http://orcid.org/0000-0003-4899-0616
Figures
Abstract
Objective
To determine the presence of a gender gap in the scientific production among Peruvian physicians and analyze either gap is associated with the presence of observable factors or the presence of prejudices against female physicians.
Methods
We analyzed data from the National Survey of User Satisfaction in Health 2016, a nationally representative survey that collected information about medical professionals working in health institutions in Peru. The outcome of interest was the number of publications in indexed journals. We estimated the gender gap in scientific production using the Oaxaca-Blinder (OB) decomposition method.
Results
From the 2216 physicians surveyed, 252 reported published at least one article in an indexed journal. From physicians with scientific production, 37.7% were women. The analysis of OB decomposition showed a gap of 2.11 indexed publications, disfavoring female physicians (p<0.01). Likewise, the explained component was 1.36 publications, representing 64.5% of the total gap (p<0.05).
Citation: Amaya E, Mougenot B, Herrera-Añazco P (2019) Gender disparities in scientific production: A nationwide assessment among physicians in Peru. PLoS ONE 14(11): e0224629. https://doi.org/10.1371/journal.pone.0224629
Editor: Andrew Carl Miller, East Carolina University Brody School of Medicine, UNITED STATES
Received: May 20, 2019; Accepted: October 17, 2019; Published: November 5, 2019
Copyright: © 2019 Amaya et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data underlying the results presented in the study are available on OSF HOME (a free open platform to upload research projects and supporting information), the link to download the database is: https://osf.io/t3vkr.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Scientific research is an essential component in the training of health professionals, since its practice generates a better understanding of clinical management and, consequently, improves the healthcare provided to patients [1].
Although the number of women pursuing careers in health has significantly increased [2], several dimensions of health scientific research show gender disparities. For instances, the underrepresentation of women as speakers in medical conferences [3], the low proportion of women being first author in cases reports and articles, even in predominantly female specialties [4,5].
Most visible indicator of gender disparities in health scientific research is the number of publications in indexed journals. Medical articles produced by women has grown significantly [6–8], however there is still a substantial gender gap [9–12], where women produce a third of medical articles worldwide [13]. Several factors can explain this gap [6–8]. Most studies highlight the presence of observable factors that disadvantage women, such as barriers in access to resources and academic opportunities, as well as cultural norms that overcharge home chores for women, especially in child care responsibilities [14–19]. On the other hand, there are also non-observable factors, associated with prejudices and sexism against women, which limits their academic development [7,20].
The gender gap in the number of articles vary according to the context [21]. In low-scientific production countries, this gap is diluted and there are even cases where publications are predominantly made by women [13]. Peru is a low-scientific production country, where biomedical areas contribute the most [22]. According to Scival, during the period 2013–2018, 13,719 articles were published, where articles in "Life Sciences and Medicine" represented 50.4% of the total. On the other hand, women enrollment in medical careers has increased in recent years, the proportion of women over 17 years enrolled in medicine went from 1.5% in 2011 to 2.6% in 2016 of the total population [23].
Despite these facts, there are no studies that have measured the gender gaps in scientific production among Peruvian physicians at national level. Moreover, although several studies have measured the gender gap in the number of articles for some medical specialties, these have been limited to quantifying the raw gap without analyzing the factors that explain it [7,14,20]. To the best of our knowledge, this is the first study that measures the gender gap in the number of publications in indexed journals for a low-scientific production country.
The goals of this study are, first, to determine the existence of a gap in scientific production by gender among Peruvian physicians; and, second, to explore whether this gap is associated with the presence of observable factors or the presence of prejudices that disadvantage female physicians.
Materials and methods
Study design
Cross-sectional secondary analysis that used data from the National Survey of User Satisfaction in Health 2016 (ENSUSALUD-2016), which was conducted by the National Institute of Statistics and Informatics of Peru (INEI) and the National Superintendence of Health of Peru (SUSALUD). The ENSUSALUD-2016 was a nationally representative survey aimed at collecting information on the operation of the Health Services Provider Institutions (IPRESS), through the information provided by the users and providers of said services [24].
