6

Chronic Conditions That Predominantly Impact or Affect Women Differently

This chapter describes the research progress on and advances in understanding chronic conditions present in both women and men that predominantly impact or affect women differently. As mentioned in Chapter 1, the committee selected certain chronic conditions to highlight to provide an overview of the evidence base and research gaps around them. This chapter covers 16 chronic conditions involving mental health, human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS), cardiometabolic disorders, pain disorder, autoimmune disorders, neurocognitive disorders, and musculoskeletal conditions. Chapter 4 provided data on the impact of select chronic conditions. This chapter highlights the specific role of biological factors, with specific emphasis on estrogens, sex chromosome effects, and influences of the maternal in utero environment, in the development of the disease or condition; emerging differences in how these disorders and conditions present themselves; diagnosis, treatment, and management in women; disparities by racial and ethnic groups and sexual orientation in terms of effects on susceptibility and outcomes; and research gaps.

DEPRESSION

Depression is more common in women, affecting them across the life course and causing significant morbidity and mortality (Bromet et al., 2011; Kessler et al., 1993). This report primarily discusses major depressive disorder (MDD), with some mention of postpartum depression (PPD) and hormonal-associated depression, which includes premenstrual dysphoric

disorder (PMDD)/premenstrual syndrome (PMS) and perimenopausal depression. Differences in the prevalence and effects of depression are likely related to biological, environmental, and societal factors (Yuan et al., 2023). It is also associated with developing other chronic conditions in women, including Alzheimer’s disease (AD); adverse cardiovascular events, including myocardial infarction, stroke, and death; and post-stroke depression, including greater severity, poorer neurological outcomes, increased risk of stroke recurrence, and mortality (O’Neal, 2023; Sibolt et al., 2013; Volz et al., 2021; Wang et al., 2017; Wu et al., 2019). The co-occurrence of three major chronic diseases—MDD, CVD, and AD—has a significant effect on women (Goldstein et al., 2021).

Biological Factors

Research has yet to fully elucidate the biological mechanisms underlying the susceptibility of women to depression. Several lines of evidence indicate roles for the gamma-aminobutyric acid (GABA), serotonin, and dopamine neurotransmitter systems and the stress-responsive hypothalamus-pituitary-adrenal axis in depression. Studies also show that female gonadal hormones, such as estrogens, and sex chromosome effects modulate the function of these neural systems, which may partly explain why women are more likely to experience depression.

Sex Chromosome Effects

In contrast to the plethora of evidence supporting a role for gonadal hormones in modulating mechanisms that lead to depression, limited evidence shows that sex chromosomal genes play a role in women. One study demonstrated that genetic polymorphisms¹ on the X chromosome influence the expression levels of three proteins expressed in GABAergic neurons—somatostatin, a hormone that affects neurotransmission, and two enzymes involved in the synthesis of the neurotransmitter glutamate (Seney et al., 2013a). This study also found that women with depression have significantly lower levels of somatostatin in the brain compared to men with depression. With studies showing less GABA-mediated inhibition of neural activity in depression (Luscher et al., 2011; Sanacora et al., 1999; Sequeira et al., 2009), sex chromosome genes can influence GABAergic neural activity by regulating protein expression in GABAergic neurons. Experiments using the Four Core Genotypes (FCG) mouse model, which decouples sex chromosome and gonadal hormone effects (see Chapter 2),

___________________

¹ Polymorphisms are two or more variations of the same gene or protein that occur among different individuals or populations.

also support this premise, demonstrating that sex chromosome effects regulate expression of genes related to GABA and the serotonergic and dopaminergic neurotransmitter systems that also modulate depression (Seney et al., 2013b).

The Effects of Estrogens

The menstrual cycle plays an etiological role in depression (Albert, 2015). Fluctuating sex hormone levels are linked to changes in brain morphology, function, and neurochemistry (Kundakovic and Rocks, 2022), with dynamic alterations in the epigenome triggering changes in gene expression that mediate these effects (Kundakovic and Rocks, 2022). Research in mice suggests that hormones and genes interact to change the brain structure in a way that results in an increased risk of depression in women (Kundakovic and Rocks, 2022). Evidence also links depression to a loss of synchronization between estrogens and cortisol phases of the reproductive cycle (Butler, 2018). Some studies demonstrate a genetic association between different estrogen receptor (ER) polymorphisms and depression (Li et al., 2022a).

Animal studies show that the ovarian hormone 17β-estradiol regulates neural activity in key subcortical regions in the stress response circuitry linked to MDD (Jacobs et al., 2015). Furthermore, inflammation has been linked to MDD (Haapakoski et al., 2015), and estrogen deficiency is linked to increases in several inflammatory markers (Klein and Flanagan, 2016). These increases may serve as a viable biomarker for identifying the risk of MDD in women. Perimenopausal depression includes complex relationships between hormonal fluctuations and neurotransmitter activity, brain-derived neurotrophic factors, chronic inflammation, and oxidative stress that warrant further study (Liang et al., 2024).

Other Biological Factors and Mechanisms

Cortisol is an anti-inflammatory hormone whose dysregulation can lead to inflammation, which has been implicated in depression and depressive disorders associated with PMS and PMDD (Bannister, 2019). It is released during stress and modulates activity in distinct brain regions, including the amygdala, prefrontal cortex, and hippocampus. Genetic evidence demonstrates sex differences in cortisol receptor expression and suggests an association between MDD in women and a specific polymorphism in a cortisol receptor (Teo et al., 2023).

Another possible mechanism may involve intestinal dysfunction, such as in inflammatory bowel disease. Experiments have demonstrated that chronic stress in female mice can disrupt the intestinal barrier, which can trigger the release of inflammatory molecules that affect neural circuits in the brain

involved in depression (Doney et al., 2024). A recent scoping review noted sex-specific differences in the composition of the gut microbiome and associations with depression severity in patients (Niemela et al., 2024).

The serotonergic system, the primary target of many antidepressant medications, shows differential expression in women and men of neurotransmitter receptors and some transcription factors. This finding may be related to evidence suggesting that selective serotonin reuptake inhibitors (SSRIs) are more effective in treating depression in women (Seney et al., 2022).

A recent meta-analysis confirmed that PPD involves several genes and is at least partially heritable. This analysis also identified a plausible biological pathway involving GABAergic neurons (Guintivano et al., 2023).

The Maternal in Utero Environment

The maternal in utero environment is an important determinant for susceptibility to depression, especially in female offspring. Clinical studies have linked maternal anxiety (Van den Bergh et al., 2008), higher levels of inflammatory markers (Lipner et al., 2024), and infection during pregnancy (Pallier et al., 2022) to depressive symptoms in adolescent female offspring (Lipner et al., 2024; Van den Bergh et al., 2008). Preclinical rodent models have also demonstrated that exposure to maternal stress-induced inflammation led to an enhanced sensitivity to stresses affecting the hypothalamus-pituitary-adrenal axis in female offspring (García-Cáceres et al., 2010).

Low birthweight and preterm birth are also related to the maternal in utero environment and considered to reflect fetal well-being during pregnancy (Kim et al., 2015). Research examining these features and their relationship to depression later in life have yielded mixed results. One study found no association, but another study comparing outcomes by sex found that low-birthweight female babies had a greater risk of depression than male babies by the age of 21 years (Alati et al., 2007). A recent analysis of data from the Women’s Health Initiative (WHI) supports the idea that low birthweight or preterm birth is associated with depression in women later in life (Rahalkar et al., 2023).

Specific Risk Factors in Women

Reproductive Milestones

Reproductive milestones serve as windows of vulnerability for depression. Clinical and population-based studies have shown that increased prevalence of depression correlates with hormonal changes in women, particularly during puberty, prior to menstruation, after pregnancy,

at perimenopause, and in menopause, suggesting that changes in hormonal levels may explain the differences in increased prevalence of depression in women (Figure 6-1) (Albert, 2015; Kundakovic and Rocks, 2022). Although girls entering puberty do not necessarily have lower levels of estrogen, fluctuations in those levels may play a role in developing depression at this age (Kundakovic and Rocks, 2022). PMDD is associated with cyclical mood, behavioral, and physical indications that correspond to menstruation, specifically the luteal phase in women. PMDD more commonly affects reproductive aged women but can begin to affect women in teen years (Mishra et al., 2024b). Studies suggest that in PMDD, hormonal patterns are normal but that women may have heightened sensitivity to cyclical variations in estrogen and progesterone levels (Hantsoo and Epperson, 2015; Mishra et al., 2024b).

Recently, a population-based cohort study analyzing data from 264,557 women suggested that the use of hormonal contraceptives, particularly in the first 2 years, is associated with a higher risk of depression, and that use during adolescence may elevate depression risk at older ages (Johansson et al., 2023). This is in contrast with a more recent study

in a smaller cohort of 6239 women, which indicated that current hormonal contraceptive use was associated with lower odds of depression (Gawronska et al., 2024). Another study indicated that prior depression associated with hormonal contraceptive use may be linked to a higher likelihood of developing PPD (Larsen et al., 2023). Such findings suggest that individual differences in hormonal sensitivity may influence susceptibility to depression across the life course in some women (Larsen et al., 2023).

The incidence of depression during the menopausal transition increases, even among women who have a history of depression (Maki et al., 2019). Longitudinal studies, such as the Study of Women’s Health Across the Nation (SWAN), Penn Ovarian Aging Study, and Seattle Midlife Women’s Study, have shown that perimenopause has an increased likelihood of depressive symptoms and disorders, with likelihood increasing in the late stage of perimenopause as levels of estrogen continue to drop (Maki et al., 2019). Studies have shown that removing the ovaries increases the likelihood of depression (Bräuner et al., 2022; Hassan et al., 2024; Hickey et al., 2021).

Vulnerabilities

Researchers have used the affective, biological, cognitive model of depression, based on a vulnerability-stress model from childhood to adolescence, to examine gender differences in the developing depression (Hyde and Mezulis, 2020). Evidence shows that affective factors, such as temperament differences between boys and girls, may explain gender differences in depression incidence. However, this correlation is based mainly on cross-sectional rather than longitudinal studies (Hyde and Mezulis, 2020). Cognitive vulnerabilities for depression specific in women include negative cognitive style and objectified body consciousness, an internalization of cultural measures about ideal body and appearance. This model also explains how trauma, sexual harassment and abuse, and victimization, which are more prevalent in women, affect the development of mental disorders, including MDD (Hyde and Mezulis, 2020; Klein and Martin, 2021; Krahé and Berger, 2017).

Gender Roles

The association between gender inequalities and depression has been well studied. Negative early-life experiences resulting from gender inequalities have been associated with the increased vulnerability of women to depression (Remes et al., 2021). Studies have shown that women are more likely to experience complex and varied adverse childhood experiences

(ACEs)² (Finkelhor, 1987; Silverman et al., 1996; Weiss et al., 1999). These exposures are associated with an increased risk of depression (Bochicchio et al., 2024; Weiss et al., 1999), with inflammation implicated as a possible underlying biological mechanism (Iob et al., 2020).

Certain events in the reproductive life cycle, including menarche, pregnancy, the postpartum period, and menopause, come with new responsibilities, role transitions, body image considerations, societal pressures, and hormonal changes and are also associated with higher rates of depression (Albert, 2015; Kundakovic and Rocks, 2022). Developmental changes in a woman’s life-span, starting from childhood, can lead to shifts in gender roles, societal pressures, vulnerability, discrimination, and variable social support, all of which can influence risk of depression in women (Sachs-Ericsson and Ciarlo, 2000; Vafaei et al., 2016). Special considerations that increase risk include ideas around sexuality, conception, and pregnancy, including carrying out a pregnancy, pregnancy loss, and infertility (Salari et al., 2024).

Changes in body appearance and function starting during adolescence can be associated with societal pressures and experiences of sexual and physical vulnerability. Life events related to reproduction in women, such as miscarriage, are associated with significant trauma and grief (Quenby et al., 2021). Evidence suggests that infertility is a susceptibility factor for depression (Salari et al., 2024), although the association is likely complex (Bagade et al., 2023). For example, female hormonal fluctuations are associated with sleep disorders, which are associated with infertility, which is associated with a higher rate of mood disorders (Baker and Lee, 2022).

Reproductive function itself is not a solely biological prospect; feelings and expectations surround it, and in societies in which motherhood is perceived as an essential component of a woman’s identity, women who do not achieve motherhood may experience significant stress (Bagade et al., 2023). Efforts to conceive are unsuccessful for one in four couples worldwide, and infertility is associated with emotional, social, relational, and financial burdens (Warne et al., 2023). One study demonstrated an association between assisted reproductive technology and current and past mental disorder diagnoses, with onset during assisted reproductive technology in up to one-third of patients undergoing it (Warne et al., 2023).

In addition to hormonal fluctuations, women may experience pressures and expectations about body appearance, which can be conflicting and

___________________

² ACEs are defined as “emotional, sexual, or physical abuse, emotional or physical neglect, and five types of household dysfunction including household substance abuse, a violent home environment, family mental disorders, parental incarceration, and parental separation or divorce” and contribute to the development of risk factors related to morbidity and mortality during adulthood (Christensen et al., 2021; Felitti et al., 1998; Nelson et al., 2017).

contradictory at baseline; pregnancy is associated with noticeable changes, including increased abdominal girth, overall weight gain, and other appearance changes that may be at odds with societal and internalized notions of beauty (Crossland et al., 2023). During pregnancy, women in general are at increased risk for intimate partner violence (IPV). One meta-analysis demonstrated an approximately 9 percent rate of physical and sexual IPV for pregnant women in North America and a nearly 30 percent rate of psychological IPV (Román-Gálvez et al., 2021).

Condition Presentation and Diagnosis

Physical symptoms associated with depression are more common in women and may include fatigue, appetite, weight disturbances, and musculoskeletal pain (Lara-Cinisomo et al., 2020). A major depressive episode during the postpartum period and whether to delineate this from a separate diagnosis of MDD is a point of debate, and the literature varies based on how the investigators defined the postpartum period, epidemiological aspects, and etiology and treatment aspects (Batt et al., 2020). The Edinburgh Postpartum Depression Scale is the most used screening tool for PPD (Sit and Wisner, 2009). Ideally, these tools should incorporate symptoms unique to postpartum experiences that are not captured in MDD. Evidence is mixed that depression with an onset of 8 weeks postpartum might be different than MDD (Batt et al., 2020).

Although sometimes used interchangeably, perinatal depression is different from PPD. Perinatal depression refers to major or minor depression that starts during pregnancy or up to 12 months postpartum (Batt et al., 2020). The U.S. Preventive Services Task Force (USPSTF) recommends screening and identifying women at increased risk of perinatal depression and referring them to interventions involving counseling (Curry et al., 2019).

As with other psychiatric conditions, research has failed to develop consistent and reliable biomarkers for depression. Certain neurotransmitters and inflammatory markers show promise based on physiological findings in persons with depression, but little is known about how sex affects potential biomarkers (Carvalho Silva et al., 2023; Strawbridge et al., 2017).

