Secondary Logo

Journal Logo

Research Report

Gender Differences Among Physician–Scientists in Self-Assessed Abilities to Perform Clinical Research

Bakken, Lori L. PhD; Sheridan, Jennifer PhD; Carnes, Molly MD, MS

Author Information

Abstract

In 1998, Nathan and colleagues reported that the proportion of physicians who apply for clinical research grants from the National Institutes of Health had declined from 40% to 25% over the previous three decades, threatening the survival of the clinician–scientist and the development of new treatments and therapies to treat human disease.1,2 This trend has occurred as the proportion of women entering medical school and holding faculty appointments has increased.3,4 Lack of appropriate training programs, inappropriate timing of training, and insufficient protected time discourages physicians, especially women, from careers in clinical research.1,2 Several national organizations have acknowledged these limitations in the academic career development of women physicians as problems requiring attention.5–16

For years, self-efficacy has been studied and found to be an important variable in learning and career development.17–20 More recently, gender has been reported as a critical factor in self-efficacy related to career choice and decision making.21,22 However, self-efficacy has not been well studied in professional populations and, to the best of our knowledge, has never been studied in clinical researchers.

Between 1999 and 2000, the National Institutes of Health funded 59 academic medical centers to develop clinical-research training programs to increase the number of physician–scientists. With this award, the University of Wisconsin-Madison (UW-Madison) developed the Clinical Investigator Preparatory Program (CIPP). Because nearly half of CIPP trainees are women, in this study we aimed to determine whether gender differences exist in the trainees’ self-assessed abilities (self-efficacy) to perform 35 learning objectives that reflect knowledge, skills, and attitudes necessary to succeed in clinical research. Our research questions in this study were:

1. Are there differences in the way men and women physicians perceive their abilities to conduct clinical research?

2. If gender differences are found, what are the potential implications for the education and career development of successful physician–scientists?

Method

Data-Collection Instrument

Six major clinical research competencies and 35 learning objectives (see List 1) were established for physician–scientists based on previous studies1,23 and on the expert opinion of senior clinical research faculty at UW-Madison. As part of a baseline evaluation,24 each trainee completed a written questionnaire to self-assess his or her ability to perform each of the 35 learning objectives (0 = no ability to 4 = high ability). Each respondent’s gender and postgraduate year were also collected using this instrument.

List 1 The Six Major Clinical Research Competencies and 35 Learning Objectives Established for Physician–Scientists, University of Wisconsin-Madison
List 1 The Six Major Clinical Research Competencies and 35 Learning Objectives Established for Physician–Scientists, University of Wisconsin-Madison:
List 1 The Six Major Clinical Research Competencies and 35 Learning Objectives Established for Physician–Scientists, University of Wisconsin-Madison

Study Population and Design

Seventeen physician postgraduate trainees who entered the CIPP two-year didactic training program between December 2000 and April 2002, and 40 physician postgraduate trainees who participated in an intensive four-day introductory clinical research workshop in July 2001 or 2002, completed the CIPP self-assessment before commencing their training. Twenty-eight of the 57 study participants (49%) were women. The workshop was conducted at a privately owned center operated by a nonprofit corporation that supports the activities of the UW-Madison Business School and other campus functions. It consisted of 27 didactic sessions (30–75 minutes each) that covered biostatistics, study design, scientific writing, bioethics, research management, the responsible conduct of research, and scientific presentation. The 40 trainees (20 women, 20 men) who participated in the workshop repeated the self-assessment after the workshop.

We excluded from our analyses trainees who were in years one or two of postgraduate training (medical residents, n = 11), participants with titles (e.g., research manager, research coordinator, clinical instructor, or professor; n = 41), and postgraduate trainees who attended the workshops but did not complete both a preworkshop and postworkshop self-assessment questionnaire (n = 17).

Data Collection and Analysis

We collected responses using SurveyPro 2.0 (Apian Software) and performed the statistical analyses using SPSS statistical software (version 11.5). The mean and standard deviation were calculated for each objective after testing for normality. Bivariate comparisons were determined by independent sample t tests on the ratings for each objective answered by female and male respondents. For those objectives with statistically significant mean differences (α = .05), a posthoc univariant analysis of covariance with each objective as the dependent variable, gender as an independent variable, and postgraduate year as a covariant was performed to control for year in postgraduate training because year in postgraduate training may affect the trainee’s perceived ability to perform the 35 clinical research learning objectives.1, pp 161–164 A statistically significant difference was accepted at the p < .05 level, adjusting for multiple comparisons using a Bonferroni correction.

