Challenges with embedding an integrated sex and gender perspective into pain research: Recommendations and opportunities

The focus of this article, within this BBI horizons special issue, is on sex, gender, and pain. We summarise what is currently known about sex-and gender-related variations in pain, exploring intersectional biological and psychosocial mechanisms, and highlight gaps in knowledge and understanding. Five key challenges with the exploration of sex and gender in pain research are presented, relating to: conceptual imprecision, research bias, limitations with binary descriptions, integrating sex and gender, and timely adoption/implementation of good research practice. Guidance on how to overcome such challenges is provided. Despite clear evidence for sex and gender differences in pain, there are conceptual and methodological barriers to overcome. Innovation in methods and approach can help develop more effective and tailored treatment approaches for men, women, boys, girls, and gender-diverse people.


Introduction
Pain is a highly disabling biopsychosocial phenomenon.It represents a significant global health challenge with around 20 % of people living with chronic pain (Rice et al., 2016).Despite its ubiquity, pain is a variable experience, reflecting a complex interplay between sensory, cognitive-emotional, and social mechanisms (Raja et al., 2020).There are consistent differences in pain found between men and women (Fillingim et al., 2009;Keogh, 2022;Mogil, 2021;Osborne and Davis, 2022).Women 1 report a greater sensitivity to experimental and clinical pain, show greater opioid analgesia, but also more side effects (Fillingim et al., 2009;Niesters et al., 2013;Sharp et al., 2022;Todd et al., 2019).There is emerging evidence for sex/gender differences in psychological therapies for pain, although the precise pattern is somewhat unclear (Boerner et al., 2017;Flegge et al., 2022).Sex/gender differences are also reported in placebo and nocebo effects, but with mixed patterns (Vambheim and Flaten, 2017).Differences in chronic pain mortality are reported pre-clinically, but again with mixed findings in humans (Holmberg et al., 2020;Macfarlane et al., 2017;Millecamps et al., 2023).
A sex-based approach to pain has led to explorations into biological mechanisms, including structural, hormonal, and immune differences between men and women (Gulati et al., 2023;Mogil, 2021;Osborne and Davis, 2022).A focus on gender points towards psychological and socialcontextual factors (Boerner et al., 2017;Keogh, 2021Keogh, , 2022)), with gender impacting how people respond to their own pain.Gender also points to the interpersonal context, in that gender beliefs can impact how people view and respond to the pain of others (Daheim et al., 2020;Robinson et al., 2003a;Robinson et al., 2003b).While the evidence for sex differences in pain is well-established, the literature on gender and intersectionality is sparse and not well-integrated in biomedical research.
In this article we will present five challenges associated with the exploration of sex and gender in pain (see, Fig. 1) and make recommendations on how best to move forward our understanding.This review will show how innovation in methods and approach can help close knowledge gaps, to develop more effective and tailored treatment approaches for men, women, boys, girls, and gender-diverse people.
Throughout, we attempt to highlight progress to date, balance pragmatism and idealism, and offer actionable solutions.

What's in a name? Conceptual imprecision with definitions
A critical starting point when embedding sex and gender into pain research is clarifying terminology, definitions, and assumptions.This represents the first challenge, as there is conceptual imprecision in how the terms sex and gender are used (Boerner et al., 2018;Keogh, 2021); meaning the literature is fraught with confusion and inconsistencies.Fortunately, guidance exists to help (CIHR Institute of Gender and Health, 2014;Springer et al., 2012).

Conceptual definitions of sex and gender
We begin by modelling how to provide a conceptual definition of the terms sex and gender by describing our use of the terms in the present paper: we use the CIHR Institute of Gender and Health, 2014 (CIHR-IGH) definitions, which focus on possible mechanisms that contribute to variations in health.They note that sex refers to variation in biological factors between males and females, such as genes, chromosomes, physiological structures in the brain and body, including endocrine and immune systems.In humans, sex generally refers to sex assigned at birth based on observation of external genitalia.In contrast, gender is used when referring to socio-cultural influences, and includes self-identity, expectations, beliefs, and biases, as well as the wider social context.Whilst it is common to refer to men/women and boys/girls, gender it is not restricted to binary groups, and can encompass a broad range of identities.We also use the term intersectionality to describe the complex interaction of an individual's identities and social positions (e.g., race, socioeconomic status, nationality, ability) that influences their health outcomes (Crenshaw, 1991).