Treatment and Management

Early and effective treatment of MDD is crucial to achieve positive outcomes, especially in women, but MDD treatments are underutilized (Goodwin et al., 2022). Studies have shown differential responses to certain types of antidepressants and note that many patients do not respond adequately to traditional antidepressants (Dwyer et al., 2020). Several factors, including body composition and hormonal influences, attenuate the

differential response and pharmacokinetics in women. Studies have shown that women respond better to SSRIs, and men to tricyclic antidepressants (Sramek et al., 2016). However, that increased response to SSRIs appears to be age dependent and is not observed in menopause (LeGates et al., 2019). Other studies have shown more rigorous therapeutic benefits of certain medications after hormone therapy (Parry, 2010). Combination therapy has also demonstrated increased benefit in women over men (Parry, 2010). SSRIs in addition to estrogens are usually more beneficial in improving mood than SSRIs or estrogen treatment alone for major depression, whereas the selective norepinephrine and serotonin reuptake inhibitors do not require estrogens to exert their antidepressant effects in menopausal depression (Parry, 2010).

In 2019, Food and Drug Administration (FDA) approved brexanolone, a naturally occurring molecule that modulates GABAergic activity and is associated with significant improvement in depressive symptoms in adult women with PPD (Epperson et al., 2023). It is administered by intravenous infusion in a health care facility over a 60-hour period. In 2023, FDA approved zuranolone as the first oral drug to treat PPD (FDA, 2023).

Other research findings support the benefits of lifestyle-based approaches and alternative therapies in preventing and managing depressive symptoms. Meta-analyses of randomized controlled trials (RCTs) support how exercise improves depressive symptoms and quality of life (Blumenthal and Rozanski, 2023). Numerous studies also indicate that high-quality sleep (Cui et al., 2024), diets low in proinflammatory foods (Choi et al., 2024; Khosravi et al., 2020; Oddy et al., 2018; Tolkien et al., 2019), and the microbiome composition (Burokas et al., 2017; Radjabzadeh et al., 2022) may prevent or reduce risk and symptoms of depression. Interventions such as strong social support, cognitive behavioral therapy, acupuncture, and mindfulness-based therapies have been shown to improve symptoms (Cui et al., 2024; Li et al., 2022b; Witkiewitz and Bowen, 2010; Yang et al., 2023).

Disparities

Racial and Ethnic Disparities

Many studies have shown that hormonal, biological, and psychosocial factors related to depression vulnerability are exacerbated in adolescent girls identifying as lesbian, gay, bisexual, transgender, queer, intersex, asexual (LGBTQIA+) and in certain racial and ethnic groups (Patil et al., 2018). The interplay of gender and racial or ethnic socialization contributes to overall identity formation, which in turn informs coping strategies that are also related to the brain structure and function underlying or protecting against depression (Patil et al., 2018).

Although studies have reported that Black and African American and Hispanic/Latina women report lower levels of depression, symptom presentation differs. A recent narrative synthesis explained that Hispanic/Latina women reported more physical symptoms than White women (Phimphasone-Brady et al., 2023). Furthermore, a systematic review showed that African American women were more likely to express physical symptoms than White women (Lara-Cinisomo et al., 2020). Researchers have not examined symptom presentation in PMS or PMDD in girls and women by race and ethnicity (Phimphasone-Brady et al., 2023).

Lack of culturally competent care for depression also contributes to mental health disparities. For example, women identifying as Black, African American, American Indian, Alaska Native, and Hispanic/Latina placed a strong emphasis on having a health care provider of similar racial and cultural background compared to men of the same racial and ethnic backgrounds (Eken et al., 2021). Several studies support that racial and ethnic concordance between the health care provider and patient may lead to better communication, perceptions of care, and health outcomes, although there are issues of heterogeneity and a lack of quality in the extant evidence (Moore et al., 2023; Shen et al., 2018). Women from these racial and ethnic groups may also experience greater stigma in health care settings, which may be linked to lower access of mental health care and services (Misra et al., 2021). More research specifically focused on women is needed in this area.

Research Gaps

Despite immense strides in acknowledging sex differences in depression and gaining insight as to how sex gonadal hormones, particularly estrogens, heighten the risk in women, significant knowledge gaps remain regarding the mechanisms that underlie sex differences. As most experimental animals do not have a menstrual cycle, alternative models are needed to better understand the molecular basis of the intrinsic, hormonally driven vulnerability to depression (Kundakovic and Rocks, 2022). More research is also needed to study sex hormone and gene interactions (Kundakovic and Rocks, 2022) and understand how sex chromosome effects play a part in the pathophysiology of depression in women. Findings from such research can be leveraged to develop diagnostics and treatments that are most efficacious in women.

As with other chronic conditions, inflammation appears to play a mediating role between depression inducing factors and MDD. More research is needed to understand inflammatory triggers for depression in women (e.g., trauma, ACEs, stress) to guide the development of future therapies.

In addition, more studies on the bidirectional influences of depression and other chronic conditions (e.g. AD and CVD), would yield insight into how negative affectivity and rumination could lead to physiological alterations that increase the risk of developing other chronic conditions. The understanding of how depression, including PMDD and PPD, manifests throughout reproductive milestones and across the life course, such as during the menopausal transition, need to be further explored.

A better understanding of how environmental influences or extrinsic factors influence depression in women is needed. Research on external factors affecting mood and depressive symptoms have multiplied in recent decades, and it is important to understand the factors underlying why some individuals are more susceptible. Further research on biopsychosocial resilience factors, such as nutrition/nutrients, microbiome, lifestyle (e.g., exercise), and family/community/medical caregiving support, and how these counter risk factors for depression in women, is important.

More research is needed to develop appropriate screening tools (e.g., questionnaires and biomarkers) that discern the diagnostic features of perinatal depression, PPD, and MDD to better identify women at risk during the reproductive stages and initiate timely interventions. Adapting routine screening tools for use in racially and ethnically diverse women remains paramount, as research has demonstrated racial differences in depression expression and presentation (Salihu et al., 2022). Similarly, research needs to incorporate racial, ethnic, and cultural considerations into clinical trials to treat depression in women at all stages of the life cycle, since racial and sex differences exist in screening and treatment practices (Hyeouk et al., 2015). Network-based approaches, involving collaborations across multiple institutions, may help facilitate identifying more effective treatment strategies for women (Flynn et al., 2018).

SUBSTANCE USE DISORDER

Women with substance use disorder (SUD) face complex, multifactorial challenges across the life course. Although more men than women meet diagnostic criteria for SUDs (Fonseca et al., 2021; Koob and Moal, 2006)—a difference that has been shrinking for the last few decades (Fonseca et al., 2021; Keyes et al., 2008)—women progress from initial use to dependence at a faster rate and may exhibit more severe symptoms (Brady and Randall, 1999; Kosten et al., 1993).

The early stages of the United States opioid epidemic had a significant increase in nonmedical use of prescription opioids (Hirschtritt et al., 2018; Mazure and Fiellin, 2018). This trend was primarily driven by the treatment of chronic pain, which was more prevalent in women than men

(Racine et al., 2012; Templeton, 2020). Women were found to develop opioid use disorder (OUD) more rapidly than men (Hernandez-Avila et al., 2004; Mazure and Fiellin, 2018), often because clinicians were more likely to prescribe them to women (Mazure and Fiellin, 2018). Research indicates that it typically takes only 5 days to become physically dependent on opioids (Shah et al., 2017).

Biological Factors

Progress in understanding mechanisms linked to addiction-like behaviors have relied on rodents as experimental models. In general, female rodents, compared to male rodents, will self-administer substances of abuse and alcohol more rapidly, are more motivated to take cocaine (Cummings et al., 2011; Quigley et al., 2021a; Roberts et al., 1989), and show a higher drive to self-administer morphine and heroin (Cicero et al., 2003; Lynch and Carroll, 1999; Roth et al., 2002). They also consume more alcohol, relative to body weight, and engage in higher levels of cue-mediated alcohol-seeking behaviors (Cofresí et al., 2019; Crabbe et al., 2009). After periods of extended access to substances of abuse, female rats tend to exhibit greater symptoms of withdrawal, apart from alcohol and possibly opioids. In an animal model of relapse after forced abstinence, female rats compared to male rats exhibited greater cocaine-induced relapse (Lynch and Carroll, 2000), and greater drug- and cue-induced relapse (Cox et al., 2013).

Sex Chromosome Effects

Research in animal models has demonstrated the influence of sex chromosomes, independent of gonadal sex hormones, on addictive behavior. For example, preclinical studies using the FCG mouse model (see Chapter 2) show that after exposure to alcohol, XY mice developed a pattern of alcohol habit formation, and XX mice exhibited reward-seeking addictive behaviors regardless of gonadal phenotype (Barker et al., 2010). Additional research using this model also distinguished between sex chromosome and sex hormone effects on other aspects of alcohol addictive behaviors, including intake, preference, tendency for relapse, and habit formation. The results showed that the XX chromosome complement was associated with higher consumption initially and at relapse (Sneddon et al., 2022).

Studies analyzing cocaine addiction in the FCG animal model found both independent and interactive effects of sex chromosomes and sex gonadal hormones (Martini et al., 2020). Several addiction-associated genes are on the X chromosome—for example, genes encoding monoamine oxidase, an enzyme that regulates various neurotransmitters and implicated

in addiction mechanisms, and receptors for the neurotransmitters, GABA and glutamate (Krueger et al., 2023)—and leaky gene expression from the inactivated X chromosome in females could potentially explain sex differences in addictive behaviors (Krueger et al., 2023).

The Effects of Estrogens

Research has shown that estradiol, the most potent estrogen produced by the ovaries, modulates activity in the brain’s reward system in females and influences the acquisition, cravings, self-administration, and intake of substances. Female rats show enhanced motivation for cocaine at elevated estradiol levels, suggesting that estradiol facilitates their motivation for drug taking (Becker and Hu, 2008; Hu et al., 2003; Nicolas et al., 2019; Roberts et al., 1989). In gonad-intact female rodents, motivation is greatest during periods of the estrous cycle when estradiol is elevated (Becker and Hu, 2008; Becker and Koob, 2016; Roberts et al., 1989). Females in estrus exhibit greater drug-primed relapse for self-administering cocaine compared with those not in estrus and males, suggesting that estradiol enhances drug cravings that may contribute to the persistence of cocaine-seeking behaviors long into abstinence (Kippin et al., 2005). This notion is supported by studies showing that without estradiol replacement, adult females whose ovaries were removed have lower motivation for cocaine than those with estradiol (Becker and Hu, 2008; Hu et al., 2003; Nicolas et al., 2019). Sex differences in cocaine self-administration are evident, however, even in animals without gonadal hormones, with females without ovaries exhibiting greater intake than males (Hu et al., 2003). Exogenous estradiol is sufficient to enhance cocaine acquisition in females without ovaries (Hu and Becker, 2008; Hu et al., 2003; Lynch et al., 2001) but does not facilitate or enhance it in males (Jackson et al., 2005). Finally, sex differences in cocaine self-administration develop in animals that are not exposed to testosterone during the perinatal period and are exposed to ovarian hormones during puberty (Perry et al., 2013), indicating organizational influences of gonadal hormones on the brain that affect sex differences in this behavior.

Administered estradiol also facilitates the opioid acquisition in female rats without ovaries (Cicero et al., 2003; Lynch and Carroll, 1999) and enhances behavioral responses to and reinforcing effects of amphetamine (Robinson and Becker, 1986). During nicotine withdrawal, female rats with intact ovaries exhibit greater anxiety-like behaviors and stress responses compared with rats without ovaries (Torres and O’Dell, 2016). Intake of ethanol is higher during the stage of the estrous cycle in intact females when they are no longer receptive to the male (Forger and Morin, 1982; Roberts et al., 1998; Witt, 2007).

Other Biological Factors and Mechanisms

The transition from casual use of substances to developing a SUD in either sex is mediated by the brain’s reward system, with neurons responsive to the neurotransmitter dopamine activated in response to stimuli necessary for health and reproductive success, such as food consumption, sexual behavior and social interactions (Everitt et al., 1999; Robinson et al., 2015; Schultz, 1986; Wise and Rompre, 1989). Substances act on the reward pathway to induce dopamine neurotransmission either directly or indirectly, and repeated drug use causes physiological changes in the brain that include enhanced dopamine release (Engel and Jerlhag, 2014; Koob and Le Moal, 2008; Robinson and Becker, 1986). Repeated exposure to the acute rewarding effects of drugs can lead to escalated intake and compulsive drug-taking behaviors (Ahmed and Koob, 1998; Chartoff et al., 2009; Wise and Koob, 2014). Sex-related differences occur in neural responses to external stimuli and internal endocrine and physiological signals. Consequently, even when behaviors are the same in both sexes, sex related differences may be present in the neural circuitry activated and the brain’s response to a given situation (Farrell et al., 2013; Orsini et al., 2022; Quigley et al., 2021b).

The Maternal in Utero Environment

Exposure to addictive substances in utero predisposes offspring to developing SUD later in life. Research using animal models has demonstrated that maternal exposure to addictive substances, such as opioids or a high-sugar diet, increases the vulnerability of offspring to drug-seeking behaviors in adulthood (Abu and Roy, 2021; Gawliński et al., 2020). Human studies have shown that prenatal substance exposure plays a critical role in developing addictive behaviors as offspring get older. Alcohol exposure during pregnancy predicts drug dependence behaviors in adolescent and adult offspring after adjusting for certain confounding factors, including family history, prenatal smoking, current parental drinking habits, the environment, and other sociodemographic factors (Baer et al., 1998, 2003; Dodge et al., 2023). Similarly, prenatal cocaine exposure is associated with developing SUD in adolescent and adult offspring.

Although these studies either found no sex differences (Min et al., 2014) or involved small sample sizes that precluded sex-based analysis (Min et al., 2023), more recent studies have provided evidence of sex bias. In one study, prenatal substance use exposure was directly related to substance use progression in girls but not boys during adolescence (Marceau et al., 2021). In addition, findings from the Helsinki Birth Cohort Study have shown that certain measures of prenatal growth—considered a surrogate marker for exposures in utero—was associated with hospitalization for SUD in female adult offspring, suggesting that the fetal period may

be a critical developmental period during which women could develop sensitivities for SUD (Lahti et al., 2014). Overall, several lines of evidence support the maternal in utero environment as a major determinant in SUD susceptibility. However, as developing an addiction involves multiple factors, further research is needed to understand the sex-related vulnerability of women to SUD.

Specific Risk Factors in Women

Higher alcohol intake is associated with higher premenstrual or menstrual distress and negative affective states in women. This supports the hypothesis that women with PMS have higher levels of alcohol use or abuse (Evans and Levin, 2011; Kiesner, 2012; Mello et al., 1990; Svikis et al., 2006). Furthermore, women who have a history of, or are currently using cocaine, opioids, or alcohol have evidence of menstrual cycle disruptions (Emanuele et al., 2002; Mello et al., 1990, 1997; Schmittner et al., 2005). When the menstrual cycle is disrupted, it is more difficult to discern how, or if, ovarian hormones are influencing drug taking or motivation for it. In women aged 18–35, estradiol enhances the positive subjective effects of stimulant drugs, such as amphetamines, similar to findings from studies using animal models (Justice and de Wit, 1999).

Women often report using drugs initially to cope with negative affective states, such as depression (Kuntsche and Müller, 2011; Müller and Kuntsche, 2011), whereas men usually cite peer influence or sensation-seeking (Kuntsche and Müller, 2011). Other sex-related risk factors for substance use and SUD include IPV, which is more likely to affect women, and other social stressors such as death of child or loved one and divorce (NIDA, 2020). Mental health conditions are also more common among women who use substances of abuse. In particular, posttraumatic stress disorder has a strong relationship with developing a SUD, especially among young girls (NIDA, 2020).

