Secondary Logo

Journal Logo

Original Research Articles

Analysis of H-index in Assessing Gender Differences in Academic Rank and Leadership in Physical Medicine and Rehabilitation in the United States and Canada

Yang, Hsin Yun MSc; Rhee, Gaeun BHSc; Xuan, Lisa BSc; Silver, Julie K. MD; Jalal, Sabeena MD; Khosa, Faisal MD, MBA, FFRRCSI, FRCPC

Author Information
American Journal of Physical Medicine & Rehabilitation: June 2019 - Volume 98 - Issue 6 - p 479-483
doi: 10.1097/PHM.0000000000001129

Abstract

Gender disparities in academic medicine have a long and persistent history. Despite the increasing representation of women physicians in medical schools, disparities can be seen in areas such as promotion, leadership positions, and compensation.1 Based on the 2014 Association of Academic Medical Colleges survey, women comprise close to 50% of the total medical applicant pool, whereas only 38% of the full-time academic medicine faculty is made up of women.2 The underrepresentation of women is more prominent in higher academic ranks, where only 15% and 16% of women hold permanent department chair and dean positions, respectively.2,3

According to the most recent physician gender data reported by the Association of Academic Medical Colleges in the 2014 Physician Specialty Data Book, 34.7% (n = 3083) of active physical medicine and rehabilitation (PM&R) specialty physicians were female.4 The Association of Academic Medical Colleges also reported that full-time female PM&R faculty members reached the 30% mark in 19925 and 41% in 2014.6 Of 49 chair positions available in PM&R departments across medical schools in the United States (US) in 2014, less than 16% belonged to women.2

In the field of PM&R, there have been several reports highlighting gender disparities in the academic physician workforce.7–13 In 2007, Wagner et al.8 from the American Academy of Physical Medicine and Rehabilitation research committee looked at the impact of gender and career development in the field and found that women physiatrists applied for fewer grants, had lower publication rates, and were underrepresented at higher academic ranks and leadership positions when compared with men physiatrists. The authors also found that the women were compensated less. Following this, Bickel9 suggested that this report was a “wake-up call” for rehabilitation medicine and specifically called for more “chair support and accountability.” The same year, Bowles et al.10 evaluated gender differences in attitude, aspiration, and priority of Association of Academic Physiatrists (AAP) and found that women valued scholarly skills, activities, and promotion less than their male counterparts. More recently, Hwang et al.11 studied gender disparity trends among PM&R academic faculty and found persistent underrepresentation for female associate and full professor positions. Although women faculty had increased in proportion at all ranks and in comparison with other specialties such as orthopedics, disparities continue to exist despite PM&R having a narrower gap. For example, studies by Silver et al.7,12 on the recognition awards from the American Academy of Physical Medicine and Rehabilitation and the AAP found that women physicians were underrepresented both historically and recently. Silver et al.13 demonstrated that women physicians received zero or near zero levels of recognition awards in numerous categories for a period of years and then demonstrated in a subsequent report that the marked underrepresentation of recognition awards for women physicians was also found in other medical specialties. A recent report released by the Women in Academic Physiatry task force for the AAP highlighted numerous gaps in the inclusion of women in PM&R including leadership and faculty promotions, recognition awards, plenary speaker invitations, and journal editors.14

To further understand the gap, we decided to evaluate associations between physician gender and research productivity using bibliometric data. Bibliometrics include a group of metrics that provide the ability to conduct a statistical analysis of an academician's journal publications to assess his or her scientific scholarship including productivity and impact.15 Bibliometrics are routinely used in the academic medicine promotions process.15 These metrics include, but are not limited, to citations, impact factor, h-index, e-index, m-index, and Eigenfactor score. The Hirsch index (h-index) is considered a valuable measure of assessing academic productivity and has been studied in various medical and surgical specialties with strong associations between higher h-indices and academic promotion as well as with National Institutes of Health awards.16 Indeed, the h-index has received considerable attention in the published literature.15–18 All bibliometrics have well-documented advantages and limitations, and the h-index is no exception. However, in a recent study focusing on h-index benchmarks for academic psychiatry, the authors explained, “Regardless of any controversies or criticisms of bibliometrics, they are increasingly influencing promotion, merit increases, and grant support.”18 Therefore, in this study, we used the h-index as a metric to consider any workforce gender differences in the process of evaluating potential gender disparities in the PM&R field.