Assumptions behind terms
The terminology used in sex and gender research is plagued by misunderstandings and challenges in searching and indexing the literature.Many of these misunderstandings can be traced to assumptions associated with sex-and gender-related terminology.For example, sex and gender have been used interchangeably, with gender inappropriately used in the context of non-human investigations (Boerner et al., 2018).The terms men and women are used to exclusively categorise humans, whereas females and males can be used to categorise both humans and non-human animals.This overlap can lead to the terms male/female inadvertently implying human differences when discussing evidence from pre-clinical studies.The use of male/female fails to acknowledge the wider variability that exists in sex, such as intersex populations.Studies of human pain that list female sex as a risk factor imply a biological vulnerability that ignores potential psychosocial, structural, or systemic influences.
The separation of sex and gender can lead to another unintended inference -that they are independent.However, sex and gender can operate in interaction (Berenbaum et al., 2011;Polderman et al., 2018).Biological processes are shaped by psycho-socio-environmental influences (e.g., epigenetics), and psychological processes (e.g., gender self-identity) can be influenced by biological factors (e.g., sex hormones).Given the difficulties in clearly separating sex and gender, some use sex/gender to avoid ambiguity around mechanism, and where it is difficult to separate biological and environmental influences (Hyde et al., 2018;Morgenroth and Ryan, 2021).

Recommendations
• Clearly define all terms related to sex and gender.
• Carefully check the conceptual assumptions behind terms such men/ women or male/female, especially in epidemiological and observational studies.• Consider using the term sex/gender where no direct inference is being made about mechanism or refer to situations that acknowledge their interaction.

Overcoming systematic bias in research
Despite a clear imperative to explore sex/gender effects in pain, these factors are often ignored or dismissed as unimportant.In some cases, there is a deliberate sampling bias without any good reason for doing so.The second challenge is therefore to overcome biases in research practice by ensuring sex/gender are considered in the design, recruitment, measurement, and analysis of pain studies.

Historical context, gender bias in recruitment
Women and female animals have long been excluded from clinical research and trials more broadly (Wizemann and Pardue, 2001), and pain research specifically (Mogil, 2020;Mogil and Chanda, 2005;Plumb et al., 2023).When Mogil and Chanda (1991)  E. Keogh and K.E. Boerner studies published in the journal PAIN, 79 % were on just male samples; this remains an issue in 2023 (Plumb et al., 2023).Similarly, a review of immunity studies found 75 % of papers published did not specify the subject sex (Beery and Zucker, 2011).Given how widely sex/gender differences in pain are reported, a lack of awareness is not likely to fully explain this systematic bias.Concerns about variability based on the estrous cycle (the animal equivalent of the human menstrual cycle) are suggested as reasons for exclusion, even though this is not a sufficient rationale (Beery, 2018;Levy et al., 2023;Prendergast et al., 2014).
In human studies, explanations for excluding women have taken an overly protective approach, including concerns about harm during pregnancy (Yakerson, 2019).Even when a diverse sample has been recruited there is a failure to consider sex/gender differences, or sex/ gender is co-varied out (Beery and Zucker, 2011;Plumb et al., 2023).A systematic review by Plumb et al. (2023) of human studies published in PAIN (2012PAIN ( -2021) ) found that although recruited, samples increasing reflected a balance of the sexes, less than 20 % considered sex differences in the analysis.We suggest that where possible, analysis of sex/ gender differences should be routine, and at minimum included as supplementary or open-access materials for published manuscripts.
There are also possible sampling differences based on prevalence and referral bias.Women are more likely to attend pain clinics, seek out support, and consume analgesics (Robinson et al., 1998;Ruau et al., 2012).In a systematic review of psychological treatments for pain in children, twice as many girls as boys were represented within the included studies (Boerner et al., 2017).However, in experimental pain studies, masculine-identifying individuals are more likely to volunteer for experimental pain studies (Feijo et al., 2018).There are calls to make pain research more inclusive, with guidance (Boerner et al., 2023;Janevic et al., 2022;Palermo et al., 2023) -to consider language/visual choices how research is advertised, such as using inclusive language, non-gendered imagery, and statements regarding how research ensures the safety of diverse participants.This also highlights a role for inclusion of diverse voices in research design, in both the research team and patient/public involvement.
The failure to report treatment data in tables disaggregated by sex/ gender is common, making it difficult to conduct meta-analysis on moderators.It also results in a more laborious and inconsistent approach in contacting authors for information (Boerner et al., 2017).Anecdotally, scientists who conduct sex differences in pain research refer to having discovered sex differences "by accident", rather than having set out intentionally to study sex or gender.Pre-registration and analysis of possible sex/gender differences could be included as routine in pain research to help understand what represents publication bias vs. a true effect.Given the move to open access availability of data, there is also an opportunity to ensure routine provision of disaggregated data.This is of particular importance when considering gender-diverse participants where meta-analysis may be necessary to aggregate a sufficient sample size.Sex/gender differences also exist in dyadic interactions around pain (Keogh, 2021).This is particularly relevant in both the research and clinical context, given the inherent power differential between participant/patient and researcher/clinician. Researchers should clearly report sex/gender for both participants and researchers/clinicians, to explore for experimenter/therapist effects.

