Sex Inequalities in Medical Research: A Systematic Scoping Review of the Literature.

A three-phase approach to searching the literature was employed. Initial literature searches were conducted by one author (L.M.) using PubMed, Science Direct (Elsevier) PsychINFO, and Google Scholar, in line with current recommendations (9). Key words were used for each database and combined with Boolean operators AND and OR. A second search of each database was performed using different terms. All search terms are outlined in Table 1 . Finally, a citation search was performed to identify studies that may not have been captured in the search terms. Search Terms Selected articles included all peer-reviewed journal publications published between January 01, 2009, and January 01, 2019. Studies were restricted to English language articles, and those where full text was not available were excluded. Opinion pieces, editorials, and nonpeer-reviewed publications were also excluded. An article was deemed to display misogyny if it discussed/assessed the female aesthetic in terms of health, but either did not measure health, or if the research could not be deemed useful or beneficial for clinical practice. A total of 30 full texts were assessed by two authors using stringent inclusion and exclusion criteria ( Table 2 ); a further 13 were excluded, of these 6 were editorials or reports rather than research, 2 were based on the gender gap in scientists rather than patients, and the remaining 5 examined clinical care and interventions rather than the presence of a gender gap in the literature. A total of 17 articles were included in the qualitative analysis. Inclusion and Exclusion Criteria The results were merged and duplicates removed using Endnote X9.3. Data were collected from the 17 included articles using a data extraction tool ( Table2 ) to collect the study name, lead author and year, methods, aims and objectives, and results. Aims and objectives were taken as direct quotes from the articles where possible, to avoid any interpretation bias. Analysis was conducted according to the 2009 Ppreferred Rreporting Iitems for Ssystematic Rreviews and Mmeta-Aanalysis (PRISMA) checklist. 24 As demonstrated in the PRISMA flow chart ( Fig. 1 ), 2176 articles were identified by the study search, 2146 were excluded due to lack of relevance. Thirty full texts were assessed for eligibility, with 13 excluded as they did not meet the inclusion criteria. Seventeen articles were included in the analysis. PRISMA flow chart of the search strategy. PRISMA, Ppreferred Rreporting Iitems for Ssystematic Rreviews and Mmeta-Aanalysis.

Of the 17 included articles, 12 examined the gender gap in medical research and 5 demonstrated misogyny, assessing female attractiveness for alleged medical reasons. Three studies were reviews of the published literature 25–27 defined as collation of all empirical evidence that fits prespecified eligibility criteria, 28 nine were cross-sectional analyses 29–38 (observational studies of the published literature over a defined time period 39 ), and the five remaining studies were within-subject experimental design. 40–43 Of the 12 studies examining gender gaps ( Table 3 ), 11 were conducted in the United States 25 , 26 , 29–33 , 35–38 and 1 in Canada. 34 Results: Representation of Female Participants in the Contemporary Medical Literature BLA, biological license application; CVD, cardiovascular disease; LPCT, late phase clinical trial; RCT, randomized controlled trial. The five experimental design studies were the same five studies that demonstrate misogyny ( Table 4 ). These five studies were conducted in the United Kingdom ( n  = 2), 40 , 41 Poland ( n  = 2), 42 , 43 and Italy ( n  = 1). 21 Results are presented in Tables 3 and 4 . Results: Evidence of Misogyny in the Contemporary Medical Literature Three themes were identified: females remain under-represented in biomedical research, sex and gender are poorly analyzed and reported in research, and several contemporary research articles display ideas that can be construed as misogynistic. A total of seven published articles analyzed the proportion of females–males participating in research trials. 25 , 30–32 , 34–36 Under-representation of women was noted, largely because of discrepancies between specialities, with some specialist fields recruiting more female participants than male participants, and others recruiting fewer females than males. Kong et al. observed that in cell biology, 60% of studies gave no information on the sex of the cells studied. 33 Two studies noted that specifically cardiovascular research appeared to under-represent women. 25 , 32 Six published articles examined the analysis and reporting of results by sex or gender. 26 , 29 , 31 , 33 , 34 , 38 These articles found that analysis by sex or gender was somewhere in the range of 6%–38%. Sugimoto et al. noted that articles with authors' with female names were more likely to report results by sex or gender. 38 Five articles displayed ideas that could be perceived as misogynistic 21 , 40–43 with three of these five studies stating that they assess female attractiveness as a marker of health. 21 , 40 , 42 Pietruski et al. conducted an analysis of visual tracking of the aesthetic of breasts, however, also stated their method is potentially flawed in that there is limited evidence as to how visual patterns translate to assessment of attractiveness. 43