Processes
We used the second questionnaire of the ENSUSALUD-2016, which was directed to the personnel in medicine and nursing that worked in one of the 184 IPRESS selected. The IPRESS were selected between establishments of the Ministry of Health (MINSA), Social Security in Health (EsSalud), Army/Police and Private Institutions, which had hospitalization service (I-IV level). The questionnaire was prepared by specialized personnel of SUSALUD and was conducted in person from 13th May to 9th July of 2016 by trained interviewers.
The ENSUSALUD-2016 used a two-stage probabilistic sampling by conglomerates and stratified by 25 administrative regions of Peru. In the first stage followed a random sampling where the unit was the IPRESS and in the second stage followed a systematically sampling where the unit was the health professionals with a minimum of one year working in the IPRESS selected.
Population
We include in the analysis only the medical professionals who reported having at least one publication in an indexed journal, that means registered in the ISI, SCOPUS or MEDLINE databases. Physicians also reported full information about their demographic and labor-market characteristics.
Outcome variable: Number of publications
The outcome variable was the number of publications in an indexed journal, measured as a numerical variable. This information was self-reported. In a first instance, a question was asked with a dichotomous answer (yes / no): "Have you published an original article in a journal indexed to ISI, SCOPUS or MEDLINE?" In a second instance, for physicians who offered an affirmative answer, they asked themselves: "How many original articles have you published in a journal indexed to ISI, SCOPUS or MEDLINE?" The answer to this question took values of one or more than one.
The databases cover several areas of science literature. The ISI database covers mainly the social sciences and the humanities. The Scopus database is significantly larger in size and scope, and covers more of the international literature but excludes the Humanities. The MEDLINE database covers more of the science and the biomedical literature.
Exposure variable: Gender of the physician
The exposure of interest was the gender of the medical professional (male / female); this information was collected by direct observation of interviewees.
Other variables
Other variables used in the analysis are: Institution in which the medical professional worked (MINSA / EsSalud / Other), age groups (23–40 / 40–55 / 55 plus), having a medical specialty (yes / no), teaching activity in higher education (yes / no) and being accredited as a researcher in the National Registry of Researchers of Science, Technology and Innovation of Peru-REGINA (yes / no). A researcher qualified as REGINA must publish academic books, articles in indexed journals, registered patents and presentation in national/international conferences.
Years practicing the medical profession, obtained by the difference between the year in which the survey was taken (year 2016) and the year he reported having obtained the university degree, later the variable was categorized into tiles. Having time for family life, which was obtained from the question "does your workload give you enough time for your personal and family life?" And was redefined in a categorical variable, in which the answer was affirmative when the Professional responded to be (totally agree / agree) and was negative when the professional gave another answer.
Statistical analysis
We performed the analysis using Stata 15.0 statistical software (StataCorp LP, College Station, TX, USA). We considered the complex structure of the sampling using the command svy [25], weighed by conglomerate using the variable of the identifier of the IPRESS and by stratum using the variable region, and we used the survey factor expansion.
We described the continuous variables by averages and standard deviations, applying a Wald test to compare them according to gender. For the categorical variables, we used absolute frequencies and proportions, applying a χ² Pearson test for gender comparison.
We estimated the gap in scientific production by gender using the Oaxaca-Blinder (OB) decomposition method [26,27], which is expressed in the following equation:
(1)
Where the superscripts M and F represent the group of male and female, respectively. The left side of the Eq 1 shows the average gap in scientific production by gender. The first component on the right side of the Eq 1 is the explained component, which represents the proportion of the gap in scientific output that is attributed to differences in observable characteristics among male and female physicians, for our case all the adjustment variables presented previously (for example, age group, being a REGINA researcher, among others). The second component on the right side of the Eq 1 represents the proportion of the unexplained gap, which can be attributed to omitted variables (unobserved), unconscious biases, and discrimination against female physicians.