Results

Table 1 summarizes the trainee sample by gender and postgraduate year for all 57 physician trainees and for the 40 physicians who attended the training workshops. In the self-assessment before training, women rated their ability to perform 22 of the 35 learning objectives lower than men rated their own abilities on these same objectives (see Figure 1). This difference was statistically significant (F = 3.97, p = .05, 95% CI = −0.886 to −0.003) for the item, “Spend sufficient time developing and advancing one’s own area of scientific knowledge and research,” after controlling for postgraduate year as a covariant. Nearly identical results were obtained for the 40 trainees who completed the assessment before attending a workshop (data not shown). On 13 objectives, women rated their abilities the same as or higher than men rated their own abilities; however, none of these differences were statistically significant.

Table 1
Table 1:
Number of Postgraduate MD Trainees Who Completed the Self-Assessment of Their Clinical Research Skills, by Gender and Postgraduate Year (PGY), University of Wisconsin-Madison, 2000–2002
Figure 1.
Figure 1.:
Mean difference in ratings between women and men for each learning objective before the clinical research workshop. Fifty-seven physician postgraduate trainees at the University of Wisconsin-Madison who entered the Clinical Investigator Preparatory Program (CIPP) between December 2000 and April 2002 completed a self-assessment of clinical skills before commencing training. Numbers above and below bars refer to learning objectives (see List 1). *The gender difference in the rating of objective 22, “Spend sufficient time developing one’s own research,” was statistically significant (p = .05) after controlling for postgraduate year as a covariant.

Self-ratings on 34 of the 35 objectives for women and 35 of 35 objectives for men increased for the 40 trainees who attended the four-day workshop. However, the pretest–posttest differences were slightly greater for men (mean difference = 1.0, SD = .83) than for women (mean difference = 0.8, SD = .85). After the workshop, women (n = 20) rated themselves lower than men rated themselves (n = 20) on 33 of 35 objectives (see Figure 2). This difference was significant (p < .05) for seven objectives as shown in Table 2. After taking the workshop, the difference between women and men in their perceived ability to “[Spend] sufficient time developing and advancing one’s own area of scientific knowledge and research” actually increased (F = 6.69, p = .01, 95% CI = −1.163 to −0.141), although the 95% confidence intervals for the two means slightly overlap. Women showed significantly less confidence than men did in one other objective related to time: “Plan and adhere to a timeline for research projects” (F = 6.71, p = .01, 95% CI = −1.097 to −0.133). We found statistically significant differences between women and men in the trainees’ abilities to design research projects, report research results, hire research staff, present research findings, and disseminate their knowledge through teaching (see Table 2). The only item on which women scored higher than men did was “Apply rules of English usage, style and composition,” although this difference was not statistically significant at the p < .05 level.

Table 2
Table 2:
Learning Objectives That Showed Statistically Significant Mean Differences Between Ratings of 40 Men and Women Who Attended a Clinical-Research Training Workshop, University of Wisconsin-Madison, 2000–2002
Figure 2.
Figure 2.:
Mean difference in ratings between women and men for each learning objective after the clinical research workshop. Forty of the 57 Clinical Investigator Preparatory Program trainees in this study participated in an intensive four-day introductory clinical research workshop in July 2001 or 2002 and repeated the self-assessment after the workshop. Numbers above and below bars refer to learning objectives (see List 1). *Significant (p < .05) gender differences existed in the rating of seven objectives (4, 15, 18, 22, 24, 26, and 32) after correcting for postgraduate year.

Discussion

Although a number of potential barriers to women’s success in academic careers have been described,6,8,16,25–29 ours is the first study to find a gender difference in self-assessed competency (self-efficacy) among physicians in the knowledge and skills required to become a clinical investigator. Previous studies have shown that gender differences exist in self-assessed abilities to perform various tasks30–33and that self-efficacy is an important factor in determining career decisions, persistence in a career, and one’s perceived abilities to overcome barriers to career development.17–19

The clinical research workshop at the University of Wisconsin-Madison was designed to promote interest in clinical research careers. Self-efficacy increased for both men and women after the workshop, a factor important for sustaining career interest.20 A surprising finding of our study is that while both women and men rated themselves higher on the objectives after the workshop than they did before their training, the gender differences were more pronounced after the workshop. There are several possible explanations for our findings. Previous studies have shown that while men tend to overrate their abilities to perform various tasks, women tend to underrate them.31 Furthermore, men may resist acknowledging their weaknesses owing to fears of threatened self-identity because work is so strongly related to their self-identity.34