Amid calls for Australian research policy to align with those in Europe and United States and increase equality in sex and gender recruitment in medical research, 1 the sex and gender gap in medical practice is drawing increasing media attention. 2 , 3 Females account for >50% of the global population and, therefore, a significant proportion of the patient population, 4 yet women wait longer than men for both a diagnosis 5 and pain relief, and 6 , 7 are more likely to be misdiagnosed or discharged during serious medical events. 5 Historically, medical studies have excluded female participants and research data have been collected from males and subsequently generalized to females 8 and those who are intersex and do not have the reproductive anatomy characteristic of female or male. 9 There are several postulated reasons for preferring males in research, including concerns for decreasing fertility or harming pregnancy, 10 researcher bias from predominantly male researchers, 11 and perception of the male as representative of the human species and, therefore, the norm. 12 The gender gap in medical research, alongside overarching misogyny, results in real-life implications for female patients. 13 The “Yentl syndrome,” named after the central protagonist in the 19th century story by Isaac Beshevis Singer, is the phrase coined by Dr. Bernadette Healy that describes how, for a woman's illness to be taken seriously, she must prove herself to be as unwell as a male counterpart. 14 Yentl syndrome as a cause of delayed care for female patients is well documented in the medical literature. 13 Lack of research evidence from female patients may result in delay in treatment; applying inappropriate, ineffective, or harmful treatments; or the withholding of effective treatments. The resulting detriment to women's health is evidenced in the contemporary literature; recent studies have demonstrated that women with acute myocardial infarction (MI) present with different symptoms to men and are less likely to be identified during angiography than men, simply because they are often underinvestigated and subsequently less likely to be managed appropriately. 15 Six-months after admission for MI, female patients, especially younger women, were more likely to suffer major adverse events and overall mortality. 15 , 16 Androcentricity in medical research has historically disadvantaged and damaged female patients, from inaccurate diagnoses of “hysteria” and related barbaric treatments such as clitoridectomies and extended periods of enforced bedrest, to the more modern discrepancies observed in diagnoses and management of female patients. 17 The hysterical discourse is often used colloquially, with terms such as “mad” and “crazy” used to describe “difficult” women who do not respond to treatment or diagnosis as expected. 18 Young et al. examined the discourse surrounding patients with endometriosis, determining that the pain and experiences of these patients were often dismissed as psychological. 18 In keeping with this, a recent study in the United States determined an average time to diagnosis of endometriosis from first consultation was 4.4 years and this was shorter in women who were aged <18 years and 40–49 years and those referred to and investigated by gynecologists. 19 The concept of females being “difficult” is embedded in societally and culturally ingrained misogyny. 17 Historically, the perception of beauty has been used to oppress women, whereby maintenance of fertility and aesthetics was of paramount importance. The use of beauty as a tool of oppression has become known as the “beauty myth” and was analyzed by feminist writer Naomi Wolf in her book “The Beauty Myth: how images of beauty are used against women.” 20 This myth is evident in the medical literature today, highlighted by recent mainstream media controversies over doctors rating the aesthetic attractiveness of their endometriosis patients. 2 , 21 Wolf asserts that the “beauty myth” is a tool of oppression and a political weapon against women. 20 The “hidden curriculum” may be an important contributor to the outcomes of the androcentricity of medical research on the experiences of female patients. The hidden curriculum pertains to the aspects of medicine that are not formally taught to medical students: the attitudes and values collected from their experiences with senior clinicians on the wards and in general practice. Perceptions from androcentric medical research, whereby female patients do not fit the male mould and, therefore, are “difficult” or “mad,” may be passed down through the hidden curriculum to junior staff members, thereby continuing the cycle. 22 In addition, medical students arguably become acculturated to the historically masculine medical environment that demands conformity. 18 As more is learnt about the gender gap and misogyny in medical research, and the impact this has upon female patients, it is important to determine whether misogyny continues to permeate the narrative in modern medicine. It is also important, while examining misogyny in medical research, to realize that sex and gender are distinct, yet the terms are often used interchangeably. Sex refers to the biological and physiological characteristics that define humans as male, female, or intersex. Gender, rather, is a societal construct that refers to roles, activities, and behaviors, and encompasses a wide range of identities beyond male, female, and intersex. 23 In this systematic scoping review, we aim to explore the extent of study into sex and gender gaps in the published literature and assess whether misogynistic characterization is prevalent in contemporary medical research.