Condition Presentation and Diagnosis

Diagnostic criteria for SUD are the same for men and women, despite gender differences in how SUDs manifest. Some research has demonstrated that women entering SUD treatment programs present with a more severe clinical profile and an array of related social, medical, psychological, and multiple mental disorders (Mazure and Fiellin, 2018). Research has shown that depression, which is more prevalent in women, is associated with drug cravings and relapse (Curran et al., 2007; Zilberman et al., 2007), making it important to account for depressive symptoms when women present with concern about an SUD. Furthermore, although women may have used

substances for a shorter duration and in smaller amounts than men, the acceleration of SUD severity and deterioration in functionality occurs more rapidly (Fonseca et al., 2021; Mazure and Fiellin, 2018).

Treatment and Management

Despite women accounting for one-third of substance users, only one-fifth of the individuals receiving SUD treatment are women (UNODC, 2015). Women face greater barriers than men in accessing SUD treatment and harm reduction programs, such as syringe service programs, opioid overdose services, and drug testing. Based on National Survey of Drug Use and Health (NSDUH) data, less than 10 percent of women of reproductive age in need of SUD treatment were receiving it (Martin et al., 2020). Figure 6-2 shows that from 2007 to 2014, treatment receipt for SUD remained relatively stable, whereas it increased for depression. Systemic, structural, social, and cultural barriers that limit access to SUD treatment include stigma, punitive attitudes, discrimination, rigid admission requirements, and inflexible scheduling (see Chapter 7).

Disparities

The literature on differences in prevalence and mortality from overdose by race, ethnicity, and sexual identity is sparse (Schuler et al., 2020). A few studies have reviewed factors that may be associated with SUD for different

racial and ethnic groups. For example, studies have found that Latinas mainly from urban and metropolitan communities and born in the U.S. had greater odds of SUD, including OUD (Castañeda et al., 2019; Takada et al., 2024). A 2019 study based on data from the NSDUH found that Native Hawaiian and Pacific Islander women had a higher prevalence of SUD, whereas Asian American women had higher rates of stimulant use compared to other racial and ethnic groups (Wang et al., 2023a). When comparing the relative risk of heavy episodic drinking and marijuana use, lesbian and bisexual women of color had an increased risk of use (Schuler et al., 2020).

Few studies have considered the role of both gender and race and ethnicity on treatment use (Pinedo et al., 2020). Studies have shown that Black and Hispanic women are less likely to receive treatment for SUD (Martin et al., 2020) and OUD (Iobst et al., 2024; Scheidell et al., 2024). Black, Latina, and other women of color from low-income communities wait longer before entering treatment and have a shorter length of stay in methadone treatment programs compared to men (Marsh et al., 2021).

Research Gaps

Differences in the biology of substance use are evident, influenced by factors such as ovarian hormones, drug type, dose, and exposure history (McHugh et al., 2018). It is also critical to understand how sex chromosome effects differentially influence addiction patterns in male and female individuals, which likely starts early in development via organizational effects on brain anatomy and function. Given the complexity of sex differences across bodily systems, multiple levels of analysis are needed. (McHugh et al., 2018).

Women develop drug dependence faster than men, but most research on risk factors and rates of progression from first use to SUD in women have been retrospective and not yielded a good understanding of the risk and protective factors in women and men. The Adolescent Brain Cognitive Development study, funded by the National Institute on Drug Abuse, will provide prospective data and analysis, and may begin to fill several of these research gaps (NIDA, 2023).

Despite no differences in how SUDs are diagnosed in men versus women, differences in gender and sex appear related to when they present for diagnosis (Mazure and Fiellin, 2018). Women tend to present symptoms of chronic pain and increased pain sensitivity and are thus more likely prescribed painkillers by physicians to alleviate symptoms (Mazure and Fiellin, 2018).

Diagnostic criteria recognize that SUD may manifest differently in women and men, but research is needed to determine if different diagnostic criteria are needed. Research assessing early interventions in women who

experience negative affective states, such as depression, or are exposed to adversities, such as ACEs, that may make them prone to drug use needs to be a priority. Considering the complexity of SUDs and the opioid epidemic, enhancing surveillance systems to encompass the intersectionality of factors such as gender, race, ethnicity, and geographic location will enhance screening and treatment approaches for diverse groups of women (Barbosa-Leiker et al., 2021).

HIV/AIDS

Chapter 3 noted that the rates of new HIV diagnoses and persons living with diagnosed HIV infection are much lower in women than men (CDC, 2023b). Although women represent a small proportion of new cases in the U.S. population living with HIV/AIDS, they have important and distinct characteristics of infection. To reduce the burden of HIV infection in both men and women, the White House Office of National AIDS Policy developed the National HIV/AIDS Strategy and implementation plan for Ending the HIV Epidemic in the United States (CDC, 2023a; HIV.gov, 2023). The stated goal is to reduce the number of new HIV infections and includes recommendations across four pillars of action: test everyone for HIV or DIAGNOSE infection, TREAT those who test positive with antiretrovirals and ensure viral suppression, PREVENT those who test negative from becoming infected by providing pre-exposure prophylaxis (PrEP) medication and syringe services for those who inject drugs, and RESPOND by predicting new outbreaks (CDC, 2023a).

Biological Factors

Sex Chromosome Effects

Several important male versus female sex chromosome differences can influence various aspects of HIV infection. Differences in the sex complement can lead to variations in gene expression patterns, including those related to immune response and susceptibility to infections such as HIV (Moran et al., 2022; Scully, 2018). Furthermore, specific X chromosome genes encode proteins that play a crucial role in the innate immune response, including toll-like receptor-7 (TLR-7) and TLR-8, which are involved in detecting viral infections (Moran et al., 2022). As noted in Chapter 2, females undergo X chromosome inactivation to balance gene expression between the sexes. If the second X chromosome escapes inactivation, that leads to differences in gene dosage, which could contribute to variations in HIV susceptibility, immune response, and disease progression (Moran et al., 2022; Scully, 2018).

Epigenetic mechanisms, such as DNA methylation and histone modifications, can regulate gene expression without altering the underlying DNA sequence. Sex-specific epigenetic patterns may influence immune responses to HIV and contribute to differences in disease outcomes (Chlamydas et al., 2022). Understanding these sex chromosome differences and their interactions with hormonal and epigenetic factors is crucial for elucidating sex-specific disparities in HIV infection and developing tailored prevention and treatment strategies for both sexes.

The Effects of Estrogens

Female sex hormones have profound effects on multiple systems throughout the body and over the life-span. Female sex hormone fluctuations, such as during the menstrual cycle, in pregnancy, and from the use of contraceptives, influence the risk for HIV infection directly (Szotek et al., 2013) or indirectly through modulatory effects on the immune response (Swaims-Kohlmeier et al., 2021). During the menstrual cycle, there are changes in the vaginal tissue physiology and microbiome that can increase susceptibility to infection (Boily-Larouche et al., 2019). High levels of estrogens during the menstrual cycle may be protective and are associated with lower HIV acquisition (Asin et al., 2008; Moran et al., 2022; Wira et al., 2015). Progestin-only contraceptives, however, have been associated with an increased risk of acquiring HIV, although the precise mechanism is unknown (Moran et al., 2022; Ralph et al., 2015; Rodriguez-Garcia et al., 2021).

Pregnancy induces notable alterations in immune responsiveness (Akoto et al., 2021). The proinflammatory immune states observed during the first and third trimesters contrast with the anti-inflammatory state in the second trimester, and these changes could influence susceptibility to HIV infection (Rodriguez-Garcia et al., 2021). In addition, the molecular signals involved in mounting an immune response and the antimicrobial environment of the reproductive tract change significantly during pregnancy, which can also influence susceptibility to HIV infection. The cervical plug formed during pregnancy contains immune factors that exhibit potential anti-HIV activity (Mhlekude et al., 2021). Despite these insights, research has yet to definitively discern whether pregnancy heightens or decreases the risk of HIV infection and indicates an important research gap (Rodriguez-Garcia et al., 2021).

Sex hormone levels fall during menopause, affecting many immune biomarkers in the female reproductive tract. Research using models of cervical tissue has provided evidence of immune activation and increased HIV replication in menopausal tissue, indicating that menopausal women may have specific biological risks for acquiring HIV (Rodriguez-Garcia et al., 2021).

Specific Risk Factors in Women

Sexual behaviors and drug use are major risk factors associated with HIV infection among women. Eighty-four percent of new diagnoses among women are attributed to heterosexual contact and 16 percent to intravenous drug use (CDC, 2021b). A recent study of HIV risk behaviors in women found that 7 percent of cisgender women with HIV had had condomless sex in the past 12 months (CDC, 2022), which increases the risk of not only HIV but also other sexually transmitted infections, such as chlamydia, gonorrhea, and syphilis (NASEM, 2021). It is now well understood that sexually transmitted infections can increase susceptibility to HIV infection and render persons with HIV more infectious (NASEM, 2021).

Intravenous and other drug use has been increasing in the United States, leading to new HIV infections among persons who use drugs (Springer et al., 2020a, 2020b); sharing and reusing needles, syringes, or other injection equipment can increase the risk of getting or transmitting HIV. One study reported that women who share needles had a significantly higher likelihood of contracting HIV compared to men, with the authors attributing that to multiple and overlapping sexual and drug use partners (Riehman et al., 2004; Valente et al., 2001). Substance use can also affect judgment and lead to sexual behaviors that increase the risk of HIV, such as unprotected sex or sex with multiple partners (Riehman et al., 2004; Valente et al., 2001).

Condition Presentation and Diagnosis

In the early stage of infection, women have higher CD4+ T cell counts and lower HIV plasma RNA loads, but both men and women progress to AIDS at similar rates. Women, however, have been seen to progress to AIDS at higher rates compared to men of equivalent HIV RNA load. This has implications for screening, testing, and treatment (Farzadegan et al., 1998; Moran et al., 2022; Ziegler and Altfeld, 2016).

Early in the AIDS epidemic, the Centers for Disease Control and Prevention (CDC) set the case definition of AIDS based primarily on conditions seen in gay men and without fully capturing the experience of women (McGovern, 1994; OTA, 1992). Gynecological conditions, such as cervical dysplasia, pelvic inflammatory disease, and recurrent yeast infections, more common in HIV-infected women were not included. After significant protests by women and others, CDC changed its definition of AIDS; candidiasis and invasive cervical cancer are now among the listed conditions affecting women (CDC, 2021a).

Treatment and Management

PrEP and Treatment

PrEP is the standard approach to HIV prevention for people who are at risk. The regimen consists of three FDA-approved medications that prevent infection by almost 99 percent: two daily oral formulations and one injectable medication administered every 8 weeks. USPSTF released new guidance in August 2023 (Liu et al., 2023; USPSTF et al., 2023) with an “A” recommendation, indicating high certainty of substantial benefit from these medications to prevent HIV. Although the national goal is to increase the estimated people with PrEP indications actually being prescribed PrEP to at least 50 percent by 2025 and remain at 50 percent by 2030, only 10 percent of women who could benefit from PrEP were prescribed it in 2019, indicating major barriers to access for women (CDC, 2021d).

Another goal is to provide life-saving combination antiretroviral therapy³ (ART) to all persons within 7 days of a positive HIV test (Gandhi et al., 2023) and to achieve 100 percent viral suppression. Staying in HIV care is important to achieving and maintaining viral suppression and is improving health and morbidity for the individual and reducing transmission to uninfected individuals (Eisinger et al., 2019). In 2020, 23 percent of cisgender women with HIV reported missing at least one medical appointment in the past 12 months (CDC, 2023c), and approximately 63 percent of them with HIV reported taking all of their doses of ART over the last 30 days (CDC, 2023c). In 2019, only 64 percent of women with diagnosed HIV in 44 states and the District of Columbia were virally suppressed (CDC, 2021d).

Having access to needed ancillary services could reduce barriers to achieving and maintaining viral suppression; the top three services cisgender women with HIV reported needing but not receiving in the past 12 months in 2020 were dental care, the Supplemental Nutrition Assistance Program and Special Supplemental Nutrition Program for Women, Infants, and Children, and shelter and housing services (CDC, 2023c). Six percent of cisgender women with HIV reported homelessness in the past 12 months, and 21 percent reported symptoms of depression or anxiety (CDC, 2023c), both of which research has shown to affect the ability to access health care, HIV care, and ART and maintain viral suppression.

___________________

³ ART is a regimen for treating HIV-positive individuals that reduces the levels of HIV in the body. PrEP is meant to prevent infection and taken before a person thinks they might be exposed.

With appropriate ART, women with HIV can live longer. In fact, life expectancy for people with HIV is approaching that of the general population. Some individuals living with HIV experience multiple comorbidities such as cardiovascular disease (CVD), diabetes, renal disease, neurocognitive disease, and cancer. In one study, women with HIV had significantly more ageing-related comorbidities than men with HIV. Women had higher prevalence of diabetes, bone disease, and lung disease than men (Collins et al., 2023). Multiple chronic conditions pose challenges for women, health care providers, and health care systems involved in managing them. These challenges are discussed further in Chapter 8.

Disparities

Disparities in HIV prevention and care for women persist, particularly among sexually, racially, and ethnically minoritized populations. Although these populations are well studied and progress has been made along the HIV care continuum, further work is needed to understand their needs for prevention and treatment, as the majority of research has focused on mother-to-child transmission (Nwangwu-Ike et al., 2023).

Racial and ethnic differences in HIV incidence in women are well known and captured by national surveillance data (see Chapter 4) (CDC, 2023b). Given that the prevalence and incidence is greatest among Black women, followed by multiracial, and Hispanic/Latina women, this disparity clearly needs to be addressed. One study following a prospective cohort of women from the HIV Outpatient Study (Buchacz et al., 2020) found that the odds of using and adhering to ART was lower among Black women compared to Hispanic/Latina and White women (Geter et al., 2019). Furthermore, the rate of achieving viral load suppression was lower among Black and Hispanic/Latina women (see Figure 6-3).

Qualitative studies have described barriers to care encountered by Black and Hispanic/Latina women living with HIV. Among Hispanic/Latina women, a systematic review cited emergent barriers to seeking HIV services that included lack of social support, insurance coverage, out-of-pocket fees, mental health outcomes, HIV-related stigma, and unique barriers such as language; fear of legal consequences, including deportation; and limited access to documentation (Geter Fugerson et al., 2019). Key features of quality of care for Black and Hispanic/Latina women included care that is integrated and coordinated with other support services, compassionate and nonjudgmental care, and shared decision making (Rice et al., 2020). Thus, HIV health care services require tailoring for these populations.

Research Gaps

Significant research gaps exist regarding the safety and efficacy in women of the medications used to prevent and treat HIV, in part because women have been and continue to be underrepresented in research studies of PrEP and ART. A 2016 systematic literature review reported that women represent a median of 19.2, 38.1, and 11.1 percent of participants in ART and PrEP, prophylactic vaccine, and HIV cure strategies studies, respectively (Curno et al., 2016). One reason for this low participation is that most PrEP studies focus on enrolling solely men because they represent 80 percent of people living with HIV (CDC, 2021c). Other reasons include greater stigma affecting women with HIV compared to men (Karim et al., 2022) and the lack of women-specific recruitment strategies to educate and enroll women.

Without adequate representation of women in clinical trials of PrEP and ART, significant gaps remain in understanding gender differences in the virological, immunological, and clinical presentation of HIV. Specifically, studies that consider women-specific biological factors and their potential influence on PrEP are needed. Such studies would seek to better understand the role of sex hormones, genital inflammation, genital microbiome, sexually transmitted infection, and sex differences in immune responses

and mechanisms and how they may influence the safety and effectiveness of PrEP and ART (Karim et al., 2022). Clinical studies are also needed to better understand sex-related issues and PrEP, including for transgender women, and women who use drugs during the reproductive period along with hormonal contraception, during pregnancy and breastfeeding, and during menopause (Karim et al., 2022).