The objective of this study was to build on the current literature and further evaluate the associations between physician gender, leadership positions, and research productivity in academic PM&R faculties across the US and Canada. This report is the first to include the Canadian perspective. Notably, there are reports in other fields that have looked at these relationships. For example, Pashkova et al.19 showed that a higher proportion of leadership positions belong to male anesthesiologists with an associated higher academic and research productivity determined by h-index. Similar leadership and scholarly productivity disparities were present in other medical fields, such as dermatology and radiology.2,20–22

METHODS

An institutional review board approval was not needed for this retrospective study because publicly available data were evaluated. Data were collected from April to June 2017, on all academic and administrative faculty members of PM&R departments across the US and Canada. Different resources were used to generate the database for analysis: The Fellowship and Residency Electronic Interactive Database (FREIDA Online) provided a total of 83 Accreditation Council for Graduate Medical Education accredited PM&R programs of the American Medical Association in the US. Three pediatric programs were also included. For the PM&R programs in Canada, the Canadian Resident Matching Service Web site provided a total of 13 programs. Using the stated resources, a list of PM&R programs that offered faculty listings on their respective Web sites was compiled. Programs that either did not have the faculty listing available or had no mention of the administrative or faculty rankings were excluded. After excluding the programs that did not fulfill the study's selection criteria, 72 programs (69 physiatry and 3 pediatric ones) were included in the US and 9 programs in Canada.

Data collection began by reviewing Web sites of the selected programs for their faculty listing. Inclusion criteria were full-time faculty members with an academic ranking of professor, associate professor, assistant professor, or junior faculty (instructors and lecturers) with a Doctor of Medicine (MD) degree listed on their university Web site. Because Doctor of Osteopathic Medicine programs are not offered outside of the US, this study focused on faculty with MD degrees for more consistency between Canadian and US data. Faculty with departmental leadership roles such as chair, vice-chair, director, associate/assistant directors, department head, and chief were also included. Missing gender data were further searched through Google, LinkedIn and Doximity profiles. Faculty members whose academic ranks could not be determined were excluded. Adjunct and retired faculty, faculty without an MD degree (or unstated), pending staff members, and faculty whose gender could not be identified were also excluded. Elsevier's SCOPUS was used to gather bibliometric data pertaining to the publications, h-index, citations, and tenure of productivity of each faculty member. The data were then analyzed using Strata Version 14.2. This study conforms to the STROBE guidelines accordingly (see the Supplemental Checklist, Supplemental Digital Content 1, https://links.lww.com/PHM/A724).

Data Analysis Procedure

The data gathered was tested for normality. Log transformation was performed for the continuous variables of h-index, number of citations, and number of publications, which were found to be skewed in distribution. At the univariate level, simple linear regression was applied. Each variable was regressed independently with h-index, the assumptions were checked, and their significance was reported. Gender was the primary factor of interest. Variables significant in the univariate regression were gender, number of publications, number of citations, years of active research, academic ranks, and leadership ranks; these were therefore selected for inclusion in the multivariable linear regression analysis. Multicollinearity between independent variables was checked for using a correlation coefficient. Cramer's V test was used for one nominal and one ordinal variable, whereas Spearman test was used for one continuous variable and one ordinal variable. A correlation of 0.8 was treated as the presence of multicollinearity, and there was no multicollinearity seen overall. Main effects were identified using a stepwise selection strategy based on the P value. The multivariable analysis supported the inclusion of gender, number of citations, number of publications, academic rank, leadership rank, and years of active research in the preliminary model. The final step was to check for interactions. Interaction terms were created between each of the main effects in the model, which demonstrated no significant interaction. Academic rank, number of publications, and number of citations were not confounders for h-index, because they did not fulfill the criterion of being confounders. A final model was created: y(x) = β0 + β1 (female) + β2 (number of publications) + β3 (number of citations) + β4 (years of active research) + β51 (academic rank-associate professor) + β52 (academic rank-professor) + β6 (leadership rank second in command). This prediction equation accounted for major variabilities in the model as adjusted R2 = 0.9076, F test = 47.28, and P ≤ 0.001.

