Guidelines on the use of sex and gender in cardiovascular research

Until as recently as the early 1990s, women were generally excluded from research studies because of the preconceived and untested notion that sex hormone concentrations in women fluctuated to such an extent that experimental results would be highly variable, resulting in uninterpretable data. In addition, there was the perception that research could pose risk to women of childbearing potential ( 1 , 2 ). It was also believed that results from men could simply be generalized to women, a concept that is not evidence based and which is potentially dangerous to women and gender-diverse individuals, particularly those from diverse ethnic and racial groups ( 3 ). The United States Food and Drug Administration found that among all medications withdrawn from market between 1997 and 2000, more were withdrawn due to harm incurred to women than to men ( 3 , 4 ). In addition, as a result of the lack of focus on sex and gender in human research, there are currently no sex-specific clinical care guidelines ( 4 ). This underscores the importance of understanding and studying sex and gender differences and the need to implement effective and specific strategies for the prevention and treatment of cardiovascular diseases for all people. Although sex and gender are terms often used interchangeably, sex refers to biological attributes (e.g., chromosomes, gonadal organs, and genetics) whereas gender refers to socially, psychologically, and culturally constructed factors that shape behaviors, stereotypes, and attitudes across societies over time ( 5 – 8 ). In genome-wide association studies, the autosomal signals are often studied, and the sex chromosomes are not considered. In addition, sex stratification is not always performed and only adjustments are accounted for in regression models ( 9 ). Sex exists on a spectrum, and includes those who are male and female as well as those who are intersex. Gender also exists on a spectrum, and includes men, women, nonbinary people, and agender people, among additional identities ( 6 ). Where sex may influence the biological differences and effects to a medication or intervention in an individual, gender encompasses a wider spectrum of factors including behavior and social interactions ( 10 , 11 ). Often, sex is also associated with disease outcome(s). For example, despite being more likely to seek medical care, women receive treatments for similar complaints to men less often than men do, because of clinician implicit bias, thus affecting outcome(s) ( 11 , 12 ). Sex and gender influences are also found in clinical care experiences as women are more likely to be misdiagnosed or have symptomology ignored, and are less often prescribed medications and procedures known to have beneficial effects in such diseases as diabetes and heart disease even when clearly indicated ( 13 ). Socioeconomic forces must also be considered because women and gender-diverse people are more likely to experience poverty ( 14 ). Socioeconomic factors also have implications for nutritional differences, which can translate to differences in the gut microbiome. The gut microbiome is increasingly recognized as an important modifier for sex and gender differences ( 15 ). As a point of clarification, because of the lack of standardized definitions for the guidelines, we use woman to include any individual who considers themselves to be a woman. The same consideration is given when referring to men. Because prior studies have not always defined terms within their articles, in general the assumption can be made that the term women refers to cis-gender women (i.e., women whose gender is consistent with their sex assigned at birth) and the term men refers to cis-gender men (i.e., men whose gender is consistent with their sex assigned at birth). We recommend in the future that these terms should be more clearly defined in methods, results, and discussion sections ( 8 , 16 ). Investigators should clearly state terminology used (women/men or female/male) and should be consistent throughout the paper as to not conflate sex and gender, and also to state if transgender individuals are included within the participant groups. For transgender adults using hormone therapy, we refer the reader to published guidelines from the American Heart Association ( 17 ). Since 1993, the National Institutes of Health (NIH) has implemented several initiatives and mandates to include women in research. These initiatives include attempts to enhance reproducibility through rigor and transparency by requiring researchers to account for sex as a biological variable (SABV) in research design, analysis, and reporting for both human and animal studies. Other government-based international research funding agencies, including the Canadian Institutes of Health (CIHR) and the European Commission (EC) also require that researchers integrate sex and gender into biomedical research ( 1 , 18 , 19 ). By 2016, it had become standard to consider sex and gender for human studies, although this standard mainly pertained to binary sex, excluding intersex and gender-diverse individuals. Furthermore, research using cells and animal models had not begun considering SABV. Thus, in 2016 NIH instituted an initiative to encourage researchers to incorporate the use of SABV for animal studies, including research using cell-based assays, an often-overlooked consideration. The expectation of NIH was that all applicants proposing animal studies would factor SABV into their experimental design, data analysis, and reporting—or—provide strong scientific justification for single sex investigation (see NIH NOT-OD-15-102, https://grants.nih.gov/grants/guide/notice-files/not-od-15-102.html ). More recently, journals including American Journal of Physiology-Heart and Circulatory Physiology ( AJP-Heart and Circ ) have supported NIH efforts by mandating that researchers include and consider SABV and gender in their studies unless there is strong scientific justification for an exception ( 20 , 21 ). Despite recent improvements in reporting SABV, sex omission, cisgender, heteronormative, male bias, and lack of analyses on how cell sex may affect the results remain widespread ( 21 ). The following guidelines focus on how to incorporate sex, gender, and their interactions on cardiovascular outcomes, and complement other resources published by the AJP-Heart and Circ ( 20 , 22 , 23 ). We have confined the scope to include cell- to individual-based research and have not included population research, which has been previously addressed ( 24 – 29 ). Although each section has a particular focus, an underlying and common thread that weaves through each segment is the concept to start simple and build from there.