The proportions of the explained and unexplained components of the OB decomposition vary depending on the group considered as a base category (male / female), as this defines the group subject to prejudice. In our analysis we use the correction of Fortin et al. [28], which is robust to the choice of the base category.
Ethical aspects
Our study was a secondary analysis of public data (http://portal.susalud.gob.pe/blog/base-de-datos-2016/). The interviewers solicited a verbal consent from each participant, indicating that the survey was anonymous and voluntary. Only information from the volunteer participants was collected.
Results
Description of the population
From the 2216 physicians surveyed, 252 reported having at least one publication in an indexed journal (11.4%). From physicians with scientific production, 37.7% were women, 43.4% worked at MINSA, 42.3% belonged to age group 23–40 years and 38.3% were in the lower title regarding time exercising the profession. Likewise, 71.7% had a specialty, 8.6% were REGINA researchers, 36.4% were professors and 32.3% had time for family life (Table 1).
Differences by gender in doctors with scientific output
From physicians with scientific production, we found significant differences by gender in the number of publications, the age groups, the years practicing the medical profession, being a REGINA researcher, being a professor and having time for family life. According to scientific production, male physicians had on average more publications than their female counterparts (p<0.05). In the case of age, 59% of women and 32.1% of men were concentrated in the age group of 23 to 40 years (p<0.1). With respect to time exercising the profession, 58.6% of women and 32.3% of men were in the group with the fewest years practicing the profession (p<0.1). In the case of REGINA researchers, 0.6% of women and 13.4% of men were registered as REGINA researchers (p <0.01). In the case of teachers, 11.1% of women and 51.7% of men reported working as teachers (p <0.01). Finally, with respect to family life, 11.7% of women and 44.7% of men had time to have a family life (p<0.01) (Table 2).
Oaxaca-Blinder decomposition
Through the OB decomposition, we found a gap of 2.11 indexed publications, which disadvantages women (p <0.01). The explained component was 1.36, representing 64.5% of the gap (p <0.05). On the other hand, the unexplained component was 0.75, representing the remaining 35.5%, although it was not statistically significant (Table 3).
In general, the results suggest that the gender gap in scientific production is explained by differences in observable characteristics that disadvantage women, such as the years practicing medicine, being REGINA researchers and being professors. The number of years practicing medicine is the main factor, representing around 50% of the total gap.
Discussion
Our study shows that 1 in 10 Peruvian physicians report having publications in an indexed journal. From physicians with scientific production, 4 out of 10 were women. In addition, the OB decomposition pointed out that there is a gap of 2.11 indexed publications that disadvantages women. This gender gap can be explained mainly by observable factors, such as years practicing medicine, being accredited as researchers in the REGINA and being a professor.
The frequency of publications in our study is similar to that found in a study for physicians of the residency program [1]. However, both studies are based on self-reported information, so this indicator may be overestimated. On the other hand, the low frequency of publications is compatible with the Peruvian educational system; in which the regulatory institutions of the medical profession and university authorities do not promote research efficiently [29,30]. For this reason, it seems necessary to propose a system of incentives for health scientific research [31,32], complemented with improved in the practices of research mentorship in settings such as medical undergraduate, both could improve research practice among Peruvian physicians [33–35].
Like the global trend, we find a gap in scientific production by gender [9–12]. A study that included several branches of science points out that although the gap has been narrowing, only a third of the global publications were produced by women [13]. On the other hand, a study for high-impact medical journals found that the proportion of women who figured as first authors went from 27% in 1994 to 37% in 2017 [7]. This trend has been sustained in most medical specialties and in other areas of knowledge [6–8,13], though certain variations are reported in some medical specialties [36]. A study that evaluated the gender gap in the first authors of English medical journals indicates that this gap decreased in gynecology journals, with a smaller decrease in Gastroenterology journals, presumably because of differences in the choice of specialty associated with gender [36]. These differences in the choice of the area of knowledge are a key factor, since it is more frequent to find women authors in subjects related to nursing, language, education, social work, whereas men are predominant in military subjects, and robotic or engineering sciences [13].