Measurement complexity and bias
Even when sex and gender are explored, measurement choice can introduce bias.Reporting demographic characteristics of samples rarely goes beyond "male/female/other" options, and the failure to clarify these options for participants can affect on the integrity of data and conclusions drawn.For example, offering 'male/female/other' without explanation raises the question for a trans masculine individualshould they report their sex assigned at birth (female), their gender identity (male), or select the potentially stigmatizing option of "other" to indicate their transgender experience?This is not a negligible issue given increasing awareness of the disproportionate impact of pain on marginalized populations (Craig et al., 2020;Janevic et al., 2022).One solution, known as the 2-step method, is to ask about sex and gender, with accompanying options and descriptive text to reduce ambiguity and increase precision (Beischel et al., 2023;Hyde et al., 2018;Tate et al., 2013).For example, sex can be defined as a person's assigned sex at birth, including options for intersex.Additional questions on gender identity can be added, again with options or open-ended text.This approach does raise the question of how best to manage samples that are small/unequal in size.Such challenges are solvable but need innovation in methods development (Hyde et al., 2018;Keogh, 2021).
Researchers have often dealt with the problem of small samples by combining all gender-diverse identities into an "Other" category.This is problematic, as it can stigmatize gender diversity.It can also pose a threat to scientific validity, as it ignores variability in gender identities e. g., genderfluid vs. agender vs. non-binary.Some ignore differences in gender identity or removing gender-diverse people from data analysis entirely.This practice contributes to the historical erasure and marginalization of gender-diverse people in science.Additionally, failing to consider intersectional perspectives results in an oversimplification of the meaning of gender categories (Janevic et al., 2022).For example, gendered norms, expectations, and biases may differ based on cultural context.We must understand systemic and structural biases associated with a specific gender identity (e.g., transphobia, gendered violence) within the context intersection of that identity with other positions of marginalization.There is a need to move beyond reporting these categories separately, and acknowledge the broader social context associated with intersectional identities (Palermo et al., 2023).
Applying the measurement of gender to pain research is complex.Contemporary approaches to gender often adopt qualitative methods, which can be difficult to integrate within a biomedical approach.Gender measurement tools can be limited, as popular measures of masculinity and femininity (e.g., Bem Sex Role Inventory) were developed in the 1970-1980 s, and meaning has changed over time.Others are based on predominantly US-based college samples, limiting relevance and generalisability (Keogh, 2021).A new generation of tools are being developed, reflecting different demographic and cultural contexts (Thompson and Bennett, 2015).Some tools go beyond masculine-feminine dimensions, for example, to consider gender-relations and power (Nielsen et al., 2021).Unfortunately, most gender scales have not been developed for use with pain and there are few measures of femininity relative to the plethora of masculinity measures.There is an urgent need to develop gender-based measurement tools for pain, with careful attention placed on item validity and relevance across pain conditions and age groups (Ghodrati et al., 2021).