This review highlights several issues regarding sex and gender in the medical literature. Females remain broadly under-represented in the medical literature, sex and gender are poorly reported and inadequately analyzed in research, and misogynistic articles continue to permeate the narrative. Several studies determined that women are largely under-represented in medical research. Geller et al. analyzed 86 randomized controlled trials (RCTs) across 9 journals and found female representation to be just 37%, with only 3 studies noting the limitations of lack of diversity. 29 Perception of the disease, rather than actual sex prevalence, appears to drive the representation of females in medical research. 44 For diseases that are perceived to affect men more than women, androcentricity dominates the research picture. Melloni et al. examined the representation of women in cardiovascular disease prevention RCTs. Although most studies in Melloni's analysis recruited both sexes and the proportion of women participants was noted to be increasing, there were still 20 studies noted to have recruited only male participants, versus just 1 study with only female participants recruited. Importantly, female representation was higher in diseases perceived to affect women more, such as hypertension and stroke. 25 Similarly, Scott et al. also examined the participation of women in cardiovascular drug trials, and determined that in diseases believed to affect women in greater numbers than men, such as hypertension, atrial fibrillation, and pulmonary hypertension, women were either adequately represented or over-represented. In diseases perceived to affect greater number of men than women, such as coronary disease and heart failure, females were under-represented. 32 Kalliainen et al. noted that the female–male recruitment ratio in hand surgery research was dependent on the pathology, but the ratios recruited were roughly in keeping with sex prevalence of the disease, 35 noting a need to increase members of the less-represented sex to enhance statistical power. The 1993 National Institutes of Health Revitalisation Act recommends that women and men be included in clinical trials based on the sex prevalence of the disease and to provide data on the efficacy of treatment in each sex, 45 which may help account for the under- and over-representation of females depending on disease. However, the perceptions of sex-related prevalence appear frequently to be outdated; for example, coronary artery disease is the commonest cause of death in both men and women 46 and women experience greater functional disability and symptoms burden and a higher prevalence of nonobstructive coronary artery disease than men. 47 Similarly, women make up 30% of the gout disease population; however, they make up just 5.3% of gout clinical drug trial participants. 30 Geller et al. conducted a cross-sectional study of RCTs across 14 journals published in 2015 and determined that the guidelines from the 1993 National Institutes of Health Revitalisation Act have not resulted in significant increases in reporting results by sex. 36 Indeed, Poon et al. noted in a similar study that female participation in clinical trials has decreased since the 1990s. 30 Not only is it important to acknowledge outdated gender-based beliefs, the representation and participation of women in medical research are important because medical research informs the development of clinical guidelines. Clinical guidelines directly impact the lives of patients, therefore, if there are sex and gender differences in presentation, management, and clinical response to management, it is vital these are described. Melloni et al. conducted an analysis of published RCTs and determined that in clinical trials used to inform guidelines for cardiovascular disease prevention in women, female participation was just 30%. 25 Kannan et al. explored recruitment bias using a cross-sectional study of patients volunteering for research through an online portal and concluded that under-representation of women in clinical trials is not due to the unwillingness of women to volunteer, but rather owing to bias within trial design and recruitment. 37 Lack of female representation in research was noted consistently throughout review of the literature, however, another related issue was the lack of sex reporting or inclusion in analysis of many medical and biomedical studies. Kong et al. evaluated sex bias in published dermatological research, much of which is on nonhumans such as cell lines or animals; 60% of the cellular studies from both humans and animals provided no information regarding sex. Where sex was declared, 70% of the cell lines studied were from males. 33 Welch et al. supported this observation in a cross-sectional study of RCTs in Canada, finding that no studies considered the influence of sex and only 6% of studies performed a subgroup analysis for sex. 34 Similar findings are noted consistently in other studies, and sex reporting appears particularly poor in animal and cellular studies. 26 , 31 , 38 Analyses of results by sex are seemingly poor across all study types. 26 , 31 , 34 Some of the gender discrepancy has been explained by Hankivsky et al.; examination of statements surrounding sex and gender inclusion from 45 health-research funding agencies and 10 sex/gender health journals determined there is little consistency in whether sex and gender are mentioned in funding and publication guidelines. There is also significant variation in the conceptualization of sex and gender and how researchers address this in research. The criteria set by agencies fail to address the complex relationship between sex and gender and health. 27 A total of five studies exhibited misogyny. Vercellini et al. utilized a likert scale for clinicians to subjectively assess the attractiveness of women undergoing surgery for endometriosis. 21 Although this publication caused outrage in the global media and across social platforms such as Facebook and Twitter, 2 it is far from unique. Three studies proposed to assess the attractiveness of the female face as a measure of health, however, none of them utilized objective measures of health. 