Our findings might indicate that the workshop’s format better serves men’s educational needs than women’s. Sullivan35 reported that a career development intervention incorporating performance accomplishments, explicit experiences, emotional excitement, and verbal persuasion increased career decision making self-efficacy and commitment in 61 college-age women. Hugo36 and Mott27 recommend that teaching practices be adapted to promote women’s thinking by including experiential forms of learning, such as modeling, apprenticeship, and teamwork. At our institution, trainees are exposed to very few women physicians performing scientific research. This paucity of women role models may contribute to lower self-efficacy for women in demonstrating laboratory skills. In the workshop, none of the principal investigator speakers were women. This factor may have contributed to women’s lower self-efficacy in their abilities to give oral research presentations.

Many women investigators assume family responsibilities coincident with research career development,29,37 which may explain the significantly lower self-ratings reported by women for two objectives. One objective that was rated significantly lower for women than for men before the workshop and became more significant after the workshop was spending sufficient time developing and advancing one’s expertise and research. Furthermore, time may be a factor that contributes to the second objective found to be statistically significant after the workshop: “plan and adhere to a timeline.…” The family–professional role conflict that many women face combined with an absence of professional role models in our training program suggests important ways to focus program planning and future research. Plans are underway to predominantly use women presenters at a future workshop to determine how this influences responses in this assessment. Qualitative studies could also help unveil variables that influence existing barriers, such as time, for both women and men developing clinical research careers.

Our study had several limitations. Our findings are limited to one institution and a predominately European-American population. The distribution of men and women at different levels of training varied, although our analyses were controlled for postgraduate year. Flores and O’Brien21 and Luzzo and McWhirter22 reported that women of ethnic minorities tend to perceive more barriers to career development; therefore, they have lower self-efficacy compared with European-American women. Additional research is needed to extend the generalizability of our findings and further explore the role of self-efficacy in physicians’ career selection. Plans are in progress to expand our studies to include clinical researchers from other institutions, particularly those that have larger populations of underrepresented minorities.

In summary, a previously unexplored gender difference in self-perceived competency (self-efficacy) may indicate an additional barrier women face in academic career development. Given the large number of women entering medical schools, educational program planning models attending to gender differences in adult learning and career development should be investigated as an effective strategy to increase the number of physician–scientists.