RESULTS

A total of 1045 faculty members who met the inclusion criteria were identified for this study. Of these individuals, 653 (62.49%) were men and 392 (37.51%) were women. Faculty academic ranks were available for 1042 physicians, of which 84 (8.06%) were junior faculty, 582 (55.85%) were assistant professors, 211 (20.25%) were associate professors, and 165 (15.83%) were professors. A total of 249 faculty members (23.8%) were found to be holding leadership positions. Among the faculty members with leadership positions, there were 213 men (85.54%) and only 36 women (14.46%), indicating a significant gender differences within this group. These results are summarized in Table 1.

T1
TABLE 1:
Men versus women representation in PM&R faculties in the US and Canada

Analysis of the data revealed that men faculty members outnumbered women across all academic ranks in the field of PM&R. Specifically, men faculty comprised 61.9% of junior faculty, 57.71% of assistant professors, 63.33% of associate professors, and 79.14% of professors (Table 1). An expected decline in the number of men beyond the junior rank was observed when comparing them with preceding ranks, whereas women held a consistently low number of positions across all ranks (Table 1). Within the group of faculty members holding leadership roles, a significant male representation (73.08%) was observed across roles, which were first-in-command, including chief, chairperson, and program director (Table 1). Women physicians, although still in much lower numbers compared with men, had a slightly higher representation (31.43%) in second-in-command ranks such as deputy chief, deputy chairperson, or associate program director (Table 1). Nonetheless, women remained a minority in all these ranks.

Citation data were available for 626 faculty members. Overall, the median number of citations was 101.5 (range = 0–27,325). Among men faculty members, the median number of citations was 143 (range = 0–8315), whereas the median number of citations for women was 69.5 (range = 0–27,325). Furthermore, publication data were available for 695 faculty members. The overall median number of publications was 7 (range = 1–353). The median number of publications was 9 for men (range = 1–353) and 5 for women (range = 1–242). Finally, h-index data were obtained for 647 faculty members. It is important to note that higher academic ranks for men have higher h-index compared with the ranks for women faculty. However, the median h-index among men overall was 5 (range = 0–96) and the median h-index among women was 4 (range = 1–58), with h-index increasing in correlation with academic rank for both genders. These results are shown in Table 2.

T2
TABLE 2:
Bibliometric data (citations, publications, and h-index) for men versus women PM&R faculty members

A prediction equation was created to determine the relationship between gender and academic rank, leadership role, number of citations and publications, and years of active research. The equation produced a variability model with an adjusted R2 value of 0.85 (P ≤ 0.001). The remaining variability in the model may be explained by factors such as full- versus part-time employment, years of employment, and contract versus tenure positions; however, this is beyond the scope of the report as data that were available on the Internet was used. The study's model determined that the odds of men PM&R faculty having a higher h-index than women PM&R faculty was 0.78 (95% confidence interval = 0.24–0.87), after adjusting for number of publications, number of citations, academic rank, leadership rank, and interaction between gender and publications and gender and citations. Furthermore, for the h-index based on gender overall (faculty n = 647), the median h-index for men was 5 (n = 452) and the median h-index for women was 4 (n = 222). Thus, the results demonstrate that h-index for the combination of academic ranks is comparable between men and women faculty members in PM&R departments.

