest and activity during family practice residency. Fam Med. 1994;26:93–7.
31.Salive ME, Stein DH. Predictors and correlates of prevention research careers: a national study of residency graduates. Prev Med. 1999;28:430–6.
32.Crooks KA, Pato MT, Berger RC. A curriculum for producing resident researchers. Acad Med. 1999;74:614.
© 2003 Association of American Medical Colleges