Research on improving HIV prevention and treatment for women is lacking, including research focusing on patient preferences regarding oral versus injectable ART or PrEP. Such studies would inform clinicians’ understanding of patient preferences about the settings in which to start ART and how to link and differentiate care by gender, comorbidity, and setting. Randomized prospective trials and studies powered to detect sex differences in access to treatment and prevention are also lacking.

Trials on HIV prevention and treatment that have an insufficient number of women do not have the power to differentiate medication effects by sex, which leave gaps in clinician knowledge around safety and efficacy (Gandhi et al., 2023). Without rigorous published studies that prospectively evaluate the effect of sex differences on preventing and treating HIV, recommendations specific to the needs of women will remain limited.

MIGRAINES/HEADACHES

Migraine disproportionately affects women in the years when they are most productive and active. Studies have shown that it is a complex, multifactorial disease with genetic factors increasing susceptibility and non-genetic factors regulating clinical presentation (Ferrari et al., 2015). The higher prevalence of migraine in women involves many factors, discussed next.

Biological Factors

During a migraine attack, multiple parts of the central and peripheral nervous systems are activated simultaneously (Charles, 2018). Migraine has several key mechanistic features. One mechanism involves central nervous system (CNS) processing of sensory input from the trigeminocervical complex (Bartsch and Goadsby, 2003) that activates the brain’s pain-processing center and the release of neuropeptides (Charles, 2018; Goadsby and Holland, 2019). The developmental differences in male and female pain-sensing systems and the effect of circulating hormones are the main physiological factors driving the higher prevalence of migraine among women (Pavlovic et al., 2017).

Cortical spreading depression—a slowly propagating wave of altered brain activity that involves changes in neuronal, glial, and vascular function (Charles and Baca, 2013)—is another important mechanism. Animal studies have shown a lower threshold to initiate cortical spreading depression in

female compared to male mice, regardless of the phase of estrous cycle (Brennan et al., 2007). Functional magnetic resonance imaging (MRI) studies have found structural changes in the specific brain regions involved in pain processing in women and men with migraine compared to healthy women and men (Maleki et al., 2012). Another study found reduced functional connectivity in sensorimotor networks and one of these same brain regions in women during migraine attack (Araújo et al., 2023).

Genetic factors may contribute to the threshold for triggering migraine in women compared to men. Population-based studies identified 38 genomic loci (Gormley et al., 2016) and genome-wide association studies (GWAS) identified 13 genes (Ferrari et al., 2015) as disease-associated gene variants. However, researchers have not found associations of these genetic variants with sex differences that explain the high prevalence of migraine in women, suggesting that other factors, perhaps hormonal differences, play significant roles in the clinical presentation.

Sex Chromosome Effects

Limited results address sex chromosome effects on the pathobiology of migraine. However, genetic studies have revealed several migraine susceptibility loci on the X chromosome, indicating possible roles of its genes. One study identified such a locus on Xq12, which includes the HEPH gene that encodes an enzyme involved in iron homeostasis, and others have implicated Xp22, Xq27, and Xp28 (Maher et al., 2012a; Maher et al., 2012b; Wieser et al., 2010). A case-control study has shown a link with the X chromosome gene SYN1, which encodes synapsin, a protein involved in neurotransmitter release at neuronal synapses (Quintas et al., 2020).

The Effects of Estrogens

Research has shown that female hormonal changes during puberty, pregnancy, and perimenopause can trigger migraine episodes, suggesting that ovarian hormones, notably estrogens, play an important role (Pavlovic et al., 2017). The “estrogen withdrawal effect” is the most commonly accepted theory for menstrual migraine, and women with histories of migraine exhibit faster declines in estradiol levels during the late luteal phase of the menstrual cycle (Pavlovic et al., 2017). Reductions in circulating estrogens before and during menstruation are associated with migraine episodes, suggesting that estrogens somehow hinder the abnormal neural activity associated with migraine (MacGregor and Hackshaw, 2004; Somerville, 1972).

Estrogens exert multiple effects on neural pathways involved in migraine, including the trigeminovascular pathway and central pain

pathways that traverse the central and peripheral nervous systems. These pathways contain neuronal populations that express ERs (Pavlovic et al., 2017). Animal studies have also identified ERs in trigeminal ganglion neurons (Bereiter et al., 2005; Puri et al., 2006). Estrogens have been reported to affect the excitability of second-order trigeminal sensory neurons involved in craniofacial pain syndromes, which was proposed as a mechanism shared in migraine (Cairns, 2007). In addition, estrogens, as well as progesterone, regulate serotonergic and GABAergic neurotransmitter systems, which are also involved in migraine pathophysiology (Vetvik and MacGregor, 2021).

Research has provided more detailed mechanistic insight into how estrogens are involved in migraine. It is thought to regulate migraine episodes by increasing expression of the oxytocin receptor and levels of the analgesic hormone oxytocin (Amico et al., 1981; Miller et al., 1989; Murata et al., 2014), which has also been implicated in preventing migraine attacks (Krause et al., 2021; Phillips et al., 2006; Tzabazis et al., 2017). Patterns of fluctuating plasma levels of oxytocin and estrogens are similar during the menstrual cycle (Engel et al., 2019), suggesting that reductions in oxytocin may also trigger migraine (Krause et al., 2021). In addition, evidence shows that estrogens modulate levels of calcitonin gene-related peptide (CGRP) (Cetinkaya et al., 2020; Labastida-Ramírez et al., 2019), a known contributor to migraine through its effects on vasodilation and neurogenic inflammation (Karsan and Goadsby, 2015). Research has shown that CGRP release is lower when levels of estrogen are high (Pavlovic et al., 2017), suggesting that estrogen withdrawals during certain phases of the menstrual cycle may increase CGRP and the likelihood of triggering an attack. Estrogens also regulate other analgesic neuropeptides and hormones, such as vasopressin (Lagunas et al., 2019), prolactin (Avona et al., 2021; Franchimont et al., 1976), and orexin (Porkka-Heiskanen et al., 2004), indicating that estrogens may mediate its effects on migraine through multiple mechanisms.

Specific Risk Factors in Women

Reproductive Milestones

The prevalence of migraine changes throughout a woman’s life (Pavlović, 2021). Migraine first appears during puberty, occurs around the time of menstruation, fluctuates through pregnancy, worsens during the perimenopausal stage, and usually improves after menopause (Burch, 2020; Pavlović, 2021). The high prevalence of migraine and severe disability rate that accompanies it during the productive ages can reduce productivity at work and impair family and social functioning (Buse et al., 2019).

Migraine and Oral Contraceptive Use

The risk of stroke is twofold higher in women experiencing migraine with aura as compared to women without migraine and sixfold higher in women using estrogen-containing contraceptives (Lee et al., 2023a). CDC’s U.S. Medical Eligibility Criteria of Contraceptive Use states that estrogen-containing contraceptives have an unacceptable health risk for women who experience migraine with aura, but the benefits for women who have migraine without aura outweigh the risks (CDC, 2016). At one time, research tied estrogen-containing contraceptives to a high risk of stroke (Calhoun, 2017), but today’s contraceptives contain much lower levels of estrogens, which appears to eliminate that excess risk. They can help control menstrual-related migraine and painful menstruation by stabilizing levels of estrogen (Nappi et al., 2022).

Gender Roles and Mental Health Factors

The 2010 ACEs study reported a higher prevalence of frequent headaches in individuals with ACEs (Anda et al., 2010). In the American Migraine Prevalence and Prevention Study, a history of emotional abuse was more common in women who experience migraine even after adjusting for depression and other sociodemographic factors (Tietjen et al., 2015), and data from the 2012 Canadian Community Health Survey-Mental Health showed that migraine was more prevalent in both women and men who experienced violence or sexual or physical abuse as a child (Brennenstuhl and Fuller-Thomson, 2015). Several studies have found that migraine is more prevalent in women who have experienced IPV compared to those who have not (Coker et al., 2000; Cripe et al., 2011; Gelaye et al., 2016; Vives-Cases et al., 2011), although more research is needed.

Despite some studies on the effects of IPV on a woman’s physical and mental health, not enough studies explore its relation to migraine. The known effects of maltreatment on the hypothalamic-pituitary-adrenal axis and stress-mediating homeostatic systems is the presumed mechanism linking childhood maltreatment and migraine (Tietjen et al., 2016).

Condition Presentation and Diagnosis

The International Classification of Headache Disorder third edition (ICHD-3) is the latest system for classifying headache disorders (IHS, 2018; Levin, 2022). The diagnostic criteria in ICHD-3 has been internationally accepted and used in headache-related studies. Migraine is a primary headache disorder with two major types, with and without aura. Migraine without aura for adults is defined as a minimum of 4 hours of headache

with specific symptoms, including nausea and an abnormal sensitivity to light and sound. Migraine with aura is identified with transient neurological symptoms accompanied by or preceding the headache.

Between 18 and 60 percent of women who experience migraine report that they occur most often during menstruation (Vetvik and MacGregor, 2021). Menstrual migraine is included in the appendix of ICHD-3, but the diagnostic criteria still require validation (Cupini et al., 2021). Based on those criteria, menstrual migraine occurs in two forms: pure and menstrual-related (Ceriani and Silberstein, 2023). Women with pure menstrual migraine report having an attack between 2 days before and 3 days after the start of menstruation and none at other times. Women who experience menstrual related migraine report attacks related to their menstruation and also at other times. Menstrual migraine lasts longer, causes more disability, and does not respond as well to acute treatment (Granella et al., 2004; Pinkerman and Holroyd, 2010).

Treatment and Management

“Throughout my life, I have been on and off treatments. Some have been helpful. Some have not. There have been significant side effects with a lot of the treatments and I faced a lot of challenges during graduate school, my working life, particularly during my pregnancies and while breastfeeding my children to find therapies that were safe and hopefully allow[ed] me to function . . . a lot of women delay pregnancy, go without treatments during certain times of life and that can negatively affect our lives and the lives of those around them. I have missed out on moments or times with family and friends.”

—Presenter at Committee Open Session

Acute Migraine Treatment

Migraine causes more than 1.2 million visits to U.S. emergency departments annually and is the third leading cause of emergency department visits in reproductive-aged women (Burch et al., 2018; Minen et al., 2018). Several families of drug—triptans, such as sumatriptan (Imitrex^®), and ditans such as Lasmiditan (Reyvow^®) are FDA approved as acute treatments (Yang et al., 2021). In March 2023, FDA approved zavegepant (Zavzpret^®), a nasal spray that blocks CGRP activity and can bring relief within 30 minutes, to treat acute migraine with or without aura (Dhillon, 2023).

Guidelines based on high-quality, evidence-based approaches are not available for emergency room acute treatment (Robblee and Grimsrud, 2020). In addition, lack of referral or follow-up plan with a neurologist or headache specialist causes high readmission rates to emergency departments (Giamberardino et al., 2020), with one study finding that more than

a quarter of initial visitors for migraine return within 6 months (Minen et al., 2018). More studies are needed on acute care in emergency rooms to develop nationally accepted treatment protocols.

Management has changed with the approval of CGRP-blocking drugs as preventive agents. These new agents include rimegepant (Nurtec^®), ubrogepant (Ubrelvy^®) and atogepant (Qulipta^®). In addition, monoclonal antibody therapies that block CGRP have recently been developed and hold much promise for preventing migraine: three FDA-approved, self-injectable drugs—galcanezumab (Emgality^®), erenumab (Aimovig^®) and fremanezumab (Ajovy^®)—and andeptinezumab-jjmr (Vyepti^®), which is an intravenous infusion every 3 months (Pope, 2023).

Migraine Diagnosis and Treatment During Pregnancy

Even though acute headache during pregnancy should always raise a concern about secondary headache disorders, migraine is the most common cause of these headaches (Burch, 2019). Although migraine is primarily a headache disorder, it increases the risk of preeclampsia, cerebral blot clots, and pregnancy-associated stroke compared to women without a history of migraine (Greige et al., 2023).

Up to 10 percent of pregnant women can start to experience migraine without aura, while the onset of migraine with aura has been reported up to 14 percent (Negro et al., 2017), and approximately 60–80 percent report that their migraine symptoms improve during pregnancy (Allais et al., 2019). Based on the data from American Registry for Migraine Research, 20 percent of women who experience migraine avoid pregnancy for fear of their migraines worsening, of having a difficult pregnancy with migraine disability, and the potential adverse effect of their migraine treatment on the fetus (Ishii et al., 2020).

The recommended first-line treatment for migraine during pregnancy is nonpharmacological therapies, including lifestyle changes, cognitive behavioral therapy, and biofeedback (Verhaak et al., 2023). However, accessing these therapies is not easy for pregnant women. Therefore, pharmacological therapies should be considered for those who either do not have access to nonpharmacologic treatments or did not benefit from them. One study supports that triptans are safe to take during pregnancy; however, a control group was not included for comparison (Ephross and Sinclair, 2014).

Based on the 2022 American College of Obstetricians and Gynecologists updated guideline, little high-quality data exist on treating migraine during pregnancy, leaving treatment decisions up to the patient and clinician (ACOG, 2022). Lack of data leads clinicians to hesitate when starting treatment for chronic migraine in pregnant patients. On a recent survey with women’s health care providers in Connecticut, 60 percent reported

they did not feel comfortable starting new preventive treatments during pregnancy, and 40 percent reported referring such patients to headache specialists or neurologists rather than starting treatment (Verhaak et al., 2023).

Disparities

Disparities among racial and ethnic groups exist regarding the severity of headaches. African American individuals reported more frequent and intense headaches and were more likely to discontinue treatment for migraine in specialty clinic settings than White individuals (Kiarashi et al., 2021). American Indian and Alaska Native individuals had the highest prevalence of migraine among all U.S. racial and ethnic groups, with the lowest rate among Asian Americans (Burch et al., 2018). A 2021 literature review found that only 46 percent of Black individuals received care for their migraines compared to 72 percent of White individuals and that only 14 percent of Black patients received a prescription for migraine medication compared to 37 percent of White patients (Kiarashi et al., 2021). The same survey found that Hispanic and Latina/Latino individuals were 50 percent less likely to receive a migraine diagnosis and were less likely to receive a prescription for migraine prevention medication than White individuals.

Transgender and gender-diverse individuals are affected significantly by migraine. Some evidence has suggested that transgender women who have received gender-affirming hormone therapy may have a higher risk of migraine resulting from the effect of estrogens on the nervous system/brain (Ahmad and Rosendale, 2022). A study based on National Health Interview Survey (NHIS) data from 2013 to 2018 found that bisexual women had 33 percent greater odds of headache/migraine than did lesbian women. The same study found the risk of migraine was 25 percent higher in sexual minority women compared to heterosexual women (Heslin, 2020).

Research Gaps

Research has long established that more women than men experience migraine, but the pathophysiological mechanisms and genetic influences underlying sex differences in the presentation and frequency of occurrence throughout the life-span are not well understood. This is in part the result of most preclinical research using animal models that only focus on males (Eisenstein, 2020). One of the very common and debilitating forms for women, menstrual migraine, has only been included in the appendix of ICHD-3 and is still waiting to be validated as a diagnosis. The lack of diagnostic certainty leads to exclusion from treatment trials; there are no specific treatment options available.