The content is solely the responsibility of the author and does not necessarily represent the official views of any of the funding agencies.

The historical failure to consider sex and gender variables in research, and to discriminate against the inclusion of women or female animals for scientifically unsound reasons, comes at a cost. Knowledge about sex differences and the role of gender in health and disease was lost, and gains in treatment, diagnosis, and management of disease have not been equally realized across sexes and genders. By comparison, women and girls have lagged behind improvements in health outcomes relative to men and boys, and those who do not identify with the binary sex and gender classification system often used in research have been largely ignored. The goals of biomedical and health research are to improve the health of all communities and provide greater opportunities for prolonged life span. The exclusion or omission of groups based on sex or gender serves as an obstacle to these goals; it creates a two-tiered system of research and healthcare that disproportionately rewards some and disadvantages others. Including sex or gender as a research variable will not only improve the health of society, it lays a solid foundation of knowledge, to build future advances in science and health that benefit all. Including SABV in study design and data collection, analysis, and interpretation improves the rigor and reproducibility of a study and improves the likelihood of study results informing clinical decision-making and improving outcomes. The study design and methods used should be considered carefully, acknowledging limitations in the applicability of findings that may arise from the model, methods, and analyses used. We have provided a framework to facilitate incorporation of sex as a biological variable when collecting, analyzing, or reporting data ( Fig. 2 ). In cardiovascular physiology, incorporating both sex and gender into human, animal, and cell studies is a necessary component of optimal study design. We acknowledge that to optimally integrate the recommended guidelines requires a significant amount of time and funding, and as such have also provided minimum requirements to help scientists as we all work toward achieving this goal. Decision tree for how to incorporate both sex and gender according to the provided guidelines. Figure created with a licensed version of BioRender.com. SABV, sex as a biological variable.

Dr. Timur O. Yarovinsky reports consulting fees from CaroGen Corporation outside the submitted work. Dr. Jeffrey R. Bender reports consulting fees from Pfizer and Esperion unrelated to the submitted work. Dr. Megan M. Wenner is a consultant for Orchestra BioMed outside the submitted work. Dr. Nisha Charkoudian is an employee of the U.S. Army. The opinions or assertions contained herein are the private views of the author(s) and are not to be construed as official or as reflecting the views of the U.S. Army or the Department of Defense. Citations of commercial organizations and trade names in this report do not constitute an official Department of the Army endorsement or approval of the products or services of these organizations. None of the other authors has any conflicts of interest, financial or otherwise, to disclose. Merry Lindsey is an editor of American Journal of Physiology-Heart and Circulatory Physiology and was not involved and did not have access to information regarding the peer-review process or final disposition of this article. An alternate editor oversaw the peer-review and decision-making process for this article.

Sex omission and male bias are still widespread in cell experiments ( 21 ), and researchers should consider using established cell lines of both sexes to assess the impact of sex-linked genes on phenotype and functional responses. Authentication of cell line sex should be an integral and routine part of scientific practice since such is recommended or mandated by many funding agencies. The American Type Culture Collection (ATCC) recommends cell authentication when a cell line is acquired, after 10 passages, after preparing a cell bank, or when in doubt. ATCC and other cell lines (e.g., European Collection of Authenticated Cell Cultures and Japanese Collection of Research Bioresources Cell Bank) strive to report the sex of human cell lines using either the sex of the isolate subject or analyses of short tandem repeats (STRs) within the amelogenin genes AMELX and AMELY, which are on X and Y chromosomes, respectively. In some cases, STR analyses of alternative genes, such as DYS319, may be required because of preexisting mutations (including deletions) in X or Y chromosomes or genetic drift following extensive passaging. Research laboratory-based STR authentication can be executed using commercially available kits (e.g., Promega GenePrint 10 and GenePrint 24) or custom-designed primers ( 158 ), comparing the STR profile with reference data at the Cellosaurus resource on the ExPASy server ( https://web.expasy.org/cellosaurus/ ) ( 159 ). However, using a commercial service provider is likely to be more reliable and time efficient. Although authentication of human cell line sex has become widely available and more routine, the sex of animal cell lines remains underreported even by cell repositories, often requiring additional effort on the part of investigators. Given the importance of SABV, that effort is necessary. When the study design involves primary cells passaged in vitro or analyzed immediately ex vivo, sex authentication may be based on that of the cell/tissue donor. If the donor information is not available, STR authentication may fill this gap. Similar to human participant and animal studies, investigators should use donors of both sexes when designing in vitro studies, or at least describe why it is not possible to do so, thereby justifying any bias. When primary cells are used, it is also important to consider the influence of gonadal hormones to which they were exposed to prior isolation. Specifying the donor history (menstrual cycle, history of hormone replacement, or gender affirming therapies), if available, may at least allow estimation of gonadal hormone influences. When the donor history is not available, study design may include analyses of associated blood or urine samples. Alternatively, gonadal hormone exposure biomarker detection could be included in the study design. Exposure of cells to gonadal hormones and hormone-like substances during in vitro culture may also introduce experimental bias with consequential misinterpretation of the results. Therefore, study design and data collection should include such a bias risk assessment, recording serum levels of gonadal hormones in the culture media, as well as concentration of the pH indicator phenol red, which is widely known to have estrogen-like effects. If the cell line sex has been determined, this should be reported in the manuscript accordingly. If the cell line sex has not been determined, it should be described in the study limitation section. If cell lines of both sexes were used, the experimental data should be reported separately with attention given to the sex differences. When possible, gonadal hormone exposure before (donor serum levels or history of exposure) and during (concentration in the culture medium) the studies should be reported. Similarly, exposure to hormone-like substances or efforts to minimize the exposure should be reported to increase study reproducibility. If the information is not available, it should be stated clearly in methods or in the study limitations. Study results and conclusions would be stronger if attention is given to sex-dependent parameters. The potential impacts of sex-linked genetic and/or epigenetic factors and environmental exposure to gonadal hormones and hormone-like substances should be assessed and discussed when possible. Furthermore, the interpretation of gonadal hormone effects on cardiovascular cell studies can be confounded by differential responses of genetically XX- versus XY-derived cells. The biology behind these differences is frequently poorly understood. Establishing clear-cut differences in carefully designed studies analyzing cells derived from donors of both sexes, separately, can lay the foundation for additional mechanistic studies defining molecular determinants of these differential responses. This can lead to paradigm-changing understanding in SABV and, in turn, gender-specific therapeutic approaches in a wide array of pathologic states. In summary, Table 3 provides recommendations for considering sex in cell-based studies. Summary of recommendations for inclusion of SABV in cell-based studies: the 4 Cs “NIH policy on sex as a biological variable” (SABV) may be found at https://orwh.od.nih.gov/sex-gender/nih-policy-sex-biological-variable .