Recommendations
• Use inclusive strategies when recruiting for pain studies, routinely including all sexes and genders unless there is a compelling reason not to.• Include 2-step measures of both sex and gender, which can include the provision of multiple options (being cautious of not "othering" ) or having open-ended questions that can be coded.• Report information on the gender context of the research environment (e.g., the gender of the researcher interacting with participants), and positionality statements to account for researcher perspectives in data analysis.
• Include details about estrous/menstrual cycle in data collections, and where possible, analyse for such effects (rather than excluding females).• Pre-register and analyse data for potential sex/gender differences where possible.Report results of sex/gender-based analyses, even if no difference is found.• Consider also examining intersection of sex and gender with other identity/social positions (e.g., race, ethnicity, socioeconomic status).
• Participate in open science practice and data sharing for meta-analysis by making data available, disaggregated by sex and gender, even if this was not a focus of the study.

Exploring beyond the sex/gender binary
Despite evidence for sex/gender differences in pain, patterns are mixed, which have been interpreted to mean small or unreliable effects (Racine et al., 2012).An alternative interpretation is that binary comparisons hides within-group or dimensional variation, highlighting the need to explore factors that have moderating and/or mediating roles.The third challenge researchers face is how to take a more nuanced approach, moving beyond binary comparisons.

Binary conceptualizations of gender
The exploration of sex and gender differences in pain mostly involves binary comparisons (e.g., males/females; men/women; boy/girl), and generally without considering biases, or the intersection with other identities or social positions.Binary, fixed, classifications of gender are overly simplistic comparisons (Beischel et al., 2023;Boerner et al., 2018;Hyde et al., 2018).For example, dichotomizing human participants based on two gender identities fails to capture the experiences of genderdiverse individuals.Issues to consider that are relevant to clinical pain in gender-diverse individuals, include increased exposure to violence and abuse, as well as therapeutic use of sex hormones (Aloisi et al., 2007;Boerner et al., 2023;Levit et al., 2021).Few studies have explored variation in pain sensitivity across a wider range of gender identities (Aloisi et al., 2007;Levit et al., 2021;Strath et al., 2020).Strath et al. (2021) found transgender women reported greater temporal summation to mechanical stimuli compared to cisgender men and women.However, they report limitations, with unbalanced groups (no transgender men) and low power.
Gender can also be used to reflect a wider range of dimensional constructs, including self-identity, expression, roles, norms/stereotypes, and expectations (Hyde et al., 2018;Keogh, 2021Keogh, , 2022)).These constructs are often explored in terms of traditional masculinity and femininity.Within a Western context, masculinity is usually associated with stoicism, independence, and strength and more likely to be reported by men, whereas femininity with emotional expressiveness, socially focused and nurturing are more likely to be endorsed by women.Whilst men generally report higher masculinity, and women higher femininity, masculinity and femininity are independent and co-existing, and not simply opposite ends of the same continuum.Masculinity has been found to be associated with pain in both men and women (Alabas et al., 2012;Keogh and Boerner, 2020).There is also a clear lack of research conducted outside of Western constructions of gender, or that considers the intersection of gender with other identity/social positions.

