Meta-Research: A 10-year follow-up study of sex inclusion in the biological sciences
Abstract
In 2016, to address the historical overrepresentation of male subjects in biomedical research, the US National Institutes of Health implemented a policy requiring investigators to consider sex as a biological variable. In order to assess the impact of this policy, we conducted a bibliometric analysis across nine biological disciplines for papers published in 34 journals in 2019, and compared our results with those of a similar study carried out by Beery and Zucker in 2009. There was a significant increase in the proportion of studies that included both sexes across all nine disciplines, but in eight of the disciplines there was no change in the proportion studies that included data analyzed by sex. The majority of studies failed to provide rationale for single-sex studies or the lack of sex-based analyses, and those that did relied on misconceptions surrounding the hormonal variability of females. Together, these data demonstrate that while sex-inclusive research practices are more commonplace, there are still gaps in analyses and reporting of data by sex in many biological disciplines.
Introduction
Studies of both males and females are essential to the advancement of human health, and the influences of sex on the prevalence, presentation, and progression of many disease states is profound. Yet, within the biological sciences, it has been a common and preferential practice to utilize male research subjects in basic and pre-clinical research (Beery and Zucker, 2011; Kong et al., 2016; Sugimoto et al., 2019; Yoon et al., 2014). This male bias stems from the misconception that female animals increase experimental variability due to cyclical fluctuating hormones and the historical belief that no major differences exist between the sexes outside of reproductive functions (Institute of Medicine, 2001). These biases are not limited to the basic sciences, but extend into clinical research as well (Geller et al., 2018; Mansukhani et al., 2016; Prakash et al., 2018; Scott et al., 2018).
Initial reports calling for the inclusion of females in research and which describe the limitations of sex-biased studies began in the 1990s and extended in to the early 2000s (Berkley, 1992; Holdcroft, 2007; Mogil and Chanda, 2005). In 2009, Beery and Zucker conducted a multi-disciplinary review of primary literature which quantified the extent of sex-bias across several research areas in the biological sciences (Beery and Zucker, 2011). Since that report, there have been numerous calls to address this issue through sex-inclusive research practices and policies (Kim et al., 2010; Klein et al., 2015; Mazure and Jones, 2015; Woodruff, 2014), culminating in 2016 when the National Institutes of Health (NIH) in the United States implemented a policy requiring investigators to consider sex as a biological variable (Clayton and Collins, 2014). The intent of the policy is to ensure equal representation of males and females in vertebrate research studies, unless there is significant justification to support the use of a single-sex. Many lauded the policy (Mogil, 2016; Shansky and Woolley, 2016), yet there were still those who saw it as unnecessary and feared that it would be time consuming, costly, increase experimental variability, and require expertise in the study of sex differences (Woitowich and Woodruff, 2019). Considering sex as a biological variable does not require investigators to power studies in order to determine sex differences nor does it ask investigators to analyze data by sex. Yet, these common misconceptions persist, despite clarifications and guidance surrounding sex-inclusive research practices (Arnegard et al., 2020; Becker et al., 2016; Clayton, 2018; Miller et al., 2017; Shansky, 2019). Recently, several studies have monitored the progress of sex-inclusive research practices following NIH policy implementation in the fields of microbiology and immunology (Potluri et al., 2017), as well as neuroscience (Mamlouk et al., 2020) utilizing methodologies similar to Beery and Zucker. Here, we present a 10 year follow-up study to the initial Beery and Zucker report by conducting a systematic review to assess sex-inclusive research practices within nine of the biological disciplines and 34 of the scholarly journals originally surveyed in 2009. We provide an updated perspective on the state of sex-inclusive research within the biological sciences, and highlight areas of improvement alongside shortcomings in the decade since Beery and Zucker conducted their original study.