Based on the data, 60–80 percent of women with migraine report improvement during pregnancy, but no literature addresses why the 20–40 percent do not. Evidence on the treatment options regarding benefits or risks during pregnancy, breastfeeding, or postpartum is insufficient. Specific data are also lacking on dosing or the effect of comorbidities for medications during pregnancy. New studies are also needed to clarify the specific formulations of combined hormonal contraceptives and the risk of stroke associated with them in different subtypes of migraine. Overall, more data are needed to develop treatment guidelines for pregnancy.

Despite evidence supporting an association between migraine and trauma (e.g., ACEs, interpersonal violence), more research is needed to explore how social factors influence susceptibility to migraine. No data exist on the effect of infertility treatment or gender-affirming hormone therapy on migraine and risk of stroke in patients with history of migraine. The effect of migraine on sexual and gender minority groups is challenging to study, and studies have conflated sex and gender when collecting this information. Nationally accepted treatment approaches in emergency room for acute migraine care are needed, as are strategies to reduce the need for emergency care.

CARDIOMETABOLIC CONDITIONS

“There has to be more attention paid to awareness [of CVD symptoms in women] . . . in the medical community and hospital emergency rooms . . . women talk so often of just being dismissed or diagnosed with GERD [gastroesophageal reflux disease] . . . and not taken seriously that they may be having a heart attack.”

—Presenter at Committee Open Session

Cardiometabolic conditions cause significant morbidity and mortality in women. These conditions often occur together, and causes are multifactorial, in which their development can be explained by various factors. This section focuses on two chronic cardiometabolic conditions with high morbidity and mortality in women—CVD and stroke—and highlight sex and gender differences in known risk factors for both. Other cardiometabolic conditions, including obesity, metabolic syndrome, and Type 2 diabetes, are also discussed as risk factors.

Biological Factors

A discussion of cardiometabolic conditions encompasses multiple major clinical heart and circulatory conditions—including stroke, brain health, complications of pregnancy, kidney disease, congenital heart disease,

rhythm disorders, sudden cardiac arrest, subclinical atherosclerosis, coronary heart disease, cardiomyopathy, heart failure, valvular disease, venous thromboembolism, and peripheral artery disease—the associated risk factors, including diabetes and obesity; and outcomes, including quality of care, procedures, and economic costs. Although many cardiometabolic conditions have overlapping biological mechanisms, some mechanisms are specific to a pathology. To summarize potential key factors that regulate the etiology of cardiometabolic conditions in women, this section highlights findings for the consensus on the more general role of key biological factors that affect women.

Sex Chromosome Effects

The contribution of the X chromosome to developing cardiometabolic conditions is not fully understood, largely because the majority of GWAS studies do not include sex chromosomes (Regitz-Zagrosek and Kararigas, 2017). However, studies in the FCG model have identified specific differences related to metabolism, with the XX, compared to the XY chromosome complement promoting fat accumulation related to increased food intake during the inactive phase of the diurnal cycle and metabolic inflexibility to switch between carbohydrate and fat fuels for energy metabolism (Chen et al., 2012, 2015; Link et al., 2020).

The XX chromosome complement in mice is also associated with elevated cholesterol levels and atherosclerosis (AlSiraj et al., 2019; Link et al., 2015), suggesting that the vascular protective effects for females comes from gonadal hormones rather than the X chromosome. This is consistent with studies showing that in female mice that have no estrogen effects after surgical removal of the ovaries, the X chromosome promoted cardiac injury from elevated expression of genes that promote inflammation, cell death, lipid oxidation, and generation of oxygen-derived free radicals (Brooks and Renaudineau, 2015; Li et al., 2014; Stamova et al., 2012). Together, these findings may provide an additional mechanism that explains how women are protected by estrogens early in life, but become susceptible to CVD, myocardial infarction, and stroke after menopause, resulting from the detrimental effects of the genes encoded on the X chromosome.

Identifying XX chromosome dosage as a risk factor for increased body fat and possibly atherosclerosis and CVD resulted in discovering candidate genes that may underlie this effect (Wiese et al., 2023). Prominent among these are genes that escape X chromosome inactivation and are thus expressed at higher levels in tissues of XX compared to XY individuals (Tukiainen et al., 2017). Two X chromosome genes, KDM5C and KDM6A, encoding histone demethylase enzymes that regulate chromatin accessibility and gene expression have been implicated in transcription of

genes influencing fat development and accumulation (Link et al., 2020) and sex-specific differences related to myocardial infarction remodeling (Li et al., 2014). Nevertheless, further research is needed to clarify how other X chromosome genes increase susceptibilities to cardiometabolic conditions in women.

The Effects of Estrogens

Estrogens play an important role in mediating vascular protective effects in women by promoting normal metabolic function, lowering systemic inflammation, and maintaining the integrity and function of the vascular endothelium—the inner lining of blood vessels—that helps prevent cardiometabolic conditions (Fried et al., 2015; Karastergiou et al., 2012; Mauvais-Jarvis, 2015; Palmer and Clegg, 2015). The role of estrogens start during early development, when they influence gene expression in the brain with lasting effects on behavior and metabolism, including the control of energy metabolism by the hypothalamus (Gegenhuber et al., 2022; Mauvais-Jarvis, 2014). Estrogens have also been shown in human and animal studies to influence appetite, with 17β-estradiol (E2) serving as the primary effector of estrogen signaling. Ovariectomized rats experiencing a sudden decline in estrogens had an increase in their food intake, while the administration of exogenous E2 normalized their feeding, presumably through its effects on appetite control centers in the hypothalamus (Rivera and Stincic, 2018). Estrogen’s anorexigenic effects have been observed in studies conducted in women, with research showing that food intake fluctuates across the menstrual cycle, decreasing most in the peri-ovulatory reproductive phase that coincides with rising estrogen levels (Buffenstein et al., 1995; Rogan and Black, 2023; Vigil et al., 2022). Further evidence is needed to support estrogen’s appetite suppressive effects and how this relates to changes in food intake behaviors during the menstrual cycle and after menopause.

In addition, estrogens act on adipose tissue to influence fat storage. Premenopausal women generally store a greater proportion of fat in subcutaneous depots, facilitating long-term energy storage and resistance to depletion during pregnancy and lactation. However, during perimenopause and after menopause, the decline in estrogens enable fat stores to switch from subcutaneous to visceral and abdominal locations, promoting abdominal obesity, which leads to higher risks of cholesterol and lipid imbalances, insulin resistance, Type 2 diabetes, and hypertension, all of which are associated with CVD (Lizcano and Guzmán, 2014). Estrogens are also a determinant of glucose homeostasis. Healthy premenopausal women demonstrate greater glucose uptake by skeletal muscle (Nuutila et al., 1995) and are more resistant than men to lipid-induced insulin resistance (Frias et al., 2001). However, insulin resistance increases with reduced or abnormally

high levels of estrogens, indicating an optimal range of levels of estrogens required for maintaining proper insulin sensitivity (Perseghin et al., 2001).

Estrogens also have protective effects on the vascular endothelium by preserving vascular tone and promoting anti-inflammatory endothelial responses. In ischemia-reperfusion experiments that model myocardial and cerebral infarction (Chen and Vunjak-Novakovic, 2018), estrogens have been linked to increases in markers indicative of endothelial viability (Chang et al., 2019; Simoncini et al., 2000; Veenema et al., 2019; Yang et al., 2000). Estrogens also activate signaling pathways that block programmed cell death and hypoxia (Jukes and Osawa, 1997; Lazaro et al., 2023; Wang et al., 2006, 2009). Although the majority of studies highlight the benefits of estrogens, others have noted that abnormal concentrations are detrimental (Smith et al., 2000; Zhan et al., 2008). For example, administering estrogens during the first week after a myocardial infarction in female rats without ovaries increases infarct size (Smith et al., 2000). This is line with studies that have shown physiological levels of estrogens improved cardiac physiology, reduced fibrosis, and increased capillary density, albeit with no significant effect on mortality (Zhan et al., 2008). Collectively, these findings are consistent with the notion that an optimal range of estrogens are needed for endothelial homeostasis and protection against injury.

Other Biological Factors and Mechanisms

Inflammation is also linked to CVD. Estrogens and ERs have cardioprotective effects via regulating multiple inflammatory pathways (Horng et al., 2017; Rosenzweig et al., 2021). In a mouse model of ischemic heart failure, levels of one specific ER decreased significantly during acute myocardial infarction but increased during heart failure (Rosenzweig et al., 2021). Estrogen signaling also has age-dependent anti-inflammatory effects. Women with an average age of 73 who developed heart failure after myocardial infarction had an increase in plasma proteins associated with macrophage-mediated production of reactive oxygen species compared to men, indicating that the protection from macrophage-mediated inflammation was lost after menopause (Deleon-Pennell et al., 2018). In a mouse model of myocardial infarction, male and female mice were found to produce specialized anti-inflammatory mediators, although female mice generated higher levels to coordinate infarct healing after myocardial infarction (Pullen et al., 2020).

Aside from estrogens, other ovarian hormones play important roles in regulating cardiometabolic and vascular function. Low levels of anti-Müllerian hormone, produced by ovarian follicles, and relaxin, secreted by the corpus luteum in the ovaries, have been implicated in maintaining cardiovascular health and are correlated with atherosclerotic plaque

accumulation (Lambrinoudaki et al., 2020; Verdiesen et al., 2022) and inflammation (Aragón-Herrera et al., 2022; Brecht et al., 2011). In addition, higher levels of the gonadotropin, follicular stimulating hormone, that increases during menopause, has been linked to increases in visceral adiposity (Liu et al., 2017), which contributes to elevated risks of CVD and stroke.

Experimental stroke models have been crucial in identifying therapeutic targets to improve drug design. Because of the complex nature of the disease, which involves the circulatory system, immune system, and endocrine secretions, multiple possible cellular and molecular targets exist. Basic science studies have shown that cell death pathways are effective targets (Liu et al., 2009), as are composite structures, such as the blood brain barrier (D’Souza et al., 2021; Okada et al., 2020), and cellular targets such as astrocytes (Sampath et al., 2023b). Stroke patients have severe gastrointestinal consequences in the acute phase, and the gut–brain axis has emerged as a new tractable target for stroke (Benakis et al., 2016; El-Hakim et al., 2021; Mani et al., 2023).

The Maternal in Utero Environment

Early-life programming (ELP) includes the idea of the maternal nutritional environment influencing the long-term health trajectory of individuals (Pannia et al., 2016). Exposure to hormonal, metabolic, nutritional, and other maternal factors during sensitive developmental stages affects the structure and function of cells and organs and influence the “programming” of health outcomes with lasting physiological and behavioral effects (Archives of Disease in Childhood, 1994; Barker, 2004; Eriksson, 2016; Warner and Ozanne, 2010). Animal models of ELP show that a maternal high-fat diet led to obesity in offspring later in life (Férézou-Viala et al., 2007). Clinical studies have also implicated elevated maternal fat stores to higher risks of metabolic conditions in offspring later in life. For example, infants born to mothers who are obese exhibit higher risks of obesity, CVD, and allergic conditions (Fleming et al., 2018; Van De Maele et al., 2018). In addition, infants born to women diagnosed with gestational diabetes have a greater risk of Type 2 diabetes, a well-known risk factor for coronary heart disease (Eriksson, 2016). Many experimental models of developmental programming report a sex difference in blood pressure, with males exhibiting a significant increase in young adulthood (Alexander, 2003; Bourque et al., 2013; Ortiz et al., 2001). Birthweight is a risk factor for hypertension and CVD, but its clinical significance is not yet considered when managing cardiovascular health. Overall, early-life exposure factors during critical developmental windows may affect lifetime risks for chronic conditions (Chan et al., 2020; Jašarević and Bale, 2019; Morgan et al., 2019; Morrison et al., 2020).

Specific Risk Factors in Women

Reproductive Milestones

Women’s reproductive milestones have been substantially studied in clinical, longitudinal, and population-based epidemiological studies that span menarche, the reproductive window, pregnancy, postpartum, and menopause with respect to risk for cardiometabolic conditions (O’Kelly et al., 2022). The key roles sex hormones play in CVD development and progression in women have been uniformly understood among clinicians and well studied in the research literature. These factors have been included in epidemiological studies, such as the Nurses’ Health Study, Coronary Artery Risk Development in Young Adults, and SWAN. Key reproductive milestones are described next.

Age at Menarche

Age of menarche has been associated with metabolic syndrome and CVD (Charalampopoulos et al., 2014; Lee et al., 2019). An analysis of 12 cohort studies found that every 1 year increase in age of menarche reduces all-cause mortality by 3 percent, including mortality from CVD, ischemic heart disease, and stroke (Chen et al., 2019). A U-shaped association exists between age of menarche and stroke, with increased risk at early and late menarche (Mishra et al., 2020). The mechanisms for why early menarche may lead to increased risk of cardiometabolic conditions is unclear, and older studies generally attributed it to lifestyle or genetic factors in young girls, such as increased body mass index (BMI), blood pressure, and cholesterol and decreased physical activity. However, age at menarche should be considered as a potential screening tool for risk of cardiometabolic conditions and adverse events associated with CVD and stroke (Lee et al., 2019).

Reproductive Window and Life-Span

During the reproductive window, fluctuations in the menstrual cycle may affect chronic conditions that lead to cardiometabolic conditions. Having a shorter reproductive life-span (the number of years between menarche and menopause), has been associated with an increased risk of CVD in midlife and specifically for women who experienced early age of menarche (Mishra et al., 2020). Menstrual cycle irregularity across the reproductive life-span is known to be associated with CVD risk (O’Kelly et al., 2022). The number of children birthed has also been linked to CVD and stroke risk (D’Errico et al., 2024e; Oliver-Williams et al., 2019; Vladutiu et al., 2017).

Polycystic ovarian syndrome (PCOS) is a gynecological condition affecting women of reproductive age that has a strong association between sex steroids and insulin resistance; at least half of women with PCOS are insulin resistant despite not being obese or overweight (Dunaif et al., 1989) and at a greater risk of hypertension, abnormal lipid levels, and metabolic syndrome (Wild et al., 2010). Despite a greater prevalence of CVD risk factors in PCOS, research has not identified a higher cardiovascular event risk or mortality rate because of study size limitations.

Pregnancy and complications

Adverse pregnancy outcomes, including hypertensive disorders of pregnancy, gestational diabetes, preterm delivery, and intrauterine growth restriction, and absence of breastfeeding are all associated with increased CVD and stroke risk (Miller et al., 2022; O’Kelly et al., 2022; Staff et al., 2024). Women who develop gestational diabetes are reported to have an increased risk of developing Type 2 diabetes later in life. Gestational diabetes is also a major CVD risk factor and is associated with other CVD risk factors, including hypertension and unfavorable changes in cholesterol and triglyceride levels (Rivero et al., 2008). Studies are needed to determine whether early interventions and CVD risk assessment during the reproductive period may reduce CVD risk at midlife.

Hypertension during pregnancy, which can be preeclampsia, eclampsia, chronic hypertension, or gestational hypertension, is associated with maternal stroke (Braunthal and Brateanu, 2019) and stroke later in life (Theilen et al., 2016). Moreover, women who develop hypertension during pregnancy have a higher mortality risk (Theilen et al., 2016), and an approximately fivefold higher odds for future stroke relative to women without this condition (Crovetto et al., 2013; Leffert et al., 2015). Disorders of pregnancy, including preeclampsia, are also associated with increased odds of stroke through midlife (Miller et al., 2022, 2023). In the Population Assessment of Tobacco and Health study, adverse pregnancy outcomes were an independent predictor of later stroke (Manrique-Acevedo et al., 2020; Miller et al., 2022; Miller et al., 2023).