Binary conceptualizations of sex
In studying human sex differences, most studies dichotomize based on the sex observed at birth.As such, little can be said regarding whether this represents chromosomal sex (XX, XY), or other phenotypic sex differences in the body, brain, and behaviour, such as sex hormones (Arnold, 2020(Arnold, , 2022)).It also hides variations in sex chromosomes e.g., trisomy (XXX, XXY, XYY).Whilst sex differences in brain structures are reported, such findings are controversial, and there is not as clear a differentiation as observed in other biological factors (Hodgetts and Hausmann, 2023;Hyde et al., 2018;Joel, 2021).Sex hormones have been shown to influence pain responses, with testosterone described as having pain-relieving properties, and mixed findings regarding estrogen and progesterone.However, within-sex variation exists in hormones levels, such as fluctuations associated with the menstrual cycle, which may account for some of the variation in pain (Gulati et al., 2023;LeResche et al., 2003).Similarly, within-sex variation in pain has been related to genetics, genes and signalling, mostly in pre-clinical, but in some human studies also (Mogil, 2021).Despite pain being a good model of gene-environment interactions, few studies have explored human sex differences in epigenetic effects around pain (Mogil, 2021).What these discussions illustrate is that there is merit in exploring within-sex variation from an integrative sex-gender perspective, which may provide valuable insights into the biopsychosocial interactions that contribute to pain.

Is all binary bad?
We caution against throwing the baby out with the bathwater, and to balance idealism with pragmatism.There are situations where there is merit to proceeding with binary comparisons, especially if no other information is available.For example, when exploring epidemiological datasets, there is often no other option but to consider binary male/female comparisons.However, authors should acknowledge the limitations with the binary approach, describe what their binary factors represent (e.g., designated sex at birth), interpret findings within a broader intersectional sex/gender context, and where possible, seek ways of including a richer range of sex/gender measures.New frameworks are being developed to help measure for, and categorize, gender diversity (Beischel et al., 2023;Pelletier et al., 2015).For example, Pelletier et al. (2015) created a proxy variable to represent gender with available psychosocial measures known to be gender-related within that cultural context.Beischel et al. (2023) have developed a short set of questions that enable researchers to categorise individuals, based on different levels of gender trajectory (cisgender, transgender, allogender) and binary relation (binary, nonbinary, allobinary).While this reflected the diversity of experience, some categories were relatively small.Simply increasing sample size is not necessarily feasible, especially when complex experimental designs are used, as is often the case in neuroscience research.There is a need for funders to support research that takes a more inclusive approach, even if they are at an increased cost.Furthermore, while a non-binary approach to gender allows for a richer exploration of factors that affect pain, gender constructs are conceptually complex, and culturally and temporally specific.Conceptual models can help explain these relationships, and be used to design studies to test causal pathways (Keogh, 2021).

Recommendations
• Consider a wider range of dimensional constructs alongside binary comparisons for human sex/gender.• Identify the specific gender-related construct being considered e.g., identity, expression, norms, stereotypes, and if it is being treated as binary.
• Consider broader interpretations and understandings of sex and gender in design and measurement, but also in interpreting findings (even in the case of binary comparisons).
• Apply an intersectional lens to understand the complexity of gender identities and the limitations of homogeneous samples.• Design studies in ways that will be inclusive of people historically marginalized from health research, such as using transgenderinclusive measures of sex and gender and conducting research in partnership with people with living experience of sex and gender diversity.

Moving from observation to explanation through an integrated sex and gender approach
The fourth challenge is how to move beyond observation and towards explanation.Only by understanding why variation occurs can we develop effective ways to manage pain.While pain and sex/gender are biopsychosocial phenomena (see Fig. 2), studies tend to focus on sex/ gender binary comparisons, and/or focus on either biological or psychosocial factors (Keogh, 2021(Keogh, , 2022)).We argue here for an integrated sex and gender approach by outlining biopsychosocial mechanisms thought to be involved, and where available, provide examples of thoughtful sex and gender integration (see, Fig. 3).

