Elsevier

Journal of Clinical Epidemiology

Volume 129, January 2021, Pages 191-197
Journal of Clinical Epidemiology

Commentary
Methods for prospectively incorporating gender into health sciences research

https://doi.org/10.1016/j.jclinepi.2020.08.018Get rights and content

Highlights

  • Sex, a biological construct, and gender, a social construct are two distinct variables that may independently influence human health.

  • Despite calls for inclusion of sex and gender into health sciences research, gender is often ignored or conflated with sex.

  • In this commentary, we provide clarification of the distinction between these two variables and concrete examples of gender-related variables that can be collected under the four domains of gender identity, gender roles, gender relations, and institutionalized gender.

  • We also provide methods for incorporating these variables into statistical analysis.

  • We hope these guidelines will help researchers in their efforts to incorporate gender into their studies, thereby meeting requirements of funding agencies and ultimately improving health equity and precision medicine.

Abstract

Numerous studies have demonstrated that sex (a biological variable) and gender (a psychosocial construct) impact health and have discussed the mechanisms that may explain these relationships. Funding agencies have called for all health researchers to incorporate sex and gender into their studies; however, the way forward has been unclear to many, particularly due to the varied definition of gender. We argue that just as there is no standardized definition of gender, there can be no standardized measurement thereof. However, numerous measurable gender-related variables may influence individual or population-level health through various pathways. The initial question should guide the selection of specific gender-related variables based on their relevance to the study, to prospectively incorporate gender into research. We outline various methods to provide clarification on how to incorporate gender into the design of prospective clinical and epidemiological studies as well as methods for statistical analysis.

Introduction

To fully understand and improve human health, it is important to incorporate sex and gender in health sciences research. Sex is a biological variable that distinguishes individuals as male or female (or intersex) based on their genetics, anatomy, and hormones [1]. Gender, on the other hand, is a social construct. It encompasses the identities, expressions, roles, norms, behaviors, and perceptions of men, women, boys, girls, and gender-diverse people [1]. It may also include the institutionalization of these norms, and how individuals are treated by society based on their identified or perceived gender, as well as relations between individuals based on identified or perceived gender [2]. A growing body of evidence demonstrates that both sex and gender may independently influence both disease risks and outcomes [[3], [4], [5], [6], [7]] and that further investigation of the role of both is necessary [5,8]. Indeed several major funding agencies now require the consideration of these variables into research proposals [1,9,10]. Nevertheless identifying methods for the incorporation of sex and gender remains difficult for many researchers, who often conflate these terms [11,12]. Despite increasing inclusion of female participants in clinical research and recognition of sex as an important variable, gender factors often remain neglected [13,14].

Gender, distinct from sex, has proven elusive from both a collection/measurement perspective as well as the relatively low frequency of its incorporation into health sciences research, despite long-standing recognition of its importance in the social sciences. The definition of gender changes with time and varies across cultures, disciplines, and among public health organizations and often includes imprecise, vague language [1], making the identification of appropriate scientific methods for measurement and analysis unclear. Because of these difficulties, few clinical researchers have attempted to quantify “gender” as a psychosocial construct and measure its impacts on health [4].

The objective of this study is to provide clarification on how to incorporate gender into the design of prospective clinical and epidemiological studies as well as methods for statistical analysis.

Section snippets

The integration of sex and gender into the research question

There are several ways in which sex and gender may influence the outcomes and/or the relationships of interest in human health studies. For example, although sex may influence an individual's biological susceptibility to an infectious disease or likelihood of developing a chronic condition, gender may influence an individual's likelihood of exposure to the disease or developing the condition through differences in social roles, responsibilities, occupation, and/or risk-taking behaviors.

Identifying gender-related variables

The Canadian Women's Health Research Network identifies four domains that encompass gender: gender identity, gender roles, gender relations, and institutionalized gender [2]. Gender identity refers to the way an individual self-identifies, which may impact their behaviors and expression of gender, as well as how others treat them. Gender roles refer to the norms and behaviors typically associated with gender. Gender relations refer to the way in which people may interact with each other based

Gender as an explanatory variable: individual vs. composite measure

If collecting many variables, researchers may investigate them in univariate and multivariate analyses or they may wish to reduce or consolidate them into a composite score of all gender domains or a score that reflects the individual gender domains they have investigated. This step would reduce the number of variables necessary to include in statistical models by creating one or a few metrics for gender rather than many intercorrelated gender-related variables [21,32,33]. Specifically, a

Discussion

Although numerous researchers and funding agencies have required the consideration of sex and gender in research designs, many researchers are struggling with how to operationalize the collection of these factors, particularly given that there is no standardized definition or measurement for gender. This commentary provides concrete examples of variables that can be collected to incorporate gender into prospective study design as well as strategies for analysis, while providing flexibility to

References (37)

  • C. Tannenbaum et al.

    Sex and gender analysis improves science and engineering

    Nature

    (2019)
  • How Sex/Gender Influence Health & Disease (A-Z) National Institutes of Health Office of Research on Women's Health

  • Promoting Gender Equality in Research and Innovation Horizon 2020

  • A.M. Weber et al.

    Gender norms and health: insights from global survey data

    Lancet

    (2019)
  • M.H.J. Bekker

    Investigating gender within health research is more than sex disaggregation of data: a Multi-Facet Gender and Health Model

    Psychol Health Med

    (2003)
  • G. Labots et al.

    Gender differences in clinical registration trials: is there a real problem?

    Br J Clin Pharmacol

    (2018)
  • N. Nowatzki et al.

    Sex is not enough: the need for gender-based analysis in health research

    Health Care Women Int

    (2011)
  • C.N. Phellas et al.

    Structured methods: interviews, questionnaires and observation

    Research Soc Cult

    (2011)
  • Cited by (0)

    Declaration of interest: none.

    Funding: The GOING-FWD Consortium is funded by the GENDER-NET Plus ERA-NET Initiative (project ref. number: GNP-78): The Canadian Institutes of Health Research (GNP-161904), La Caixa Foundation (LCF/PR/DE18/52010001), The Swedish Research Council (2018-00932), and The Austrian Science Fund (FWF, I 4209). V.R. is funded by the Scientific Independence of Young Researcher Program of the Italian Ministry of University, Education and Research (RBSI14HNVT).

    This manuscript was conceptualized by C.P.T., V.R., L.P., and C.M.N. The initial draft was written by C.P.T., with all other authors contributing to editing.

    Conflict of interests: We declare no competing interests.

    View full text