C.W.U., M.L.L., A.T.R., B.A.H., C.E.T., W.D.M., D.K., J.R.B., J.G.R., K.L.M., L.P., M.M.W., M.O., T.O.Y., N.S.S., N.C., Q.E.D., J.D.M.-B., W.G.P., and K.Y.D.-P. conceived and designed research; J.D.M.-B. and K.Y.D.-P. prepared figures; C.W.U., M.L.L., A.T.R., B.A.H., C.E.T., W.D.M., D.K., J.R.B., J.G.R. K.L.M., L.P., M.M.W., M.O., T.O.Y., N.S.S., N.C., Q.E.D., J.D.M.-B., W.G.P., and K.Y.D.-P. drafted manuscript; C.W.U., M.L.L., A.T.R., B.A.H., C.E.T., W.D.M., D.K., J.R.B., J.G.R., K.L.M., L.P., M.M.W., M.O., T.O.Y., N.S.S., N.C., Q.E.D., J.D.M.-B., W.G.P., K.Y.D.-P. edited and revised manuscript; C.W.U., M.L.L., A.T.R., B.A.H., C.E.T., W.D.M., D.K., J.R.B., J.G.R., K.L.M., L.P., M.M.W., M.O., T.O.Y., N.S.S., N.C., Q.E.D., J.D.M.-B., W.G.P., and K.Y.D.-P. approved final version of manuscript.

The authors acknowledge funding from National Institutes of Health Grants AI168968 and HL150766 (to J.R.B.); AR079591 (to J.R.B. and T.O.Y.); ARO84226 and DK124344 (to J.G.R.); GM151274 (to M.L.L.); AG062319 , AG072094 , and AG075544 (to K.L.M.); HL147998 and R15HL165325 (to A.T.R.); HL146558 (to M.M.W.); GM151274 (to M.L.L.); HL093056 and HL146833 (to B.A.H.); NR019054 (to Q.E.D. and B.A.H.); AR084221 (to Q.E.D.); and HL135790 (to W.D.M.) and from the Biomedical Laboratory Research and Development Service of the Veterans Affairs Office of Research and Development Grants I01BX000505 (to M.L.L.) and I01BX005848 (to K.Y.D.-P.). N.C. is funded by the Military Operational Medicine Research Program at the U.S. Army Medical Research and Development Command. L.P. and the GOING-FWD Consortium is funded by the GENDER-NET Plus ERA-NET Initiative under Project Ref. No. GNP-78: Canadian Institutes of Health Research Grant GNP-161904. C.W.U. is funded by the Natural Sciences and Engineering Research Council of Canada Discovery Grant RGPIN-2018-05961 and the Fonds de Recherche du Québec Santé (Chercheur-boursier Junior 1 268920), and W.G.P. is a Senior Career Investigator for Improving the Heart and Brain Health for Women in Canada (Heart and Stroke Foundation of Canada and Health Canada) and supported with a Heart and Stroke Foundation of Canada Grant-in-Aid G-21-0031543 and Natural Sciences and Engineering Research Council of Canada Discovery Grant RGPIN-2018-04732. J.G.R. is supported by the Ludeman Family Center for Women’s Health Research at the University of Colorado School of Medicine. K.Y.D.-P. acknowledges investigator-initiated research project funding from Merck and Co., Inc.