The causes that explain the gap in scientific production by gender are multiple and complex. These may be related to personal characteristics, or structural issues, such as fewer possibilities for women to occupy senior researchers positions and thus having a lower probability of obtaining funds to finance research, and being less likely to lead research projects or be first authors in the articles [7,8,13]. In this sense, being accredited as REGINA researchers, which includes requirements such as having published scientific articles, having participated in research projects and having advised thesis projects, and exercising university teaching, are variables that explain the gap in scientific production by gender.
With respect to time practicing the profession of medicine, several studies indicate that the time occupied in research work is related to the frequency of the production of articles [13]. In that sense, the time they have been exercising the profession could be considered as a proxy indicator of the dedication to research work, in a context in which young researchers produce fewer articles than senior researchers.
Although domestic chores, which include caring for children, are often included as a factor that could explain this gap [9,13], studies indicate that in the research, women with children and without children have the same productivity [18]. Finally, one factor that could explain this gap is the presence of prejudice or sexism, as has been suggested by other authors [7,20]. A study evaluating the applications for a laboratory position in North American universities showed that men were perceived as more competent and eligible than their female counterparts, even when the evaluators were women [20]. Although in this analysis the presence of observable factors predominated over unobserved ones, it is clear that in the academic field women suffer a series of disadvantages, such as lower likelihood of being main teachers, deans, obtaining research funds, salaries, among others [20,37–40]. This in turn suggests that unobserved factors play an essential role when discussing gender gaps.
Limitations and strengths
Our study has some limitations. First, and mainly, the number of publications is self-reported. Although interviewers explained that the information collected in ISI, SCOPUS and MEDLINE are indexed, responses may be affected by respondents’ recall bias, so this indicator is likely to be overestimated. Second, we have included in the analysis the factors identified in the ENSUSALUD-2016, there may be other factors that influence the gender gap in scientific production and that we have not considered. Thirdly, by using a cross-sectional survey, we are not able to establish a causal relationship between the factors and the gender gap in scientific production.
The main strengths of our study are that, first, we use a method that in addition to measuring the gender gap in scientific production, quantifies the importance of the factors that explain this gap, and, second, it is based on a nationally representative survey, so the results are valuable for policymakers when proposing policies that encourage research in medical professionals.
Conclusions
There is a gender gap in the number of articles by Peruvian physicians that disadvantages women. This gap is mainly explained by observable factors such as having less years of practicing medicine, being a REGINA researcher and being a professor. There is an opportunity to reduce this gap by promoting the career of REGINA researcher and university teaching among women who practice medicine.
References
- 1. Herrera-Añazco P, Saavedra PJO, Rondán ÁT, Gutiérrez WN, Díaz CA, Armas DJ, et al. Prevalencia y factores asociados a publicar artículos científicos durante la residencia médica en Perú. FEM Rev Fund Educ Médica. 2018;21: 9–16.
- 2.
Jolliff L, Leadley J, Coakley E, Sloane R. Women in US academic medicine and science: statistics and benchmarking report 2011–2012. Washington DC: Association of American Medical Colleges; 2012.