DISCUSSION

Gender disparities in academic medicine have been gaining increasing recognition for the past several years.2 Despite the fact that the number of women medical trainees in the US has risen from 5.1% in 1960 to 46% by 2014, a significant gender imbalance among physicians who hold senior academic ranks and leadership positions still exists today.2 Based on conclusions drawn from several studies, women advance through academic ranks more slowly, are promoted less, have a lower publication rate, and ultimately achieve lower academic productivity compared with men of similar rank.23,24 Furthermore, women are less likely to attain leadership roles because leaders are often selected among those with high academic productivity.25 Carr et al.26 followed 1273 faculty at 24 medical schools in the continental US for 17 yrs to identify predictors of advancement, retention, and leadership for women faculty and found that after adjusting for significant covariates, women were less likely than men to achieve the rank of professor (odds ratio = 0.57; 95% confidence interval = 0.43–0.78), and male faculty were more likely to hold senior leadership positions after adjusting for publications (odds ratio = 0.49; 95% confidence interval = 0.35–0.69). They concluded that gender disparities in rank, retention, and leadership remain across the career trajectories of women faculty.26 The authors recommended that institutions examine the climate for women to ensure that their academic capital is fully used and equal opportunity exists for leadership.26 Indeed, the underrepresentation of women is an issue in the field of medicine because diversity in the skill sets and attitudes among medical faculty are important to consider to train future generations of physicians.27

In the field of PM&R, women have a greater representation compared with some other medical specialties.8 However, women physicians in the academic PM&R community do not achieve a similar level of success as their male peers in terms of career advancement and research productivity. This study aimed to examine the gender differences in academic PM&R faculties in the US and Canada and to determine the associations between gender, leadership rank, and research productivity. The results demonstrated that men are found in greater numbers across all academic ranks and hold most leadership positions. Men faculty members also had a higher number of citations and publications as well as a higher overall research productivity compared with women across all academic ranks.

Research productivity, as measured by h-index, is an important quantitative factor that contributes to career advancement and promotion.18,28 The relationship between gender and academic productivity was assessed in this study, where the findings indicated that women were not significantly inferior in academic performance compared with men in the academic PM&R field overall with all academic ranks combined. Thus, the gender gap present in academic and leadership positions is likely due to other underlying factors. It has been noted that men had higher h-index at higher academic ranks specifically, which may be potentially influenced by the differences in years of research for faculty members. Women may be more focused on familial roles during their late 20s and 30s, where associate level promotions may occur around this period.29,30 Being unable to fully account for the years of research and possible gaps in between research years are limitations, and further research studying adjusted years of research can help assess this possible gender difference.

The cause of gender disparity in medicine is still being explored by researchers today, because the issue is complex and most likely multifactorial. According to Helitzer et al.,29 women face barriers in both individual and institutional domains that contribute to gender disparity in academic medicine. In the individual domain, tensions between cultural expectations of family and academic roles, as well as personal satisfaction achieved from their work are relevant factors.29 In the institutional domain, the sociocultural environment of academic health centers, lack of institutional strategies to engage women in organizational initiatives, and the influence of critical leaders on women's ascent to leadership are important elements.29 The study suggested that strategies aiming to close the gender disparity gap would include an institutional culture that is receptive to changing existing cultural practices and changes in the perception and understanding that leadership roles are appropriate for women.29 Women are still less represented in research and quantitative science careers than men.3 To decrease this gender gap, organizations must work to eliminate the incongruities of being both a leader and a woman. At the same time, women must be engaged in organizational initiatives, and critical leaders must be encouraged to foster women's ascent to leadership.