Among women who have given birth, both ever having breastfed and longer duration of breastfeeding is associated with lower risk of stroke, with some evidence that these associations vary by both age and race and ethnicity (Jacobson et al., 2018; Richardson et al., 2022).

Menopausal transition

Research has shown that changes in cardiometabolic health that lead to metabolic disorders accelerate during the menopause transition and have a

significant effect on CVD risk that are independent of chronological aging (El Khoudary et al., 2020). Menopause triggers a dramatic shift in the risk profile for stroke in women, with the decline of estrogens and other female sex steroids thought to be neuroprotective regarding ischemic stroke.

Midlife and menopausal transition are times when women accumulate comorbidities, and cardiometabolic health often worsens, marked by unfavorable changes in lipid profiles, abdominal fat stores, higher risk of diabetes and metabolic syndrome, and endovascular dysfunction (Leeners et al., 2017; Ley et al., 2017; Moreau and Hildreth, 2014; Subramanya et al., 2018; Sun et al., 2019; Thurston et al., 2018). Although research suggests that these issues result from the physiologic changes occurring during the menopausal transition, rather than chronological aging, more data is needed on reproductive aging and how this affects risk of stroke and CVD.

Early menopause is common in women, and those who have experienced early, natural menopause have higher risks of developing Type 2 diabetes (Brand et al., 2013) and CVD mortality, heart failure, and all-cause mortality (Ley et al., 2017). Surgically induced menopause also increases the risk of Type 2 diabetes according to several cohort studies (Honigberg et al., 2019; Rocca et al., 2017). Removal of both ovaries is a common cause of premature or early menopause, and cohort studies have identified an increased risk of CVD outcomes in women who had both ovaries removed before age 45 compared to women who did not (Honigberg et al., 2019; Parker et al., 2009; Rocca et al., 2017). Another cohort study showed steeper increases in weight and BMI in women who had both ovaries removed before age 46 and did not receive estrogen therapy (Gibson et al., 2013; McCarthy et al., 2013). Both premature and early menopause are associated with increased risk of stroke, though this is primarily associated with ischemic stroke and varies between natural and surgical menopause (Lisabeth et al., 2009a; Welten et al., 2021). As surgical removal of the ovaries and fallopian tubes is increasingly being used to reduce risks of genetic forms of breast and ovarian cancers (Daly et al., 2021), appropriate timing for this procedure in order to maximize benefits while minimizing risks needs further investigation (Bertozzi et al., 2023). Clinical guidance for early menopause is lacking and needed for this population (Mishra et al., 2024a).

Clinical Guidelines for CVD and Stroke in Women

Basic science and epidemiological studies have identified risk factors for CVD, stroke, and cardiometabolic conditions, such as obesity, metabolic syndrome, and Type 2 diabetes. Researchers have incorporated these risk factors, along with lifestyle risk factors for CVD and stroke that have distinct gender differences, into the Predicting Risk of CVD Events (PREVENT) calculator (Larkin, 2024).

Clinical guidelines and recommendations for preventing CVD in women have existed since 1999 and emphasize that risk factor management is different for women and men (Mosca et al., 1999). There has been increased recognition that women should have unique recommendations given the evident sex and gender differences in preventive factors that are often overlooked (Mosca et al., 2011). The American Heart Association (AHA) updated its evidence-based guidelines, shifting from pure clinical research to focusing comprehensively on primary prevention of CVD and stroke in women as observed in clinical practice (Mosca et al., 2011). Emerging nontraditional risk factors for women in the 2011 AHA guidelines included preterm delivery, hypertensive pregnancy disorders, gestational diabetes, breast cancer treatments, autoimmune conditions, and depression (Garcia et al., 2016). The updated guidelines have included the “essential eight” risk factors, adding sleep to the list, but they do not contain guidelines specifically for women (Lloyd-Jones et al., 2022).

Clinical guidelines to prevent stroke in women were established in 2014. They highlight female-specific risk factors, such as pregnancy, preeclampsia, and migraines, and factors with a more pronounced effect in women, such as migraine with aura, atrial fibrillation, diabetes, hypertension. The guidelines also address factors with similar prevalence in men and women, such as physical activity, smoking, metabolic syndrome, and diet (Bushnell et al., 2014), for which the differential effects on male versus female risk are unknown. The 2021 guidelines for the prevention of stroke from the AHA and the American Stroke Association outline the continued importance of managing vascular risk factors, including diabetes, smoking, hypertension, and atrial fibrillation, and adopting healthy lifestyle behaviors known to lower risk of stroke, such as a low-salt/Mediterranean diet, exercise, avoidance of prolonged sedentary behavior, and adherence to medication (Kleindorfer et al., 2021). New to these guidelines is the recommendation for patients to receive a “diagnostic workup” after stroke, to identify the etiological cause, and appropriate post-stroke treatment to lower the risk of subsequent stroke (Kleindorfer et al., 2021). Sex and gender considerations were also highlighted; alcohol intake of greater than one drink per day in women and two drinks per day in men should be lowered or altogether prohibited to reduce stroke risk; women with history of gestational diabetes should have their blood glucose levels monitored to lower risk for ischemic stroke; and women with migraine with aura who use estrogen-containing oral contraceptives, particularly smokers, should consider avoiding them and if they have had a previous stroke. Special approaches that address the underlying determinants of inequities in stroke risk, particularly in women, people of color, immigrants, rural populations, older adults, and sexual minority groups, was also emphasized to reduce risk (Kleindorfer et al., 2021).

Other Risk Factors

CVD and stroke risk factors that are disproportionately more severe in women include hypertension, atrial fibrillation, and diabetes (Rexrode et al., 2022). Women have a higher risk of stroke at a lower threshold for systolic blood pressure than men (Howard et al., 2019; Ji et al., 2021; Peters et al., 2020). Atrial fibrillation is the most common type of arrhythmia, and women with it are more likely to have more severe symptoms (Ball et al., 2013), with a greater associated risk of stroke and death (Bushnell et al., 2014; Emdin et al., 2016). Diabetes is a major risk factor for stroke and may be more significant in women (Peters et al., 2014). Women aged 55–64 and who have Type 2 diabetes are significantly more likely to have a stroke than are men (1.5–6-fold vs. 1.5–2-fold) (Almdal et al., 2004). Research is needed to understand the biological mechanisms of why diabetes presents a higher risk of stroke in women and whether this is linked to menopause, as mentioned (Madsen et al., 2021; Peters et al., 2020).

Condition Presentation and Diagnosis

CVD

Emerging evidence from the last 2 decades supports distinct differences in the clinical presentation of CVD in women (Vogel et al., 2021). Compared to men, who typically exhibit features of obstructive coronary artery disease,⁴ women are more likely to experience atypical symptoms, such as epigastric symptoms, shoulder, jaw and/or neck pain, and palpitations (Lichtman et al., 2018). The cardioprotective effects of estrogens most likely explain the lower prevalence of obstructive coronary artery disease in premenopausal women (Mehta et al., 2016). In women with it, atherosclerotic lesions may be smaller as a result of positive remodeling of coronary arteries, which obscures atherosclerosis (Bairey Merz et al., 2006; Pacheco Claudio et al., 2018). In the Women’s Ischemia Syndrome Evaluation (WISE) study, 79 percent of women with nonobstructive coronary artery disease showed signs of coronary atherosclerosis with positive remodeling (Khuddus et al., 2010). Distinct pathophysiological mechanisms govern atherosclerosis in men and women, with women more often developing features of plaque erosion and men more commonly having disruptions of atherosclerotic plaque (Pacheco et al., 2022b; Pacheco Claudio et al., 2018; Pasupathy et al., 2015).

Although similar proportions of women and men aged 18–55 present with angina, or chest pain, upon diagnosis of acute myocardial infarction

___________________

⁴ Obstructive coronary artery disease is defined as a greater than 50 percent blockage of the coronary arteries.

(Lichtman et al., 2018), women may present additional symptoms and early signs that appear to be distinct. These include angina with absence of coronary artery obstruction; discomfort or pain in the jaw, neck, arms, shoulder or back; weakness; fatigue; difficulty breathing; nausea; indigestion; anxiety; arm weakness or heaviness; cold sweats; and sleep disturbances (Brush et al., 2020; Khan et al., 2017; Pacheco et al., 2022b). Although these differences in clinical presentation have been reported in women for many decades, clinicians still misdiagnose CVD or dismiss symptoms in women (Gendered Innovations, 2023), and only in recent years has recognition increased of cardiovascular syndromes that preferentially affect women (Tamis-Holland et al., 2019). The following section provides an overview of recent findings related to three cardiac conditions primarily observed in women—ischemia with no obstructive coronary artery disease (INOCA), myocardial infarction with no obstructive coronary artery disease (MINOCA), and spontaneous coronary artery dissection (SCAD)—and stroke.

INOCA

Nearly half of all women experiencing angina and undergoing invasive coronary angiography with symptoms and/or signs of ischemia have INOCA (Ford and Berry, 2019), with around two-thirds of angiograms from women with suspected ischemia lacking signs of obstructive coronary artery disease (Merz et al., 2017). INOCA is more prevalent in women, particularly during midlife (Merz et al., 2017). Estimates of INOCA derived from clinical studies vary depending on the study population, inclusion criteria, and diagnostic method. For example, 62 percent versus 34.4 percent of women enrolled in the WISE study and the International Study of Comparative Health Effectiveness with Medical and Invasive Approaches trial, respectively, had no obstructive coronary artery disease (Reynolds et al., 2020; Shaw et al., 2006).

INOCA is not a benign condition, as it is associated with an increase in major adverse cardiovascular outcomes and a significant contributor to CVD mortality in women (Pacheco Claudio et al., 2018). The WISE study identified an elevated risk in all-cause mortality in women with INOCA compared to age-matched control subjects (Kenkre et al., 2017). Considering the morbidity and mortality risks, it is important for both patients and clinicians to be aware of the clinical significance of unconventional CVD symptoms in women.

Angina is a common symptom of INOCA that may be missed by anatomical coronary artery tests alone (Ford and Berry, 2019). Typically, invasive coronary angiography is the standard tool for identifying obstructive coronary artery disease, and its absence in patients presenting with angina may indicate INOCA (Ford and Berry, 2019). It is unclear whether INOCA is a precursor to angiographic stenosis or unrelated (Barsky et al., 2020).

Approximately half of women with INOCA also demonstrate coronary microvascular dysfunction, an abnormal constriction of the heart’s small blood vessels, detected by invasive coronary reactivity testing (Pacheco Claudio et al., 2018). To diagnose this, clinicians typically employ invasive intracoronary measurements or noninvasive positron emission tomography (PET) to assess abnormal coronary flow reserve or coronary blood flow (AlBadri et al., 2019; Merz et al., 2017). Most women with INOCA exhibit diffuse nonobstructive coronary atherosclerosis, likely caused by inflammatory processes (Merz et al., 2017).

The etiological mechanisms of INOCA are numerous and include vascular endothelial dysfunction, inflammation, smooth muscle cell dysfunction, remodeling of small blood vessels, and activation of the sympathetic nervous system. Other symptoms indicating decreased oxygen supply have been observed with INOCA, such as aortic stenosis, severe hypertension, and anemia (Pacheco Claudio et al., 2018). Physicians have used beta-blockers, calcium channel blockers, nitrates, and ranolazine to alleviate INOCA symptoms (Kunadian et al., 2020).

Despite a high prevalence of INOCA in women, there is a lack of clarity, standardization, and uniformity in diagnosis and management. Consequently, women are often discharged with the inaccurate diagnosis of non-cardiac chest pain and repeatedly hospitalized (Ford and Berry, 2019; Pacheco et al., 2022a). No standard definition or a set of unifying consensus guidelines exist regarding diagnostic evaluation and treatment strategies. In a published survey, most cardiologists believed that additional clinical guidance is needed to accurately diagnose and treat patients with INOCA, with over half of them wanting U.S.-specific guidelines or expert consensus for diagnosis and management (Pacheco et al., 2022a). As a result of the knowledge gaps, approximately one in three cardiologists in early to mid-career report being neutral or lacking confidence in their ability to manage patients with INOCA (Luu et al., 2022).

MINOCA

MINOCA is defined as a myocardial infarction attributed to atherosclerotic or non-atherosclerotic causes, with presentation of ischemic symptoms and lesions that occupy less than 50 percent of the epicardial coronary arterial space (Herling de Oliveira et al., 2022; Tamis-Holland et al., 2019). The mortality is approximately 4.7 percent per year (Pacheco Claudio et al., 2018; Pasupathy et al., 2015). MINOCA accounts for between 6–15 percent of all heart attacks (Reynolds et al., 2021), and occurs in 10.5 percent of women compared to 3.4 percent in men (Smilowitz et al., 2017). MINOCA also affects patients at younger ages and who identify as African American or Black (Pacheco Claudio et al., 2018; Pasupathy et al., 2015). A study

using coronary optical coherence tomography and cardiac MRI has shown that MINOCA was the underlying cause in 56 percent of women diagnosed with a heart attack. Within this group, ischemia was more than three times more common than non-ischemia (Reynolds et al., 2021). As several mechanisms may lead to MINOCA, including plaque disruption, coronary vasospasm, coronary microvascular dysfunction, SCAD, clotting disorders, blood clots, cardiomyopathies and myocarditis (Hayes et al., 2018; Reynolds et al., 2021; Tamis-Holland et al., 2019), it serves as a working diagnosis with many possible causes that requires further evaluation of symptoms to administer appropriate therapies (Pasupathy et al., 2015).

SCAD

SCAD is another underdiagnosed and poorly recognized acute coronary syndrome that primarily affects women (Méndez-Eirín et al., 2021). SCAD-triggered myocardial infarction is spontaneous and nonatherosclerotic and is caused by lack of myocardial blood flow resulting from partial or complete occlusion of the coronary artery following a vessel-wall tear or hematoma (Tweet et al., 2018). It occurs frequently in midlife women, with the average age ranging from 45 to 53 years old (Hayes et al., 2018). SCAD shares symptoms with other acute coronary syndrome conditions, and risks include myocardial ischemia, arrhythmias, and sudden cardiac death (Saw et al., 2014a). It accounts for approximately 35 percent of acute myocardial infarctions in women under 50 and is the main cause in women under 40 (Tweet et al., 2018). SCAD is also the most common cause of myocardial infarction associated with pregnancy (Tweet et al., 2017), which is more serious than SCAD in non-pregnant women and can lead to impaired heart function (Tweet et al., 2017).

Diagnosing SCAD continues to be challenging because of limited research on its causes and natural history. No biomarker reliably differentiates SCAD from atherosclerosis (Adlam et al., 2021). Noninvasive imaging, such as computed tomographic coronary angiography, is not recommended routinely because of its limitation in assessing regions of the coronary arteries that SCAD frequently affects, so clinicians use invasive coronary angiography (Adlam et al., 2021). In some confusing circumstances, non-SCAD presentations mimic the angiographic features usually associated with SCAD, which may lead to an inaccurate diagnosis (Adlam et al., 2021).