Results
In 2009, Beery and Zucker conducted a bibliometric analysis of 841 articles from high-impact journals, across ten biological disciplines which quantified the extent of male-bias in research and a noted lack of sex-based analyses when males and females were both included as research subjects (Beery and Zucker, 2011). We recapitulated the work of Beery and Zucker utilizing a similar bibliometric analysis of 720 journal articles, corresponding to nine of the original disciplines and 34 journals surveyed in 2009 (Table 1).
Journals surveyed by subject area in 2009 and 2019.
Discipline | Journal A | Journal B | Journal C | Journal D |
---|---|---|---|---|
General Biology | PLoS Biology | Proceedings of the Royal Society B: Biological Sciences | Nature | Science |
Immunology | Journal of Immunology | Infection and Immunity | Immunity | Vaccine |
Neuroscience | Journal of Neuroscience | Neuroscience | The Journal of Comparative Neurology | Nature Neuroscience |
Physiology | Journal of Physiology (London) | American Journal of Physiology – Renal Physiology | American Journal of Physiology – Gastrointestinal and Liver Physiology | American Journal of Physiology – Heart and Circulatory Physiology |
Pharmacology | Neuropsycho- pharmacology | Journal of Psychopharma- cology | The Journal of Pharmacology and Experimental Therapeutics | British Journal of Pharmacology |
Reproduction | Biology of Reproduction | Reproduction | ||
Endocrinology | European Journal of Endocrinology | Journal of Neuroendo- crinology | Endocrinology | American Journal of Physiology – Endocrinology and Metabolism |
Behavioral Physiology | Journal of Comparative Psychology | Behavioral Neuroscience | Physiology and Behavior | Hormones and Behavior |
Behavior | Behavioral Ecology and Sociobiology | Animal Behaviour | Animal Cognition | Behavioral Ecology |
Subject sex across disciplines
In 2019, 49% (n = 356) of studies reported using both male and female research subjects, resulting in a significant increase in sex inclusion demographics compared to 28% of articles surveyed 2009 (n = 232, p<0.0001; Figure 1A). Six of the nine disciplines demonstrated a significant increase in the use of both sexes (Figure 1). Between 2009 and 2019, the largest increases in sex-inclusive studies were seen in the fields of neuroscience (29% vs. 63%, p<0.0001) and immunology (16% vs. 46%, p<0.0001), followed by endocrinology (30% vs. 56%, p=0.001), general biology (34% vs. 59%, p=0.002), physiology (13% vs. 36%, p=0.001), and behavioral physiology (43% vs. 61%, p=0.018). In reproduction, single-sex studies remained the norm, and studies of both males and females increased only marginally (10% vs. 14%, p=0.35), while the number of female only research studies increased, corresponding to a female to male subject ratio of 1.6:1 in 2009 to 3.6:1 in 2019. Behavior remained the most inclusive biological discipline with 70% and 81% of studies reporting the use of both sexes in 2009 and 2019, respectively, largely driven by sex-inclusive field studies. Pharmacology was the only field to trend downward with 29% of articles reporting the use of both sexes in 2019 compared to 33% in 2009 (p=0.607). Likewise, there was an increase in the male to female subject ratio from 5:1 in 2009 to 5.8:1 in 2019.
![](https://iiif.elifesciences.org/lax/56344%2Felife-56344-fig1-v1.tif/full/617,/0/default.jpg)
Comparison of studies by field, sex, and sex-based analyses in 2009 and 2019.
(A). The proportion of articles surveyed in 2009 and 2019 which utilized male subjects, female subjects, both male and female subjects, or those that did not specify the sex of the subjects. Data are presented by individual biological discipline as well as by the sum of all nine disciplines. (B). The percentage of articles surveyed in 2009 and 2019 which utilized both male and female subjects and conducted sex-based analyses, either by including sex as a covariate or by subgroup analyses. Data are presented by individual biological discipline as well as by the sum of all nine disciplines. The source data for this figure are in Supplementary file 1.