This study has its share of limitations including potential dating or inaccuracy of information available on the PM&R program Web sites and faculty listings used during the data collection. In addition, women faculty members may have undergone changes in their last name after marriage. Unfortunately, this limitation could not be accounted for, because there was no way to discover the change and correlate the number of publications, h-index, or academic parameters before and after the name change. Furthermore, there is a lack of standardized resources to explore additional parameters, such as faculty age and length of academic service for better comparisons. Only a portion of the physician PM&R academic workforce was analyzed to maintain consistency between US and Canadian data because Canada does not offer any Doctor of Osteopathic Medicine programs. Moreover, leadership appointments outside of PM&R departments were not explored in further detail.

A thorough understanding of the extent of the underrepresentation of women physicians in academic medicine as well as its underlying causes is critical in bridging the gender disparity in the future. The study's findings reaffirm the idea that women today are still unable to reach equal standing in research and leadership compared with men, despite the fact that they are not inferior in academic performance. Systemic changes at the organizational level should take place before women can be adequately represented in the field of PM&R. Leaders, both men and women, should first recognize any gender disparities in their departments where they can better invest their time and resources into providing solutions, which may include engagement and promotions. Furthermore, leaders and organizations can facilitate focus groups to assess for any difficulties and create change strategies, such as considering leadership development programs and operationalizing measures to examine and address noted gender disparities. Lastly, an effort should be made to eliminate other potential work-related barriers faced by women physicians, including increased burnout, biases, salary inequities, and lack of protected time for research.30 As mentioned earlier, the AAP's task force for Women in Academic Physiatry recently released a report on the diversity and inclusion of women physiatrists within the medical society and laid out a plan to follow the metrics and address disparities.14 All organizations should focus on metrics-driven diversity and inclusion and rule in or out causal factors to close gaps.

CONCLUSIONS

A significantly greater number of men make up PM&R faculties across all academic ranks and leadership positions in the US and Canada compared with women. Gender differences are most prominent among professors and faculty members holding first-in-command leadership positions, where men hold a significant majority of positions. Men physicians were also found to have higher academic productivity, with a greater number of citations and publications as well as a higher h-index compared with women physicians across all academic ranks. However, the h-index based on gender, with all academic ranks combined, is comparable between men and women physiatrists, suggesting that more complex and multifactorial issues are likely influencing the existing gender differences. Leaders in their respective departments, along with systemic changes at the organizational level, can help address associated barriers as well as support leadership development and programs to minimize the noted differences.

ACKNOWLEDGMENTS

The authors thank Jessica Bui for her numerous contributions to the data collection and article.