- 3. Ruzycki SM, Fletcher S, Earp M, Bharwani A, Lithgow KC. Trends in the Proportion of Female Speakers at Medical Conferences in the United States and in Canada, 2007 to 2017. JAMA Netw Open. 2019;2: e192103–e192103. pmid:30977853
- 4. Silver JK, Poorman JA, Reilly JM, Spector ND, Goldstein R, Zafonte RD. Assessment of women physicians among authors of perspective-type articles published in high-impact pediatric journals. JAMA Netw Open. 2018;1: e180802–e180802. pmid:30646033
- 5. Hsiehchen D, Hsieh A, Espinoza M. Prevalence of Female Authors in Case Reports Published in the Medical Literature. JAMA Netw Open. 2019;2: e195000–e195000. pmid:31150085
- 6. Strand M, Bulik CM. Trends in female authorship in research papers on eating disorders: 20-year bibliometric study. BJPsych Open. 2018;4: 39–46. pmid:29467058
- 7. Filardo G, da Graca B, Sass DM, Pollock BD, Smith EB, Martinez MA-M. Trends and comparison of female first authorship in high impact medical journals: observational study (1994–2014). bmj. 2016;352: i847. pmid:26935100
- 8. Bendels MH, Müller R, Brueggmann D, Groneberg DA. Gender disparities in high-quality research revealed by Nature Index journals. PloS One. 2018;13: e0189136. pmid:29293499
- 9. Barnett RC, Carr P, Boisnier AD, Ash A, Friedman RH, Moskowitz MA, et al. Relationships of gender and career motivation to medical faculty members’ production of academic publications. Acad Med. 1998;
- 10. Elhakimi W, Al Othman A, El Yahia M, Al Dawood A, Al Sadiq S, Mosli M, et al. Female authorship in major endocrinology journals: a 25-year progression. J Endocrinol Metab Diabetes South Afr. 2018;23: 76–79.
- 11. González-Álvarez J, Cervera-Crespo T. Research production in high-impact journals of contemporary neuroscience: A gender analysis. J Informetr. 2017;11: 232–243.
- 12. González-Alvarez J. Author gender in The Lancet journals. The Lancet. 2018;391: 2601.
- 13. Larivière V, Ni C, Gingras Y, Cronin B, Sugimoto CR. Bibliometrics: Global gender disparities in science. Nat News. 2013;504: 211.
- 14. Carr PL, Ash AS, Friedman RH, Scaramucci A, Barnett RC, Szalacha LE, et al. Relation of family responsibilities and gender to the productivity and career satisfaction of medical faculty. Ann Intern Med. 1998;129: 532–538. pmid:9758572
- 15. Duch J, Zeng XHT, Sales-Pardo M, Radicchi F, Otis S, Woodruff TK, et al. The possible role of resource requirements and academic career-choice risk on gender differences in publication rate and impact. Plos One. 2012;7: e51332. pmid:23251502
- 16. Long MT, Leszczynski A, Thompson KD, Wasan SK, Calderwood AH. Female authorship in major academic gastroenterology journals: a look over 20 years. Gastrointest Endosc. 2015;81: 1440–1447. pmid:25887727
- 17. Jolly S, Griffith KA, DeCastro R, Stewart A, Ubel P, Jagsi R. Gender differences in time spent on parenting and domestic responsibilities by high-achieving young physician-researchers. Ann Intern Med. 2014;160: 344–353. pmid:24737273
- 18. Helmer M, Schottdorf M, Neef A, Battaglia D. Gender bias in scholarly peer review. Elife. 2017;6: e21718. pmid:28322725
- 19. Boynton JR, Georgiou K, Reid M, Govus A. Gender bias in publishing. The Lancet. 2018;392: 1514–1515.
- 20. Moss-Racusin CA, Dovidio JF, Brescoll VL, Graham MJ, Handelsman J. Science faculty’s subtle gender biases favor male students. Proc Natl Acad Sci. 2012;109: 16474–16479. pmid:22988126
- 21. Jagsi R, Guancial EA, Worobey CC, Henault LE, Chang Y, Starr R, et al. The “gender gap” in authorship of academic medical literature—a 35-year perspective. N Engl J Med. 2006;355: 281–287. pmid:16855268
- 22. Herrera-Añazco P, Valenzuela-Rodríguez G, Pacheco-Mendoza J, Málaga G. Scientific production of Vice Chancellors for Research in Peruvian universities with a medical school. Medwave. 2017;17.
- 23.