Sex based analyses by discipline
For articles that reported the inclusion of both sexes in 2019, data were collected on whether or not the authors conducted sex-based analyses. Out of 356 of the journal articles which used both sexes in 2019, only 42% analyzed data by sex, compared to 50% in 2009 (n = 117, p=0.3; Figure 1B). Pharmacology was the only biological discipline to demonstrate a significant increase in sex-based analyses from 19% in 2009 to 48% in 2019 (p=0.033; Figure 1B).
Description of sample size by sex across disciplines
For articles that reported the inclusion of both sexes in 2019, data were collected on whether the authors provided a description of the sample size (n) by sex. Out of the 356 articles that used both sexes, 27% failed to provide a description of the sample size by sex (Figure 2A). Neuroscience articles failed to provide a description of the sample size by sex 52% (n = 26) of the time, along with general biology at 47% (n = 22) and immunology at 43% (n = 19).
![](https://iiif.elifesciences.org/lax/56344%2Felife-56344-fig2-v1.tif/full/617,/0/default.jpg)
Percent of articles which provided the sample size (n) by sex, rationale for single-sex studies, or rationale for the lack of sex-based analyses.
(A). The percentage of articles which utilized both male and female subjects and provided a description of the sample size by sex. Data are presented by individual biological discipline as well as by the sum of all nine disciplines. (B). Categorization of articles which provided rationale for single sex-studies or the lack of sex-based analyses (n = 30) into seven distinct themes. Each theme includes representative rationale derived from the experimental methods. The source data for this figure are in Supplementary file 1.
Rationale for single sex studies or lack of sex-based analyses
For all 720 articles analyzed in 2019, data were collected on whether the authors provided a justification for the use of a single sex or rationale for the lack of sex-based analyses. Thirty articles included a range of explanations related to sex-inclusion and sex-based analyses (Figure 2B). Justifications for single sex studies included: a priori knowledge of sex-differences or sex-specific effects (n = 9), the potential for increased experimental variability (n = 8), experimental conditions which limited the use of both sexes (n = 4) and difficulties in animal husbandry (n = 2). Rationale for the lack of sex-based analyses included: limited sample sizes to determine statistical significance (n = 4) or an inability to determine the sex of the subject (n = 3). Only two studies specified that the authors did not identify any sex differences, so the dataset was analyzed in aggregate.
Discussion
Notably, the number of sex-inclusive research studies has significantly increased across most biological disciplines. At face value, this change is encouraging and suggests that the scientific community may have an increased awareness and understanding of the need for sex-inclusive research and its contribution to experimental rigor and reproducibility (Clayton, 2018; Miller et al., 2017). At the same time, close to one third of all research studies that utilized both male and female subjects failed to quantify their sample size by sex. Ironically, this is most prevalent in the fields which reported the greatest increases in sex-inclusive research (ex. neuroscience, immunology, and general biology) At best, this result indicates that investigators may not think it is important to provide a description of the sample size by sex in the absence of sex-based analyses. In a less ideal case, the representation of males and females is not well balanced, and this may be intentionally obscured. Single-sex studies are valid and warranted, provided there is evidence-based rationale for the case. Yet, several studies explicitly stated that they excluded both sexes as a means to prevent experimental variability, which is an erroneous belief and unsound research practice (Beery, 2018; Prendergast et al., 2014).
Perhaps most concerning, improvements in the inclusion of both sexes over the past decade have not been accompanied by general improvement in sex-based analyses, despite repeated calls and guidelines for such analyses (Beltz et al., 2019; Clayton, 2018; Clayton and Collins, 2014; Hankivsky et al., 2018; Prager, 2017). Sex-based analyses may uncover sex differences for a given trait, prompting the development of sex-specific prevention strategies, drug targets, or other therapies beneficial to both sexes (Yang et al., 2019). And while it is reasonable to aggregate and analyze data from both sexes if it has been established that there are no sex-differences for a given trait or condition, out of the 720 articles reviewed here, only two conveyed this information in their methods. When this information is lacking, the reader is tasked with making the assumption that either there are no sex differences or that sex-differences have yet to be examined. In either case, this can lead to redundant research efforts requiring additional time, money, and biological resources.