Sex-gender integration in hormonal pain research
Sex hormones have an impact on early development (organizational impacts on the development of key bodily systems) and during periods such as puberty (activational impacts on behaviour and physiology).Sex hormones are also known to impact pain (Gulati et al., 2023;Nahman-Averbuch et al., 2023;Sharp et al., 2022), including effects on the endogenous opioid system, and have a role in painful conditions such as osteoarthritis.Fluctuations in sex hormones can affect pain, including across menstrual cycle phases, during pregnancy, and around the menopause (Gulati et al., 2023;Keogh, 2022;LeResche, 2013;Osborne and Davis, 2022).Pain can also be moderated through the administration, withdrawal, or suppression of sex hormones (Aloisi et al., 2007;Gulati et al., 2023;Sharp et al., 2022).Whilst patterns are complex and contradictory, estrogen and progesterone seem to enhance pain experience, whereas androgens (e.g., testosterone) have protective effects (Gulati et al., 2023;Osborne and Davis, 2022).
Few studies take an integrated endocrine and gender approach to pain, although this is possible (Manigault et al., 2021).A sex and gender approach could not only explore how changes in pain are associated with the interaction of hormonally mediated body changes, the development of gender identity, independence from family and increasing importance of social networks during puberty.Hormonal changes across the menstrual cycle and during pregnancy, during the menopause, as well as following sex hormone treatments/medication, could be explored within a gender framework -focusing on understanding the shifting gendered social demands (e.g., caregiving) and identity factors (e.g., new parenthood) that occur during these life transitions (Dennison, 2022).

Sex-gender integration in neuroscience pain research
The presence of sex hormones during early brain development leads to sexual dimorphisms in the brain (Ivan et al., 2023), which may be related to pain (Gupta et al., 2017;Osborne and Davis, 2022).For example, in their review Osbourne and Davis (2022) note studies showing sex differences in brain areas, such as the periaqueductal gray, anterior cingulate cortex, and amygdala, which are relevant to pain.They also reported evidence for binary sex differences in brain responses to induced pain.Similarly, a review by Gupta et al. (2017) concluded that there was greater variation in sensorimotor areas in women, compared to men.Other approaches have been developed, which also point to central nervous system differences.For example, conditioned modulation techniques suggest sex differences in descending control, with men showing greater inhibition to experimental pain compared to women, as well as in those with clinical pain (Lewis et al., 2012;Popescu et al., 2010).However, research investigating sex and gender differences in the brain is controversial, with contradictory findings and methodological concerns (DeCasien et al., 2022).Few studies take an integrated sex and gender approach in pain neuroscience research, which may enhance our understanding of the variation in pain that occurs within and well as between sex groups.This is therefore a potentially new and exciting area for pain discovery.

Sex-gender integration in immunological pain research
Evidence that the immune system is implicated in a sex-related way to pain mostly stems from pre-clinical studies (Gregus et al., 2021;Rosen et al., 2017;Sorge and Totsch, 2017).However, there is a known female vulnerability for chronic pain conditions that have an immune component, and sex differences exist in immune responses to other health conditions (Klein and Flanagan, 2016).There are also sex differencs in experiences highly relevant to pain, such as adverse childhood events and depression (Craner et al., 2022;Kropp and Hodes, 2023).Again, few studies take an integrated sex and gender approach, although there are examples that might translate to human pain (Diamond et al., 2021).For example, Diamond et al (2021) found inflammatory vulnerabilities associated with gender-diverse populations due to stigma/stress.A related challenge how findings from pre-clinical studies might translate, if at all, to and from clinical studies.While it may be possible to bring findings from pre-clinical and experimental studies (e.g., sex hormone effects) forward into human clinical models, it is less clear how effectively backward translation (i.e., integrating gender-related concepts to pre-clinical models) could be accomplished.