References

1.Kelley WN, Randolph MA (eds). Careers in Clinical Research: Obstacles and Opportunities. Washington, DC: National Academy Press, 1994.
2.Nathan D. Clinical research: perceptions, reality, and proposed solutions. JAMA. 1998;280:1427–31.
3.Medical school faculty by rank & gender. AAMC Faculty Roster System 〈http://www.aamc.org/findinfo/infores/datars/facros/frspubs/usmsf.htm〉. Accessed 20 July 2002.
4.Women applicants, enrollees & graduates—selected years: 1949–1950 through 2000–2001. AAMC Section for Student Services, 2001 〈http://www.aamc.org/members/wim/resources.htm〉. Accessed 20 July 2002.
5.National Science Foundation. Women, Minorities, and Persons With Disabilities in Science and Engineering. Arlington, VA: 2000. Report No. NSF 00–327.
6.U.S. Department of Health and Human Services, Public Health Service, National Institutes of Health. Women in Biomedical Careers: Dynamics of Change, Strategies for the 21st Century (Full Report). Washington, DC: 1995. DHHS Publication No. 95-3565.
7.Increasing Women’s Leadership in Academic Medicine. Washington, DC: Association of American Medical Colleges, 1998.
8.Department of Health and Human Services. Council on Graduate Medical Education: Fifth Report. Women and Medicine. Washington, DC: Public Health Services Health Resources and Services Administration, 1995.
9.United States Public Health Service Office on Women’s Health. Action Plan for Women’s Health. Washington, DC: Department of Health and Social Services, 1991. Report No. 91–50214.
10.NIH Director’s Panel on Clinical Research. Report presented to the Advisory Committee to the Director of the NIH. Washington, DC: National Institutes of Health; December 8, 1997.
11.The American College of Physicians. Comprehensive Women’s Health Care: the role and commitment of Internal Medicine. Am J Med. 1997;103:451–7.
12.Rosenberg L. Physician-Scientists: endangered and essential. Science. 1999;283:332.
13.Cadman E. The academic physician-investigator: a crisis not to be ignored. Ann Intern Med. 1994;120:410.
14.Schrier R. Ensuring the survival of the clinician-scientist. Acad Med. 1997;72:589–94.
15.Martin J. Academic health centers: vulnerabilities in 1995 and beyond. Arch Intern Med. 1995;155:534–42.
16.Carnes M, Vandenbosche G, Agatisa P, Hirshfield A, Dan A, Shaver J. Using women’s health research to develop women leaders in academic health sciences: the national centers of excellence in women’s health. J Womens Health Gend Based Med. 2000;10:39–47.
17.Lent R, Brown S, Hackett G. Toward a unifying social cognitive theory of career and academic interest, choice, and performance. J Vocat Behav. 1994;45:79–122.
18.Lent R, Brown S, Hackett G. Contextual supports and barriers to career choice: a social cognitive analysis. J Couns Psychol. 2000;47:36–49.
19.Zimmerman B. Self-efficacy and educational development. In: Bandura A (ed). Self-efficacy in Changing Societies. 1st ed. Cambridge, UK: Cambridge University Press, 1995:202–31.
20.Nauta M, Kahn J, Angell J, Cantarelli E. Identifying the antecedent in the relation between career interests and self-efficacy: is it one, the other, or both? J Couns Psychol. 2002;49:290–301.
21.Flores L, O’Brien K. The career development of Mexican American adolescent women: a test of social cognitive career theory. J Couns Psychol. 2002;49:14–27.
22.Luzzo D, McWhirter E. Sex and ethnic differences in the perception of educational and career-related barriers and levels of coping efficacy. J Couns Dev. 2001;79:61–7.
23.Bland C, Schmitz C, Stritter F, Henry R, Aluise J. Successful Faculty in Academic Medicine: Essential Skills and How to Acquire Them. New York: Springer Publishing Company, 1990.
24.Bakken L. An evaluation plan to assess the process and outcomes of a learner-centered training program for clinical research. Med Teach. 2002;24:162–8.
25.Valian V. Women in academia. In: Why So Slow? The Advancement of Women. Cambridge, Mass.: MIT Press, 2000:217–50.
26.Bierema L. Women, work, and learning. New Dir Adult Contin Educ. 2001;92:53–62.
27.Mott V. Women’s career development in midlife and beyond. In: Bierema L (ed). Women’s Career Development Across the Lifespan: Insights and Strategies for Women, Organizations, and Adult Educators. 80th ed. San Francisco: Jossey-Bass Publishers, 1998:25–34.
28.Inman P. Women’s career development at the glass ceiling. In: Bierema L (ed). Women’s Career Development Across the Lifespan: Insights and Strategies for Women, Organizations, and Adult Educators. 80th ed. San Francisco: Jossey-Bass Publishers; 1998:35–42.
29.Stalker J. Athene in academe: women mentoring women in the academy. Int J Lifelong Educ. 1994;13:361–72.
30.Beyer S. Effects of gender and depression on self-evaluations of performance on academic tasks. Paper presented at the annual meeting of the American Psychological Society, Washington, DC, July 1994.
31.Beyer S. Self-consistency and gender differences in the accuracy of self-evaluations. Paper presented at the annual meeting of the American Psychological Society, Washington, DC, August 14–18, 1992.
32.Hackett G, Campbell N. Task self-efficacy and task interest as a function of performance on a gender-neutral task. J Vocat Behav. 1987;30:203–15.
33.Campbell N, Hackett G. The effects of mathematics task performance on math self-efficacy and task interest. J Vocat Behav. 1986;28:149–62.
34.Valian V. Effects on the self. In: Why So Slow? Cambridge, Mass.: The MIT Press, 2000:152–8.
35.Sullivan K, Mahalik J. Increasing career self-efficacy for women: evaluating a group intervention. J Couns Dev. 2000;78:54–62.
36.Hayes E, Flannery D, Brooks A, Tisdell E, Hugo J. Women as Learners. San Francisco: Jossey-Bass Publishers, 2000.
37.Foster S, McMurray J, Linzer M, Leavitt J, Rosenburg M, Carnes M. Results of a gender climate survey from a midwest academic medical school. Acad Med. 2000;75:653–60.
© 2003 Association of American Medical Colleges