The data presented here highlight a continued need for education, awareness, and advocacy surrounding sex-based research practices including the consideration of sex as a biological variable. We call upon academic publishers to require a description of sex, rationale for single-sex studies or lack of sex-based analyses in the experimental methods. In the absence of formal policies, reviewers can ask for these essential criteria. In addition, funders can also contribute to the advancement of rigorous sex-inclusive science by requiring grant proposals to include appropriate sex-based reporting and analyses and determine funding success on the evaluation of sex and other key biological variables. Lastly, we call upon universities to encourage the consideration of sex as a biological variable through institutional review boards (IRBs) and institutional animal care and use committees (IACUC) oversight (Duffy et al., 2020) and by providing instruction to biomedical trainees on sex-inclusion, reporting, and analyses through established responsible conduct of research modules and within medical school curricula. Only together, through concerted, tripartite efforts at the institutional, funder, and publishing levels will the consideration of sex as a biological variable become standard practice (Tannenbaum et al., 2019). Together, this will allow us to improve our understanding of health and disease for both men and women and to further the reality of personalized medicine.
Methods
A systematic sampling of journal articles from 2019 was conducted using the methodologies originally described in Beery and Zucker, 2011. All articles were reviewed and coded by one of us (NCW) in order to minimize coding bias. Briefly, journal articles were assessed for sex-inclusive research practices from nine biological disciplines and 34 journals sampled by Beery and Zucker, 2011. These disciplines included: General Biology, Immunology, Neuroscience, Physiology, Pharmacology, Reproduction, Endocrinology, Behavioral Physiology, and Behavior. Zoology, which was studied by Beery and Zucker, was excluded here due to a limited number of mammalian studies available to survey at the time of manuscript preparation. Four journals were selected to represent each discipline, with the exception of Reproduction (Table 1). For each journal, the first 20 primary research articles which met eligibility criteria were surveyed in 2019. For the two reproductive biology journals, the first 40 journal articles were surveyed for 2019. For the majority of disciplines, the first 20 research articles which met eligibility criteria were published between January and April of 2019, whereas articles from other disciplines were published between January through June (Endocrinology), August (Behavioral Physiology) and October (Behavior) of 2019.
The eligibility criteria for studies in this analysis were as follows.
Inclusion criteria (all criteria required): i) Reported use of any vertebrate mammal in some part of the experimental methods, including those which describe the generation of primary cell culture; ii) Published after January 1 st, 2019; iii) Published in the English language.
Exclusion criteria (each criterion can exclude): i) Type of article: review articles, brief communications, or viewpoints; ii) Articles published in a special or themed issue; iii) Reports utilizing fetal organisms or those restricted to immortal cell lines.
When journals were arranged by subtopics, articles were sampled evenly across several topics. In journals such as Nature and Science, only articles pertaining to the biological sciences were considered.
Articles were coded for sex. Sex was recorded as male, female, both sexes, or unspecified. Following the strategy of Beery and Zucker, 2011, coding was biased in favor of inclusivity and articles were categorized as using both sexes when different parts of a study utilized different sexes. Likewise, field studies were categorized as investigating both sexes when this was explicitly noted or could be inferred by the methods provided. Articles which utilized both sexes were further evaluated for a description of the sample size by sex and whether data were analyzed by sex, including sex as a covariate or subgroup analyses by sex. For all articles reviewed, we noted if the authors provided rationale for the use of a single sex or the lack of sex-based analyses.
Data analyses were primarily qualitative, with a small quantitative component. Descriptive statistics were used where appropriate. Nominal data were described as n (%). We compared the 2019 data to 2009 data in Beery and Zucker, 2011. Chi-squared tests were used to assess differences between the use of both sexes in 2009 compared to 2019, and the number of studies which analyzed data by sex in 2009 compared to 2019 (GraphPad Prism, version 7.0). p-values<0.05 were considered significant.