Sex-gender integration in psychosocial pain research
Gender and gender-related constructs are psychosocial in nature, and their importance in understanding the variation in pain has already been articulated.Here we describe psychosocial factors that help explain the variations in pain, and how an integrated sex-gender approach may advance the field.
Sex/gender differences exists in cognitive-behavioural factors that are important for pain (Keogh, 2018(Keogh, , 2021(Keogh, , 2022)).Pain-related anxiety is found to be more strongly related to pain in men, whereas depression is stronger for women (Edwards et al., 2003;Keogh et al., 2006).Women report higher pain catastrophizing, a greater negative appraisal of pain (Keogh et al., 2004;Unruh et al., 1999).Pain catastrophizing also mediates sex/gender differences in pain (Sullivan et al., 2000).Behavioural differences are found in health seeking behaviours, with women consulting more about health, including pain, and consuming more analgesics (Bingefors and Isacson, 2004;Wang et al., 2013).Men report using distraction and avoidance approaches, yet benefit from focusing on pain, whereas women reporting using social support more (Keogh, 2018(Keogh, , 2022)).Less is known about pain resilience factors, or how these factors present in gender-diverse populations.A better understanding of the sex-related neurobiology of stress responses and regulation, and how this is potentially reinforced by gendered expectations or norms, may advance the research on pain coping (Goldfarb et al., 2019).
The use of social support highlights the social and interpersonal context of pain (Keogh, 2018(Keogh, , 2021(Keogh, , 2022;;Samulowitz et al., 2018).Pain reports vary according to whether interpersonal interactions involve same-and opposite-sex dyads, as well as in different relational contexts e.g., friend, stranger, family members.Sex/gender differences in verbal and nonverbal pain communication have been reported (Jaworska and Ryan, 2018;Keogh, 2021Keogh, , 2022)).For example, Jaworska and Ryan (2018) found that men used fewer pain words to describe pain, but that the words they used were more emotive.There may also be sex/ gender differences in how observers view pain expressions (face, body), as well as how healthcare professionals view and judge pain (Keogh, 2021;Samulowitz et al., 2018).For example, women and girls' pain seems to be viewed more negatively and underestimated, resulting in possible treatment differences.A sex-gender integrated approach to studying these biases may help illuminate discrepancies between gendered expectations and biological outcomes, as well as to develop materials to train health professionals in understand their own biases and how it affects practice.
Like the study of biological mechanisms, exploration of these psychosocial factors could benefit from moving beyond the binary, as well as adopting an integrated sex and gender approach.There are interesting translational developments, drawing on work in humans on interpersonal interactions, and the role of social and/or environmental context.For example, sex-specific dyads and 'social' interactions have been explored in rodents, and non-verbal pain behaviours (facial Fig. 3.A visual depiction of some future directions made possible from an integrated sex and gender approach.expression, vocalisations), are observed in animals (Mogil et al., 2020).Learning, avoidance/approach behaviours, and proximity have potential for translational work, and in a sex-specific way.Similarly, there may be translations from rodent empathy and social contagion that informs human pain research (Martin et al., 2015).
Other factors, such as age, race/ethnicity, social class, and poverty, can intersect with sex and gender to influence pain (Craig et al., 2020).For example, ethnicity interacts with sex/gender, although only a few studies consider these factors together (Weisse et al., 2005).There is more evidence that sex/gender differences in pain vary across the life course, being less pronounced in early years and emerging around puberty and continuing into adulthood (Boerner et al., 2014).It is unclear whether these differences are maintained in older age groups (Lautenbacher et al., 2017).Few studies take a lifespan perspective mapping changes in sex/gender and pain over time, despite known developmental shifts in biological (e.g., hormones), psychological (e.g., identity), and social influences (e.g., dependency) (Boerner et al., submitted).

Recommendations
• Develop methods to explore individuals in pain within a broader biopsychosocial context, including biological and psychosocial interactions.• Consider using qualitative and mixed-methods to understand the individual variations in experiences of gender-related concepts in interaction with sex-related biological effects in pain.• Consider taking an intersectional approach in data analysis (at best) or interpretation of findings (at minimum) • Explore ways to enable better translation (forward/backward) between pre-clinical and clinical investigations into sex/gender influences on pain.

Moving from observation to advocacy and meaningful change
The fifth challenge is how to encourage the implementation of good research practice around sex/gender and pain, and how to use research as advocacy to initiate meaningful change.This requires researchers to place their work in the broader context of inequalities and inequities in health, and use the knowledge gained to initiate change for the better.This final section will consider ways to promote inclusive research to improve how we understand and manage pain.