Stroke

Sex differences in symptom presentation of acute stroke or transient ischemic attack (also known as “minor stroke”) are not well understood

and may underlie issues of misdiagnosis or missed diagnosis in women (Ali et al., 2022). Often, clinical symptoms of stroke presented by women may be mistaken for other conditions, like a migraine attack (Ali et al., 2022). Two recent meta-analyses indicate that, compared to men, women more frequently present with non-localized symptoms, including changes in consciousness or mental status, headache including migraine, and generalized weakness (Ali et al., 2022; Shajahan et al., 2023). However, these two studies differed in their conclusions related to sex differences in focal symptoms. Based on an analysis of 60 studies, the first review indicated that dizziness and speech slurring (with or without nausea/vomiting) occurred more commonly in women, while muscle weakness and visual aberrations were more often experienced by men (Ali et al., 2022). Another review based on 21 studies showed that men more commonly presented with speech slurring and had higher risk of dizziness (Shajahan et al., 2023). These discrepancies and the general lack of high-quality studies warrant further research to distinguish differences in stroke presentation between men and women (Ali et al., 2022), which is a crucial factor in the early recognition and management (Shajahan et al., 2023).

Women are also more likely to arrive at the hospital later than men, which likely excludes more women from clot-busting treatment with tissue plasminogen activator (tPA) (Madsen et al., 2016), and less likely to be conscious upon arrival, which suggests they are experiencing larger strokes and a longer duration of symptoms (Glader et al., 2003). Moreover, the greater prevalence of poorly understood symptoms in women may lead to diagnostic delays and missed treatment opportunities (Lisabeth et al., 2009b).

Treatment and Management

INOCA AND MINOCA

Studies have shown that specialized women’s heart centers improve outcomes in women with INOCA and MINOCA. A prospective study showed that evaluating women at these centers led to more specific diagnoses of INOCA and MINOCA that resulted in therapeutic changes that improved control of CVD risk factors, including blood pressure, total cholesterol, and low-density lipoprotein (LDL) cholesterol; decreased chest pain; and elevated the quality of life after 1 year (Parvand et al., 2022). Specialized women’s heart centers are an important resource for patients and clinicians, providing diagnostic tools, risk stratification, and management relevant to women’s heart health, as well as promoting awareness, education, and multidisciplinary collaboration (Parvand et al., 2022). They may serve as a long-term, sustainable model that can address disparities in cardiovascular risk factors and CVD in women (Pacheco et al., 2022a).

SCAD

The decision to proceed with percutaneous coronary intervention or coronary artery bypass surgery differs widely among clinicians and depends on the acute presentation (Saw et al., 2014a). A conservative approach is typically recommended given no ongoing ischemia, recurrent chest pains, hemodynamic instability, or left main arterial dissection (Saw et al., 2014a). However, this is not based on high-quality randomized data, and the acute and long-term outcomes of a conservative or revascularization approach remain unclear (Saw et al., 2014a).

Pharmacological Treatment for CVD, Stroke, and Risk Factors

Access to treatment and prescribing practices for cardiometabolic conditions differ by gender. Anti-coagulation medication, such as warfarin, is critical for preventing strokes, but clinicians are less likely to prescribe it for women despite showing equal effectiveness in women and rates of similar risk of intracranial hemorrhage between men and women (Rexrode et al., 2022). New oral anticoagulants (e.g., dabigatran) appear to be more efficacious than warfarin, but women are more likely to fill prescriptions for lower doses compared to men (Tsadok et al., 2015). Women are less likely to be prescribed beta-blockers or receive an implantable cardioverter-defibrillator (Butzner et al., 2022). In addition, research shows that women are less consistently prescribed statins, angiotensin-converting enzyme inhibitors, and beta-blockers for managing Type 2 diabetes and the concomitant risk of CVD (Vimalananda et al., 2013; Zhao et al., 2020). Many women with history of Type 2 diabetes and moderate to high CVD risk meet the criteria to initiate statin therapy according to AHA/American College of Cardiology guidelines (Stone et al., 2014). However, clinicians may address CVD risk factors less aggressively in women than men with Type 2 diabetes (Regensteiner et al., 2015). A retrospective cohort study of diabetic women veterans showed that they had lower rates of achieving a target low density lipo-protein (LDL) level than men (Vimalananda et al., 2013). This may be because of no prospective studies in women regarding primary CVD management with statins and no recommendation based on expert opinion for this as a primary prevention approach.

One FDA-approved drug for stroke exists, the clot-busting agent tissue plasminogen activator (tPA). The standard treatment regimen for acute ischemic stroke is to administer tPA within 4.5 hours from symptom onset, which consequently restricts the number of patients eligible to receive treatment (de Los Ríos la Rosa et al., 2012; Schwamm et al., 2013). However, recent studies have shown that the use of imaging tools, such as computed tomographic perfusion or MRI, may give better indications of stroke lesion

progression and could thus be used to identify patients who may benefit from tPA treatment beyond the 4.5-hour window (Broocks et al., 2023; Ma et al., 2019). Further clinical trials are needed to strengthen the evidence base that imaging modalities can help determine if tPA treatment can be safely and effectively given to stroke patients even after 4.5 hours from symptom onset (Ma et al., 2019).

Generally, women are less likely to receive tPA at primary and non-primary stroke centers regardless of stroke presentation and after adjusting for age, socioeconomic class, location, and hospital (Boehme et al., 2017; Kent et al., 2005; Reeves et al., 2009a, 2009b; Reid et al., 2008; Strong et al., 2020). A recent meta-analysis of 24 studies showed that women had 13 percent lower odds of being treated with tPA compared to men; however, disparities in tPA treatment were not consistently observed across different regions (Rexrode et al., 2022). Data from the Get with The Guidelines Stroke registry⁵, an online, interactive assessment and reporting system, showed that women who arrive at the hospital within 2 hours of symptom onset of acute ischemic stroke were less likely to receive tPA than men (55.8 vs 59.4 percent) (Reeves et al., 2009b). A later retrospective analysis of Get with The Guidelines Stroke data captured between 2003 and 2011 showed that approximately 25 percent of patients presenting with acute ischemic stroke symptoms within 2 hours of onset failed to receive tPA treatment. Based on their findings, researchers suggested that women, older adults, and racial minorities are undertreated despite being eligible for tPA treatment (Messé et al., 2016).

Experimental stroke models have also successfully modeled long-term neuropsychiatric conditions commonly seen in survivors. Poststroke depression is more common in women (Lyu et al., 2022; Poynter et al., 2009), and conventional antidepressants are not effective (Frank et al., 2022; Sun et al., 2017). More than half of stroke patients experience some level of cognitive impairment by 6 months afterward, and ischemic stroke is a leading cause of vascular dementia (Mellon et al., 2015; Vijayan and Reddy, 2016) and AD (Elman-Shina and Efrati, 2022; Filippenkov et al., 2023). Poststroke cognitive impairment is more likely to affect women than men (Pendlebury and Rothwell, 2009), and women are more likely to require care in an assisted living facility because of deteriorating health conditions that include cognitive impairment, frailty, pain, and depression (Yu et al., 2020). Vascular dementia is also worse in women compared to men (Gannon et al., 2022). Experimental stroke models have replicated post-stroke cognitive

___________________

⁵ Additional information available at https://www.heart.org/en/professional/quality-improvement/get-with-the-guidelines/get-with-the-guidelines-stroke/get-with-the-guidelines-stroke-registry-tool

impairment in males and females (Mani et al., 2023; Sampath et al., 2023a; Ward et al., 2018) and aging females (Panta et al., 2020).

Disparities

Research on racial and ethnic disparities regarding factors associated with cardiometabolic conditions is increasing. For example, investigators have examined whether lifestyle modifications and interventions can decrease the risk of morbidity in minoritized populations. In addition to accounting for BMI and central fat stores, researchers have studied how psychosocial and lifestyle factors affect disparities in the incidence of conditions such as diabetes, CVD, and stroke.

Data from the NHIS show that Black women experience higher rates of all the major risk factors associated with cardiometabolic conditions, including diabetes, hypertension, and obesity (Chinn et al., 2021). Compared with White women, Black women had lower incomes and lower rates of education, two social factors that significantly affect stroke risk and attenuated their higher risk (Jiménez et al., 2019). In the Reasons for Geographic and Racial Differences in Stroke study, achieving a higher ideal cardiovascular health score that considers lifestyle behaviors and modifications decreased the risk of diabetes among adults with normal glucose levels, with a stronger association among Black than White adults (Joseph et al., 2019). In a prospective cohort study, the cumulative incidence was greater for Black than for White men and women (Carson et al., 2021). After adjusting for age and prediabetes, this same study found that diabetes affected Black women more than White women (Carson et al., 2021); individual risk factors, such as BMI and waist circumference attenuated the racial disparity (Carson et al., 2021).

As noted, effective treatment of stroke is time sensitive. Studies have identified inequities in administering clot-busting treatments, indicating that Black individuals have less access to key treatments in the acute time window (Otite et al., 2021). Furthermore, data indicate that Black women are disproportionally affected by treatment inequities (Boehme et al., 2014).

CVD significantly affects Hispanic/Latina women. The Hispanic Community Health Study/Study of Latinos found that Hispanic/Latina women face higher than average rates of obesity, elevated cholesterol levels, hypertension, and disorders during pregnancy, such as preeclampsia and gestational diabetes (Gomez et al., 2022; Rodriguez et al., 2021). In addition, these rates differed by background, with higher rates among Cuban, Dominican, and Puerto Rican compared to Mexican American women. Hispanic/Latina women also experience higher rates of heart failure than non-Hispanic White women (Tsao et al., 2023). They were less likely to be aware of having these conditions and to have them under control,

which may result from cultural, socioeconomic, geographic, and linguistic barriers that affect access to care (Gomez et al., 2022).

Studies of risk factors for CVD among Hispanic/Latina women found that workplace exposure to environmental toxins, such as solvent, pesticide, and metal, was associated with greater levels of CVD (Bulka et al., 2019; Gomez et al., 2022). Research has also shown that immigration-related factors, such as level of acculturation, are associated with access to care and engaging in health-promoting behaviors, with Hispanic/Latina women engaging in lower levels of physical activity (Divney et al., 2019).

Despite established clinical guidelines for treating CVD, these have not been validated in Hispanic/Latina women. The AHA/American College of Cardiology Pooled Cohort Equations risk calculator has only been validated in non-Hispanic White and non-Hispanic African American men. These tools may lead to inaccurate and unreliable risk calculations in Hispanic populations (Gomez et al., 2022). Interventions for preventing CVD and addressing associated risk factors that target both youth and adult Hispanic/Latina women are needed, given that obesity rates are high among youth and adults (L’Engle et al., 2023).

Based on limited studies, American Indian and Alaska Native men exhibit higher morbidity and mortality from chronic conditions such as CVD and obesity, and more risk factors, such as smoking, whereas American Indian and Alaska Native women reported higher rates of diabetes (Breathett et al., 2020). To describe the various factors that affect the development of chronic conditions in American Indian and Native Alaska populations, researchers have used the Framework of Historical Oppression, Resilience, and Transcendence (Burnette and Figley, 2016; Burnette et al., 2019; McKinley et al., 2019). One study examining sex differences in various behavioral, physical, and psychological outcomes associated with CVD focused on how factors associated with oppression and trauma affect CVD risk for American Indian and Alaska Native women (Burnette et al., 2020), but more research is needed to understand gender differences in risk factors.

Data disaggregation has remained an issue in identifying cardiometabolic disease disparities among groups of Asian American and Pacific Islander women in risk factors for CVD and stroke (Đoàn et al., 2021). Asian American women have reported lower prevalence of diabetes and being overweight and higher rates of hypertension than men (Đoàn et al., 2021). Among groups of Asian American women, Asian Indian women reported the highest prevalence of being overweight or obese, and Filipino women had the highest prevalence of hypertension (Đoàn et al., 2021). One study found that Asian Indian, Chinese, Filipino, and Korean women had lower levels of high-density lipoprotein and higher triglyceride levels compared to non-Hispanic White individuals (Kwan et al., 2023).

Native Hawaiian and Pacific Islander women report greater prevalence of obesity and hypertension than Native Hawaiian and Pacific Islander men (Đoàn et al., 2021). Studies have found that mortality from CVD and ischemic stroke are higher in Asian American women and men compared to non-Hispanic White individuals (Đoàn et al., 2021). Data from the National Center for Health Statistics have shown that the standardized mortality rate for ischemic heart disease was higher in Asian Indian women, whereas the highest cerebrovascular disease was highest among Asian Vietnamese women compared to other Asian groups and non-Hispanic White groups of men and women (Shah et al., 2022).

In Asian Americans, the level of acculturation is a key factor researchers have studied, but few studies have disaggregated the data by sex. In addition, findings have been mixed on whether living in the U.S. for longer periods was associated with cardiometabolic conditions (Min et al., 2022). Cohort studies in the Asian American community that examined lifestyle factors include Mediators of Atherosclerosis in South Asians Living in America (MASALA) and Multi-Ethnic Study of Atherosclerosis, which only included Chinese Americans (Kanaya et al., 2014; Rodriguez et al., 2020). A study using data from MASALA and focusing predominantly on immigrant women found that self-reported menopausal status was associated with BMI and subcutaneous fat area and increased odds of hypertension (Vijay et al., 2022).

Research Gaps

The prevalence of obesity and metabolic syndrome is increasing in women, with distinct metabolic features compared to similar conditions in men. Understanding the molecular basis for the female-biased propensity to store fat is critical for both prevention and effective use of treatments to combat obesity. Preclinical models will play an instrumental role in achieving this goal. New insights will come from single-cell and spatial transcriptomic approaches to characterize sex differences in adipose tissue structure and function. Recent data on the role of certain X escape genes highlight the important role of epigenetic regulation as a determinant of sex differences and warrant further investigation.

Efforts to determine how gonadal hormones influence fat stores will benefit from studying gonadal hormone ablation in preclinical models throughout the lifespan, particularly in mice at ages similar to the age of menopause in women. Human studies of new diabetes and obesity drugs, such as Ozempic and Wegovy, and sodium glucose cotransporter 2 inhibitors (e.g., Jardiance) should analyze the effects on weight loss and long-term outcomes after stratification by sex/gender, which researchers often fail to do.

Tools to enable early and accurate diagnosis of CVD in women are needed. Myocardial infarction, stroke, and heart failure affect women,

particularly women of color, at a younger age, causing death and disability earlier in the lifetime, and it is important to diagnose CVD in the populations experiencing disparities in diagnosis (Ivey et al., 2022). Knowledge is lacking that would enable accurate diagnosis of the pathophysiological mechanisms of heart disease in women. Improving risk stratification and treatment for women will require differentiating their presentations of coronary artery disease (Pacheco et al., 2022b).

INOCA also lacks standardized diagnostic criteria, with no consensus on how to diagnose INOCA and MINOCA. Establishing diagnostic codes for both would improve treatment management, enable research, and inform policies affecting medical practice (Pacheco et al., 2022a). More research is needed to clarify the mechanisms leading to INOCA and MINOCA in women (Vogel et al., 2021), and more specialized women’s heart centers should be established, as it has been shown that they improve outcomes for both conditions (Parvand et al., 2022).

A knowledge gap continues to exist for clinicians managing SCAD, with a lack of consensus guidelines on evaluation and treatment. In addition, outcomes from conservative therapy and the extent of patient revascularization have been understudied (Saw et al., 2014b).

Disaggregated data are lacking regarding Hispanic, Latina/Latino, and Asian American, Native Hawaiian, and Pacific Islander people, which can mask differences in certain groups that may be more affected by CVD. Furthermore, national data on these populations underreport CVD prevalence and mortality and related deaths as a result of small sample sizes and issues with racial misclassification (Mehta et al., 2023). Conditions such as atrial fibrillation and ventricular arrhythmia are understudied in Hispanic women and Latinas, and more studies are needed to understand these conditions in women and validate diagnostic criteria in these populations.