Data availability
All data generated or analysed during this study are included in the manuscript and supporting files. Source data files have been provided for Figures 1 and 2.
References
-
Inclusion of females does not increase variability in rodent research studiesCurrent Opinion in Behavioral Sciences 23:143–149.https://doi.org/10.1016/j.cobeha.2018.06.016
-
Sex bias in neuroscience and biomedical researchNeuroscience & Biobehavioral Reviews 35:565–572.https://doi.org/10.1016/j.neubiorev.2010.07.002
-
Analysis of sex differences in pre-clinical and clinical data setsNeuropsychopharmacology 44:2155–2158.https://doi.org/10.1038/s41386-019-0524-3
-
Vive la différence!Trends in Neurosciences 15:331–332.https://doi.org/10.1016/0166-2236(92)90048-D
-
The more things change, the more they stay the same: a study to evaluate compliance with inclusion and assessment of women and minorities in randomized controlled trialsAcademic Medicine : Journal of the Association of American Medical Colleges 93:630–635.https://doi.org/10.1097/ACM.0000000000002027
-
Beyond sex and gender difference in funding and reporting of health researchResearch Integrity and Peer Review 3:6.https://doi.org/10.1186/s41073-018-0050-6
-
Gender bias in research: how does it affect evidence based medicine?Journal of the Royal Society of Medicine 100:2–3.https://doi.org/10.1177/014107680710000102
-
BookExploring the Biological Contributions to Human Health: Does Sex Matter?Washington, DC: National Academies Press.https://doi.org/10.1089/152460901300233902
-
Mind the gap: sex bias in basic skin researchJournal of Investigative Dermatology 136:12–14.https://doi.org/10.1038/JID.2015.298
-
Sex bias and omission in neuroscience research is influenced by research model and journal, but not reported NIH fundingFrontiers in Neuroendocrinology 57:100835.https://doi.org/10.1016/j.yfrne.2020.100835
-
Considering sex as a biological variable in preclinical researchThe FASEB Journal 31:29–34.https://doi.org/10.1096/fj.201600781r
-
Addressing sex as a biological variableJournal of Neuroscience Research 95:11.https://doi.org/10.1002/jnr.23979
-
Sex bias in interventional clinical trialsJournal of Women's Health 27:1342–1348.https://doi.org/10.1089/jwh.2017.6873
-
Female mice liberated for inclusion in neuroscience and biomedical researchNeuroscience & Biobehavioral Reviews 40:1–5.https://doi.org/10.1016/j.neubiorev.2014.01.001
-
Participation of women in clinical trials supporting FDA approval of cardiovascular drugsJournal of the American College of Cardiology 71:1960–1969.https://doi.org/10.1016/j.jacc.2018.02.070
-
Considering sex as a biological variable will be valuable for neuroscience researchThe Journal of Neuroscience 36:11817–11822.https://doi.org/10.1523/JNEUROSCI.1390-16.2016
-
Implementation of the NIH sex-inclusion policy: attitudes and opinions of study section membersJournal of Women's Health 28:9–16.https://doi.org/10.1089/jwh.2018.7396
-
Sex differences in GBM revealed by analysis of patient imaging, transcriptome, and survival dataScience Translational Medicine 11:eaao5253.https://doi.org/10.1126/scitranslmed.aao5253
Decision letter
-
Cassidy SugimotoReviewing Editor; Indiana University Bloomington, United States
-
Peter RodgersSenior Editor; eLife, United Kingdom
-
Rebecca ShanskyReviewer
-
Londa SchiebingerReviewer
In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.
Your article has been reviewed by three peer reviewers, and the evaluation has been overseen by a Reviewing Editor (Cassidy Sugimoto) and the eLife Features Editor (Peter Rodgers). The following individuals involved in review of your submission have agreed to reveal their identity: Rebecca Shansky (Reviewer #2); Londa Schiebinger (Reviewer #3).