REFERENCES

1. The State of Women in Academic Medicine: The Pipeline and Pathways to Leadership. Washington, DC, Association of American Medical Colleges, 2014
2. The State of Women in Academic Medicine: AAMC. 2014. Available at: https://members.aamc.org/eweb/upload/The%20State%20of%20Women%20in%20Academic%20Medicine%202013-2014%20FINAL.pdf Accessed September 1, 2017
3. Schor NF: Tdecanal divide: women in decanal roles at U.S. medical schools. Acad Med 2018;93:237–40
4. 2014 Physician Specialty Data Book. Washington, DC, Association of American Medical Colleges, 2015
5. Bickel J, Quinnie R: Women in Medicine: Statistics. Washington, DC, Association of American Medical Colleges, 1992
6. Table 4A: Distribution of Women M.D. Faculty by Department and Rank, 2014. American Academy of Medical Colleges. Available at: https://www.aamc.org/download/411788/data/2014_table4a.pdf Published 2014. Accessed May 10, 2017
7. Silver JK, Blauwet CA, Bhatnagar S, et al.: Women physicians are underrepresented in recognition awards from the Association of Academic Physiatrists. Am J Phys Med Rehabil 2018;97:34–40
8. Wagner AK, McElligott J, Chan L, et al.: How gender impacts career development and leadership in rehabilitation medicine: a report from the AAPM&R research committee. Arch Phys Med Rehabil 2007;88:560–8
9. Bickel J: The work that remains at the intersection of gender and career development. Arch Phys Med Rehabil 2007;88:683–6
10. Bowles AO, Kevorkian CG, Rintala DH: Gender differences regarding career issues and promotion in academic Physical Medicine and Rehabilitation. Am J Phys Med Rehabil 2007;86:918–25
11. Hwang J, Byrd K, Nguyen MO, et al.: Gender and ethnic diversity in Academic PM&R Faculty: national trend analysis of two decades. Am J Phys Med Rehabil 2017;96:593–5
12. Silver JK, Bhatnagar S, Blauwet CA, et al.: Female physicians are underrepresented in recognition awards from the American Academy of Physical Medicine and Rehabilitation. PM&R 2017;9:976–84
13. Silver JK, Slocum CS, Bank AM, et al.: Where are the women? The underrepresentation of women physicians among recognition award recipients from medical specialty societies. PM&R 2017;9:804–15
14. Silver JK, Cuccurullo SJ, Ambrose AF, et al.: Association of Academic Physiatrists Women's Task Force Report. Am J Phys Med Rehabil 2018;97:680–90
15. Garner RM, Hirsch JA, Albuquerque FC, et al.: Bibliometric indices: defining academic productivity and citation rates of researchers, departments, and journals. J Neurointerv Surg 2018;10:102–6
16. Bastian S, Ippolito JA, Lopez SA, et al.: The use of the h-index in academic orthopaedic surgery. J Bone Joint Surg Am 2017;99:e14
17. Hu J, Gholami A, Stone N, et al.: An Evaluation of h-Index as a measure of research productivity among Canadian academic plastic surgeons. Plast Surg (Oakv) 2018;26:5–10
18. MacMaster FP, Swansburg R, Rittenbach K: Academic productivity in psychiatry: benchmarks for the h-index. Acad Psychiatry 2017;41:452–4
19. Pashkova AA, Svider PF, Chang CY, et al.: Gender disparity among US anaesthesiologists: are women underrepresented in academic ranks and scholarly productivity? Acta Anaesthesiol Scand 2013;57:1058–64
20. Shah A, Jalal S, Khosa F: Influences for gender disparity in dermatology in North America. Int J Dermatol 2018;57:171–6
21. Khurshid K, Shah S, Ahmadi M, et al.: Gender differences in the publication rate among breast imaging radiologists in the United States and Canada. Am J Roentgenol 2018;210:2–7
22. Qamar SR, Khurshid K, Jalal S, et al.: Academic musculoskeletal radiology: influences for gender disparity. Skeletal Radiol 2018;47:381–7
23. Wright AL, Schwindt LA, Bassford T, et al.: Gender differences in academic advancement: patterns, causes, and potential solutions in one US college of medicine. Acad Med 2003;78:500–8
24. Ash AS, Carr PL, Goldstein R, et al.: Compensation and advancement of women in academic medicine: is there equity? Ann Intern Med 2004;141:205–12
25. Reed DA, Enders F, Lindor R, et al.: Gender differences in academic productivity and leadership appointments of physicians throughout academic careers. Acad Med 2011;86:43–7
26. Carr PL, Raj A, Kaplan SE, et al.: Gender differences in academic medicine: retention, rank, and leadership comparisons from the national faculty survey. Acad Med 2018;93:1694–9
27. Mayer AP, Files JA, Ko MG, et al.: Academic advancement of women in medicine: do socialized gender differences have a role in mentoring? Mayo Clin Proc 2008;83:204–7
28. Mayer EN, Lenherr SM, Hanson HA, et al.: Gender differences in publication productivity among academic urologists in the United States. Urology 2017;103:39–46
29. Helitzer DL, Newbill SL, Cardinali G, et al.: Changing the culture of academic medicine: critical mass or critical actors? J Womens Health (Larchmt) 2017;26:540–8
30. Bickel J, Wara D, Atkinson BF, et al.: Increasing womenʼs leadership in academic medicine. Acad Med 2002;77:1043–61
Keywords:

Physical and Rehabilitation Medicine; Medical Faculty; Gender; Leadership

Supplemental Digital Content

Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.