40–42 Marcinkowska et al. determined that women with higher parity are perceived as less attractive and healthy, but the study did not provide information regarding objective measures of health from which this conclusion was generated. 42 A further study by Pietruski et al. applied eye-tracking technology to ascertain the attractiveness of female breasts in the context of reconstructive surgery; however, authors acknowledged that the link between lingering gaze and attractiveness is unknown, 43 raising the question of why this study was conducted. Studies displaying misogyny are problematic for clinical medicine and patient care for several reasons. First, they continue to perpetuate the concept of the “beauty myth” and the utilization of the female aesthetic as a tool of oppression and prejudice. In highlighting female beauty, the authors of the articles inadvertently suggest that patient care should be altered based on the aesthetic of the patient. Despite the findings of Vercellini et al. that attractive women are more likely to have severe endometriosis, 21 it is unlikely that this will be incorporated into clinical practice given the subjective nature of the assessment. Consequentially, studies that focus on the female aesthetic add little to the care of women and may be deemed unethical. Furthermore, the beauty myth has been debunked, evidenced by the evolving nature of society's preference of female aesthetics, from the voluptuous women of the 1940s and 1950s to the emaciated females of the 1970s to 1990s, and finally to the contemporary fashion of fitness and athleticism as beautiful. As Wolf writes, “there is no legitimate justification for the beauty myth,” but rather the emphasis on beauty is about power and patriarchy, where the aging woman is most feared as advancing age is associated with increased knowledge, power, and respect. 20 Possibly as a result of the beauty myth, women's identities may be entrenched in their physical attractiveness, which ensures their vulnerability to external approval. 20 Failure to achieve the gold standards to beauty impacts on self-esteem and possibly subsequent success. 2 , 48 , 49 Studies focusing on female attractiveness may serve to reinforce false beliefs surrounding beauty and increase hostility toward women. 50 Hostile sexism may lead to perpetuation of the hysterical discourse 51 and increase time to diagnosis and occurrence of misdiagnoses. Although sexism remains present in the medical sphere, patient care is compromised. The research gap and the publication of misogynistic research may adversely affect female care and contribute to the negative perceptions of female hysteria and the gap in time to diagnosis experienced by many women. The androcentric history of medical research led to assumptions about women's health and response to treatments based solely on studies from male bodies. 10 , 17 A rapidly growing body of evidence from clinical research demonstrates that females and males can differ greatly in their susceptibility and presentation of disease and their response to treatment and profile of adverse effects. 23 A more immediate issue with studies that display misogyny is the trust patients place in the medical profession. Women reportedly feel a sense of distrust in the medical profession, with many professing acute awareness that doctors have the power to label them as “anxious” or “depressed” rather than address their physical pain or symptoms. 17 Physicians assessing physical attractiveness as part of a clinical assessment is arguably entering dangerous territory when it comes to the doctor–patient relationship. 2 Doctors are bound by the Hippocratic Oath and a set of laws from registering boards, under which the boundary of relations with patients is clear and absolute. 52 Rating patient attractiveness is, therefore, both inappropriate and extremely subjective, consequentially it must not be allowed to blur the clinical picture. Other countries have developed research policies to address the sex and gender imbalance in research and to stipulate that sex must be reported; Australia has somewhat lagged on this. 1 , 23 Until we address the gender gap and misogyny in medical research, we run the risk of the health care of female patients remaining substandard to the care of male patients. 53 The consequences of neglecting sex and gender differences are wide reaching—including inaccuracies in science, adverse health outcomes, and experiences and cost ineffectiveness. 23 , 54 This review has a few limitations, first in that sex and gender are largely considered in binary terms and there is no discussion of intersexuality, nonbinary, transgender, or any others on the gender spectrum because there is very limited literature on gender gaps in these population groups, and the authors highlight a need for further research in this area. Second, only articles from the recent decade were included, this was to allow the review to examine only the most contemporary literature, however, it is likely that more historical review would yield more evidence of misogyny and gaps in gender analysis. Finally, the definition of misogyny in research was determined by the authors, because there was no literature previously examining this topic in the medical literature, there is no gold-standard way to measure the misogyny of a research article. In much the same way as the Bechdel test for positive female representation within media, 55 the authors propose it may be of use to develop a tool to assess misogyny in medical research and to provide guidelines to avoid future publication of misogynistic research.

The gender gap and misogynistic studies, which serve little to improve women's health, remain present in the contemporary medical literature. There may be several drivers for the gender gaps and misogyny revealed by this scoping review, including outdated perceptions of sex prevalence of diseases, unhelpful discourse surrounding female patients and illness, and the ongoing perpetuation of the beauty myth. The hidden curriculum in medical education may be an important and seemingly unexplored contributor to the outcomes of the androcentricity of medical research on the experiences of female patients.