The reviewers and editors have discussed the reviews and we have drafted this decision to help you prepare a revised submission. We hope you will be able to submit the revised version within two months.
Summary:
This manuscript is a 10-year check in on a seminal paper (Beery & Zucker 2011) that systematically examined sex bias in biomedical animal research. Given the introduction of NIH's "Considering Sex as a Biological Variable" (SABV) mandate in 2016, it is important to assess whether research practices have changed over the past decade.
The authors conducted a survey of 2019 publications in ten major fields, and report that in some fields, inclusion of both sexes has increased but that for the most part this has not been accompanied by an increase in analysis of data by sex or reporting the n's of each sex used. The authors call for journals to require more detailed information about the sex of animals and justification of single sex studies. I especially appreciate the assessment of "excuses" for the single sex studies.
This study is a critical addition to the literature and will help biomedical scientists understand how well SABV is working and what needs to be done to make it even more successful.
This is an important article and it should be published once the following points have been addressed.
Essential revisions:
1) Introduction and Discussion: It was disappointing to see that the introduction and discussion did not address/include several key historical and recently published studies. These studies include some of the first studies that attempted to raise awareness regarding the lack of sex reporting (Berkley 1992; Mogil and Chanda 2005; both published before the initial 2009 study), updates to the 2009 study that robustly analyzed key fields that were documented as problematic in the 2009 study (Potluri et al., 2017; Mamlouk et al., 2020), and recently published arguments and analyses that are advancing sex-inclusive research (Shanskey, 2019; Galea et al., 2020). There are others as well (eg, Becker et al., 2016). While I am well aware that not every relevant paper can be cited/discussed, I encourage the authors to enhance the robustness of their literature search to help place their important results within the broader context.
2) I would encourage the authors to point out that SABV explicitly states that experiments need not be powered to detect sex differences or that data be analyzed by sex. However, their calls to require reporting of sex n's are well founded and will improve rigor and reproducibility.
3) The authors call upon academic publishers to require sex reporting. They should also call upon funders to require appropriate sex reporting and analysis in grant proposals--and make funding decision on the quality of that (and other) analysis. The authors should also call upon universities (medical schools) to teach sex reporting and analysis in the curriculum. In Tannenbaum et al (Nature 575: 137-146) the present reviewer [LS] and colleagues discuss the three pillars of the science infrastructure: Funding agencies, Universities, and Peer-reviewed journals. I encourage the authors here to add funders and universities to the paragraph where they discuss academic publishers.
4) Methods: "First 20 primary research articles" - Since only the first 20 articles were analyzed does this mean that most of the data was collected in studies published between January-March 2019? It would be a good idea to document this in the methods in case the data from 2019 need to be compared to future analyses that select across the entire year.
5) Methods: Interrater reliability: Did more than one person analyze studies? Were intra and inter-rater reliability controls performed to ensure accurate analysis?
6) Methods: "Organismal mammalian work" - This label is vague and could be divergently interpreted by different readers. Does this just mean whole organism behavioral analysis? Does it include all mammals, including primates and humans, or just rodents? Beyond answering these specific questions, the entire a priori article section criteria needs to be much better documented. This documentation should include how the employed protocol may potentially skew findings.
7) Methods: "Whether data were analyzed be sex." This definition is vague and requires further documentation. What analyses were considered an analysis by sex? For example, if sex was employed as a covariate, was this considered a sex analysis? What if the data or analysis/statistics were not shown? What if data were presented as disaggregated by sex but not further analyzed? Similar to the previous point, this section requires more robust documentation so that the reader can understand how this study was conducted.
https://doi.org/10.7554/eLife.56344.sa1Author response
[We repeat the reviewers’ points here in italic, and include our replies in Roman.]