Instituto Nacional de Estadística e Informática (INEI). Perú, Brechas de género, 2016: Avances hacia la igualdad de mujeres y hombres [Internet]. Lima, Perú: Instituto Nacional de Estadística e Informática (INEI).; 2016. Available: http://www2.congreso.gob.pe/sicr/cendocbib/con4_uibd.nsf/9D551FD9CE29ABE7052580790075B50A/$FILE/2.Per%C3%BA.Brecha_genero-2016.pdf
- 24.
Instituto Nacional de Estadística e Informática (INEI). Encuesta Nacional de Satisfacción de Usuarios en Salud 2016, Informe Final. Lima, Peru: Instituto Nacional de Estadística e Informática (INEI); 2016.
- 25.
StataCorp. Stata Survey Data Reference Manual, Release 13. College Station, Texas: Stata Corp LP; 2013.
- 26. Oaxaca R. Male-female wage differentials in urban labor markets. Int Econ Rev. 1973;14: 693–709.
- 27. Blinder AS. Wage Discrimination: Reduced Form and Structural Estimates. J Hum Resour. 1973;8: 436–455.
- 28.
Fortin N, Lemieux T, Firpo S. Chapter 1—Decomposition Methods in Economics. In: Card OA and D, editor. Handbook of Labor Economics. Elsevier; 2011. pp. 1–102. Available: http://www.sciencedirect.com/science/article/pii/S0169721811004072
- 29. Rey de Castro J. ¿Las instituciones médicas regulatorias del Perú apuestan por el desarrollo de la investigación científica? Rev Medica Hered. 2012;23: 269–270.
- 30. Mayta-Tristán P, Pereyra-Elías R, Mejía CR. Producción científica de los miembros vitalicios de la academia nacional de investigadores médicos. Rev Peru Med Exp Salud Pública. 2013;30: 720–721.
- 31. Ynalvez MA, Shrum WM. Professional networks, scientific collaboration, and publication productivity in resource-constrained research institutions in a developing country. Res Policy. 2011;40: 204–216.
- 32. Pavesi G, Siccardi A, Viale G, Grazioli C, Calciolari T, Tenchini ML, et al. Hedgehogs, humans and high-school science: The benefits of involving high-school students in university research. EMBO Rep. 2008;9: 208–211. pmid:18311167
- 33. Manjarin M, Cutri A, Noguerol E, Torres F, Ossorio F, Ferrero F. Enseñanza de la investigación con un sistema de tutores durante la residencia de pediatría. Arch Argent Pediatr. 2007;105: 333–41.
- 34. Ahmad S, De Oliveira GS Jr, McCarthy RJ. Status of anesthesiology resident research education in the United States: structured education programs increase resident research productivity. Anesth Analg. 2013;116: 205–210. pmid:23223116
- 35. Ticse R, Pamo O, Samalvides F, Quispe T. Factores asociados a la culminación del proyecto de investigación requerido para optar el título de especialista en una universidad peruana. Rev Peru Med Exp Salud Pública. 2014;31: 48–55. pmid:24718526
- 36. Sidhu R, Rajashekhar P, Lavin VL, Parry J, Attwood J, Holdcroft A, et al. The gender imbalance in academic medicine: a study of female authorship in the United Kingdom. J R Soc Med. 2009;102: 337–342. pmid:19679736
- 37. Amaya E, Mougenot B. The gender differences in highly paid wage: a case study of Peruvian physicians. Cad Saude Publica. 2019;35(5).
- 38. Waisbren SE, Bowles H, Hasan T, Zou KH, Emans SJ, Goldberg C, et al. Gender differences in research grant applications and funding outcomes for medical school faculty. J Womens Health. 2008;17: 207–214.
- 39. Budden AE, Tregenza T, Aarssen LW, Koricheva J, Leimu R, Lortie CJ. Double-blind review favours increased representation of female authors. Trends Ecol Evol. 2008;23: 4–6. pmid:17963996
- 40. Jagsi R, Griffith KA, Stewart A, Sambuco D, DeCastro R, Ubel PA. Gender differences in the salaries of physician researchers. Jama. 2012;307: 2410–2417. pmid:22692173