Placing sex/gender within a wider context of inequality and inequity
Exploring sex/gender is already complex when considered at an individual level, and taking a broader perspective can make exploration even more difficult.However, while there is a need to "zoom in" on the interactions of individual identities, it is also critical to take a "zoomed out" view of the systems and structures that maintain and perpetuate a range of inequities in pain care (see, Fig. 2).Key to this is multilevel and/ or intersectional approaches to study design and data analysis e.g., incorporating the multiple identities/positions an individual holds and associated experiences of marginalization.Guidance is being developed on how best to actualise this in health research (Abrams et al., 2020;Janevic et al., 2022;Newman and Thorn, 2022).Researchers at all stages of the translational spectrum can consider how their findings may be interpreted, implemented, and disseminated within systems (e.g., academia, health care) that uphold and perpetuate biases related to pain and marginalized groups.This approach opens opportunities for researchers to see how their work can operate as advocacy for change.
Conducting sex/gender science itself can also be an act of advocacy, moving forward the needle on understanding inequities in pain care.The politicization of issues related to gender parity in academic or healthcare settings, gendered biases in funding and opportunities for scientists, restricted access to reproductive healthcare or gender-affirming care, have consequences for understanding sex/gender influences on pain.These issues exert their influence in what research is funded, which researchers experience success, which researchers bear the burden for moving equity-related research forward, and which findings are translated to clinical practice and policy.This necessitates moving away from silos, partnering with public, government, and change-makers at all levels, and emphasizing meaningful knowledge dissemination.Collaborations with other disciplines who have effectively addressed issues related to social justice can inform the innovation of pain research methods, and guide how to address systems-level issues.This may feel a long way from the lab bench but is critical to translate pain research findings into actionable, equitable changes in care for people living with pain.

Research guidance and policies
There are existing guidance and policy recommendations to change research practice and distribute the responsibility.The International Association for the Study of Pain's Special Interest Group on Sex, and Gender Differences in Pain published a consensus statement and guidance (Greenspan et al., 2007).There is a need for this to be updated, to reflect current understanding.It is notable that this Special Interest Group has recently updated its name to include Racial Differences, highlighting the increasing importance of taking an intersectional approach in the study of sex/gender.More recently, general guidelines have been developed, which focus on sex as a biological variable, and gender as a sociocultural variable (Heidari et al., 2016;Nielsen et al., 2021).Leadership is key to promoting implementation.For example, the US National Institutes of Health (NIH), the Canadian Institutes of Health Research (CIHR), and the European Commission (EC) introduced sex/ gender policies into funding guidelines.However, guidance is not routinely taken up, and there is still a need to ensure data is disaggregated by sex and gender.This calls for the development of better measurement tools to study gender-related effects (Gogovor et al., 2021;White et al., 2021).In pain, there is now an agreed approach across eight of the key pain journals (e.g., PAIN, The Journal of Pain, Pain Medicine) to improve the practice of ensuring diverse representation across scientific pain work (Palermo et al., 2023).

Recommendations
• Adhere to good practice guidelines around the study and reporting of sex, gender, and pain in research.• Design intersectional studies that take a systems-level perspectives to the exploration of sex, gender, and pain.• Explore novel approaches developed in other areas of research (e.g., epidemiology, sociology) that could help explore intersectionality in pain.
• Consider partnership with policy-makers, government, and organizations that can help translate research into advocacy for change.

Conclusions
Despite clear evidence for sex/gender playing an important role in pain, these factors can often be ignored, or their importance dismissed.This variation also highlights the broader a healthcare challenge, which is to better understand and pain inequities.Without addressing this problem, inequities in pain care will persist.There is still much to learn about why sex/gender variation occurs, and how to harness this knowledge in pain management.
We have highlighted here five challenges that need to be overcome.These challenges span conceptual and methodological complexities, which are partly due to difficulties in embedding gender into sex/biological research.We make recommendations on how best to overcome these challenges and facilitate the integrated, intersectional exploration of sex and gender in pain research.We have tried to balance the pragmatics with doing research, alongside the ideal ways in which we would like to see research conducted.In doing so, we hope this will lead to more effective, tailored, treatment approaches that support all people living with pain to lead and live full and rewarding lives.

Fig. 1 .
Fig. 1.Five challenges to embedding sex and gender into pain research.

Fig.
Fig. Illustration of key biopsychosocial factors involved in sex/gender differences in pain, with examples of some, but not all, intersecting identities.