Essential revisions:
1) Introduction and Discussion: It was disappointing to see that the introduction and discussion did not address/include several key historical and recently published studies. These studies include some of the first studies that attempted to raise awareness regarding the lack of sex reporting (Berkley 1992; Mogil and Chanda 2005; both published before the initial 2009 study), updates to the 2009 study that robustly analyzed key fields that were documented as problematic in the 2009 study (Potluri et al., 2017; Mamlouk et al., 2020), and recently published arguments and analyses that are advancing sex-inclusive research (Shanskey, 2019; Galea et al., 2020). There are others as well (eg, Becker et al., 2016). While I am well aware that not every relevant paper can be cited/discussed, I encourage the authors to enhance the robustness of their literature search to help place their important results within the broader context.
We thank the reviewers for these suggestions and have enhanced our introduction to include a robust review of the literature related to the SABV policy and sex- and gender-inclusive research practices. Which, in turn, highlights the importance of this work by including additional historical context.
2) I would encourage the authors to point out that SABV explicitly states that experiments need not be powered to detect sex differences or that data be analyzed by sex. However, their calls to require reporting of sex n's are well founded and will improve rigor and reproducibility.
We have modified our introduction to ensure the clarification of this common misconception is included and provide more detailed information about the SABV policy.
3) The authors call upon academic publishers to require sex reporting. They should also call upon funders to require appropriate sex reporting and analysis in grant proposals--and make funding decision on the quality of that (and other) analysis. The authors should also call upon universities (medical schools) to teach sex reporting and analysis in the curriculum. In Tannenbaum et al (Nature 575: 137-146) the present reviewer [LS] and colleagues discuss the three pillars of the science infrastructure: Funding agencies, Universities, and Peer-reviewed journals. I encourage the authors here to add funders and universities to the paragraph where they discuss academic publishers.
We have added these additional “calls to action” and included recent work encouraging IRB and IACUC oversight as well (Duffy et al 2020).
4) Methods: "First 20 primary research articles" - Since only the first 20 articles were analyzed does this mean that most of the data was collected in studies published between January-March 2019? It would be a good idea to document this in the methods in case the data from 2019 need to be compared to future analyses that select across the entire year.
We thank the reviewers for this comment and have added this information into the Methods section. For the majority of fields, the articles analyzed here were published between January - April 2019, but some fields extended further into the calendar year, which we noted in the Methods.
5) Methods: Interrater reliability: Did more than one person analyze studies? Were intra and inter-rater reliability controls performed to ensure accurate analysis?
All studies were analyzed by the corresponding author. We have modified the Methods section to reflect this information.
6) Methods: "Organismal mammalian work" - This label is vague and could be divergently interpreted by different readers. Does this just mean whole organism behavioral analysis? Does it include all mammals, including primates and humans, or just rodents? Beyond answering these specific questions, the entire a priori article section criteria needs to be much better documented. This documentation should include how the employed protocol may potentially skew findings.
We updated the Methods section to include a detailed listing of the eligibility criteria for each article included in this study.
7) Methods: "Whether data were analyzed be sex." This definition is vague and requires further documentation. What analyses were considered an analysis by sex? For example, if sex was employed as a covariate, was this considered a sex analysis? What if the data or analysis/statistics were not shown? What if data were presented as disaggregated by sex but not further analyzed? Similar to the previous point, this section requires more robust documentation so that the reader can understand how this study was conducted.
We thank the authors for this comment and have updated the Methods section to reflect a more accurate description of sex-based analyses.
https://doi.org/10.7554/eLife.56344.sa2Article and author information
Author details
Funding
Northwestern University (Women's Health Research Institute)
- Nicole C Woitowich
- Teresa Woodruff
The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
Acknowledgements
We thank Dr Irving Zucker for his encouragement to conduct this study.
Publication history
- Received:
- Accepted:
- Version of Record published:
Copyright
© 2020, Woitowich et al.
This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.
Metrics
-
- 8,234
- views
-
- 715
- downloads
-
- 197
- citations
Views, downloads and citations are aggregated across all versions of this paper published by eLife.
Download links
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Further reading
-
Researchers tend to overlook sex differences in experiments and analyses.