MULTIPLE SCLEROSIS

“. . . I realize, of course, when I get up, the legs aren’t working quite the way they did . . . Sometimes I use my walker to get to the bathroom or my cane to get around, but then after that reality check, I still try to figure out every single day what it is that I’m able to do and then to try to live my best life around that.”

—Presenter at Committee Open Session

MS is an autoimmune disease⁶ that causes brain and spinal cord demyelination, axonal damage, and neuronal loss (Gold and Voskuhl, 2016).

___________________

⁶ Systemic lupus erythematosus (SLE) is not discussed in this chapter. The report, Enhancing NIH Research on Autoimmune Disease provides information on other autoimmune diseases including those that predominately impact or affect women differently and have multi-systemic involvement (NASEM, 2022).

Depending on the brain regions affected, diverse clinical symptoms occur, including motor and sensory deficits, urinary incontinence, and cognitive impairment (Voskuhl, 2020). MS is approximately three times more common in women than men, but disease progression appears to be faster in men (Dunn et al., 2015; Voskuhl, 2020). Women with MS experience 70 percent lower rates of relapse of symptoms during later stages of pregnancy, suggesting that pregnancy hormones may alleviate the disease (Gold and Voskuhl, 2016). Both the lower incidence in men and the alleviation during pregnancy represent larger protective effects for MS than are achieved by any conventional therapy. Thus, the clinical observations provide strong rationale for investigating how sex-biasing factors increase or decrease disease incidence and progression and suggest that the effects of sex-related variables are complex, providing relative protection to one sex or the other by acting on different disease mechanisms or phases.

The primary view of MS is that myelin-activated T helper cells expand in the periphery, enter the brain, and elicit the secretion of proinflammatory cytokines. In turn, other immune system cells, such as macrophages, enter the CNS and contribute to demyelination and neuronal injury (Dunn et al., 2015; Petermann and Korn, 2011; Prat and Martin, 2002). In the brain, response to an autoimmune attack involves microglia, oligodendrocytes, astrocytes, and neurons. Evidence suggests that several sex-biasing factors may influence MS via actions on diverse cell types, each of which can protect from or exacerbate disease, affecting incidence and progression independently (Voskuhl, 2020).

The dominant animal model of MS is the experimental autoimmune encephalomyelitis (EAE) model in mice (Dedoni et al., 2023; Voskuhl, 2020). A mouse is injected with a mixture of myelin protein antigens, along with an adjuvant and pertussis toxin, which together activate immune cells and break down the blood–brain barrier. The autoimmune attack on the brain and spinal cord has many features of the MS attack in humans. In specific strains of mice, females show greater susceptibility, and thus can be studied to isolate sex-specific factors that influence EAE outcomes and determine their cellular and molecular mechanisms of action.

Biological Factors

Sex chromosome effects

To investigate a direct role for sex chromosomes on MS, EAE was studied using the FCG model, allowing the comparison of mice with different sex chromosomes (XX vs. XY) but the same type of gonadal hormones (Smith-Bouvier et al., 2008). The mice had their gonads removed to eliminate the possibility of differences in levels of gonadal hormones when EAE was induced. XX mice showed worse clinical disease and neuropathology,

independent of their gonadal sex at birth. XY mice showed increased secretion of Th2 cytokines associated with protection. This sex chromosome effect may be caused in part by a specific X-linked gene, Kdm6a, that escapes X inactivation and is therefore expressed higher in many cell types, including T cells (Itoh et al., 2019). Conditional knock out of Kdm6a in CD4+ T lymphocytes reduces clinical disease and neuropathology in the EAE model. Another X gene escaping X inactivation, Kdm5c, a H3K4me3 histone demethylase, is also implicated in EAE (Doss et al., 2021). EAE can be induced by transferring T helper 17 (Th17) lymphocytes to recipient mice, after the Th17 cells were stimulated by myelin antigens. The resulting EAE is worse if the transferred Th17 cells are XY than if they are XX. Increasing expression of Kdm5c in XY Th17 cells reduces their pathogenicity, but inhibiting Kdm5c in XX Th17 cells increases EAE. Because the two studies implicating opposing actions of the two genes (Doss et al., 2021; Itoh et al., 2019) were performed in different strains and EAE model systems, it is not clear if both occur at the same time, they differentially regulate promoter or suppressor regions, or one relates to worse incidence in females and the other to worse disease progression in males; further work is needed.

The Y chromosome contributes to sex chromosome effects in EAE (Case and Teuscher, 2015; Case et al., 2013). EAE was induced in multiple strains of male mice that differed in the strain background of their Y chromosome but were otherwise genetically identical in their autosomes and X chromosome. Specific types of Y chromosomes had varying effects on severity of EAE. Further work is needed to discover the Y genetic elements that account for these differences and determine if the Y gene effects synergize with or counteract effects of X chromosome genes impacting EAE (Arnold, 2017).

The effect of sex chromosome complement on EAE differs depending on whether the cells affected are in the immune system or brain. Bone marrow chimeras were used to create mice with XX or XY cells in the immune system, in mice composed of XX or XY cells in the brain and rest of the body (Du et al., 2014). When the sex chromosomes in the brain were XY, the clinical response to EAE was worse, independent of the type of sex chromosomes in the immune system. Thus, the direction of the sex chromosome effect is opposite in the brain (XY worse than XX) than in the immune system (XX worse than XY). These results may help explain why progression of EAE is worse in men than women, even when incidence of MS is worse in women than men.

In studies of female bias in autoimmune diseases, increasing attention is given to XIST, the X chromosome gene that is expressed as a noncoding RNA and catalyzes the inactivation of one X chromosome in each somatic XX cell. This gene regulates the expression of a majority of X genes, many of

which regulate immune function. This gene inactivation by XIST is atypical and aberrant in some types of immune system cells, suggesting that greater incidence of autoimmune diseases in females is caused in part by reactivation and aberrant expression of some normally inactivated X genes in XX cells (Forsyth et al., 2024). Another hypothesis is that the XIST RNA forms ribonucleoprotein complexes with its interacting partners in XX cells, which act as an antigen that is attacked by the immune system in autoimmune diseases. Again, the link to this gene may help explain the female predominance of autoimmune diseases. This idea has received recent support from experiments in mice in which this gene was artificially expressed in XY male mice, increasing the amount of disease in a model of SLE (Dou et al., 2024). Further research is needed to understand the role of XIST.

The Effects of Estrogens

The three types of gonadal hormones that might influence MS are estrogens, progestins and androgens. Treatment of either males or females with the androgens testosterone or dihydrotestosterone reduces disease symptoms (Palaszynski et al., 2004). Thus, androgens may be protective, an idea supported by limited analysis of the immune- and neuroprotective effects of androgens in men in a small clinical trial (Voskuhl et al., 2020). Further work is needed to understand how androgen therapy might alleviate MS symptoms in men, including studies of the affected downstream molecular pathways that may be targeted in the clinic in men and women.

The effects of ovarian hormones are generally protective and thus unlikely to explain the greater incidence of MS in women but may explain the alleviation of MS in late pregnancy. In the third trimester, estrogens and progesterone are elevated (Voskuhl, 2020). Women report worse symptoms when levels of estrogen are low before the menstrual period and after menopause. However, removal of the ovaries has an inconsistent effect on the incidence of MS. In EAE mice, treating females with estrogens reduces clinical symptoms, but progesterone has little effect (Voskuhl and Palaszynski, 2001). Selective knockout of ER alpha (α) or beta (β) in mice indicate that ERα expression in astrocytes but not neurons is required for the neuroprotective and anti-inflammatory effects of estrogens, but that estrogens’ effects do not require ERβ expression in astrocytes or neurons (Spence et al., 2013). However, ERβ ligands can offer protection in EAE mice via actions on other cellular targets (Wisdom et al., 2013), which requires ER expression in oligodendrocytes (Kim et al., 2018). Two Phase 2 clinical trials support a protective effect of estrogens in women with MS. In one, oral contraceptives containing estradiol, taken with interferon beta, improved cognition (De Giglio et al., 2017). The other found that an estrogen plus glatiramer acetate (Copaxone^®) improved cognitive

processing speed, spared atrophy in cortical regions involved with problem solving and attention, and decreased serum neurofilament light chain levels (MacKenzie-Graham et al., 2018; Voskuhl et al., 2016, 2020, 2022).

Other Biological Factors and Mechanisms

GWAS have provided additional evidence for the role of the immune system in the pathogenesis of MS. A recent study identified more than 200 genetic variants associated with susceptibility to MS, with a large majority related to the immune response (Patsopoulos et al., 2019). In a systematic review of genome-wide transcriptome studies of MS, differential gene expression was analyzed in males and females using Gene Expression Omnibus and ArrayExpress databases that included patient data. Analysis of gene expression in the brain indicated sex-specific differences in immune patterns: females displayed differences in proinflammatory and innate immune responses related to myeloid lineage and males had changes in adaptive responses associated with the lymphocyte lineage. Other notable results were cellular metabolic alterations in females and differences in the stress response to metal ions in males (Català-Senent et al., 2023).

Specific Risk Factors in Women

Sex differences in the prevalence and severity of MS are influenced by the interplay of environmental factors with inherent biological factors. Environmental risk factors implicated in MS are Epstein-Barr viral infection, cigarette smoke, lack of sun exposure, low vitamin D levels, and obesity (Angeloni et al., 2021; Hedström et al., 2021b). A longitudinal analysis in a cohort of military service members indicated that Epstein-Barr viral infection (but not other viral infections) increased the risk of MS by 32-fold (Bjornevik et al., 2022). Environmental exposures to Epstein-Barr virus, obesity, and smoking may interact with a human leukocyte antigen risk allele, further elevating MS risk (Hedström et al., 2021a, 2021b; Olsson et al., 2017). Certain environmental risk factors that mainly affect women, including occupation, smoking, obesity, birth control, and childbirth, may influence disease onset and progression (Filippi et al., 2018).

Condition Presentation and Diagnosis

Most patients present with relapse-remitting MS (RRMS), which is based on at least two occurrences of CNS dysfunction with partial or complete resolution of symptoms. This diagnosis can also include MRI assessment of demyelinating CNS lesions. Patients with RRMS may eventually develop secondary progressive MS (SPMS), a condition marked by permanent neurological impairments with worsening disability.

A small proportion of MS patients will already present with progressive disease, known as primary progressive MS (PPMS), characterized by progressive CNS dysfunction for a minimum of 1 year and demyelinating lesions or oligoclonal bands in cerebrospinal fluid (Filippi et al., 2018). Women are disproportionately affected by RRMS, develop MS earlier, have more relapses, and exhibit an inflammatory pattern when assessed by MRI; men display rapid progression, worse disease outcomes, and less inflammatory MRI patterns (Angeloni et al., 2021).

Treatment and Management

Despite the development and approval of disease-modifying treatments for patients with RRMS, treatment options are still limited for those with PSMS and SPMS. Furthermore, disease-modifying therapies are associated with severe adverse effects, creating an unmet need for developing safe and effective therapies for the different categories of MS (Filippi et al., 2018). Recently, chimeric antigen receptor (CAR) T cell therapy entered U.S. clinical trials for MS. It has been previously shown to effectively target cancer cells in patients with blood cancers; CAR T cells engineered to target B cells that react to myelin will be evaluated as a potential therapeutic for MS (Mullard, 2024).

Disparities

It is unclear whether disparities rooted in racial or ethnic differences affect the risk of MS in women. In a nested, case-control study of an ethnically diverse, population-based cohort from the United Kingdom, female predominance was observed with MS regardless of ethnicity; furthermore, modifiable risk factors associated with MS (e.g., smoking, obesity, infectious mononucleosis, low vitamin D, and head injury) had similar effects across all ethnicities (Jacobs et al., 2024). In contrast, a multicenter study from Italy indicated that individuals with MS from non-White ethnicities experienced a more rapid clinical decline compared to their White counterparts. As data was not disaggregated by sex, it is unclear if these findings also apply to comparisons of White with non-White women (Bianchi et al., 2024). Disparities in care have been observed in a study that showed that women and individuals with lower educational levels were less likely to be treated with disease-modifying therapies (Leavitt et al., 2024).

Research Gaps

Studies of the last 20 years provide strong evidence implicating both hormonal and sex chromosomal mechanisms in sex differences in MS. These studies suggest multiple gene pathways that might be targeted clinically.

Multiple immune system and CNS cell types function differently depending on their hormonal environment and whether they are XX or XY. Little is known about how the hormonal and sex chromosome mechanisms intersect to affect MS or EAE. The investigation of sex chromosome effects in diverse tissues is still in its infancy, and much work is needed to identify X and Y genetic elements that influence disease.

The interactions between genetic predispositions to MS and environmental, or extrinsic, factors determine the likelihood of developing this disease. However, more research is needed to understand how other factors, particularly Epstein-Barr viral infection, obesity, and lack of sun exposure, prime the immune and nervous systems for an autoimmune attack of the myelin sheath, leading to MS. Furthermore, better therapies specifically catering to the unique symptomatology and progression in men and women need to be developed that target sex-specific disease pathways and address the sex- and gender-related environmental risk factors.

CHRONIC PAIN

Like many chronic conditions, chronic pain disproportionately affects women. Given that it can be general or localized, studies have reported a higher prevalence of chronic pain in women, particularly in terms of musculoskeletal pain, such as back pain, body pain, fibromyalgia, and osteoarthritis; abdominal pain; headaches; and pelvic pain.

Biological Factors

The instigation, propagation, and perception of pain involve several key physiological components: a stimulus that triggers the activation of pain-sensing or -modulating receptors on neurons, which then relay pain signaling through neuronal pathways or networks. The pain signals terminate at brain regions that process and perceive pain stimuli, resulting in either evading or tolerating pain. These components of pain processing appear to be regulated differently in males and females. For example, evidence exists that differences in brain structure between males and females commence early in embryonic development, before gonadal differentiation, indicating the important influence of sex chromosome effects (Mallard et al., 2021; Pallier et al., 2022); however, it is unclear if these brain regional differences include pain-processing centers. Depending on the sex, some brain regions may contain different populations of neurons or differentially expressed receptors involved in pain. Less apparent is if alternate pain-processing pathways are recruited according to sex—although past research has indicated sex differences in ascending pain neural transmission pathways and descending pain modulatory neural pathways (Mogil, 2012). As for pain triggers,

perhaps the best studied in the field of sex differences are the molecules released during inflammation by immune cells, a response influenced by gonadal hormones (Rosen et al., 2017).

Sex Chromosome Effects

Experiments using the FCG mouse model have demonstrated sex chromosome effects, independent from gonadal hormone effects, on pain and analgesic pathways. In one study, mice with two copies of the X chromosome⁷ exhibited higher sensitivity to two types of acute pain-inducing stimuli, regardless of gonadal type, compared to mice with only one X chromosome.⁸ As these mice had their gonads surgically removed, these findings suggested that the sex chromosomes themselves, rather than gonadal hormones, had a larger role in determining the pain threshold (Gioiosa et al., 2008). In a related study, XX mice showed less activation of pain-suppressing pathways mediated by the kappa opioid receptor irrespective of gonadal hormones, indicating the influence of sex chromosomes on analgesic mechanisms that dull the perception of pain (Taylor et al., 2022). Discovering the precise mechanisms underlying these effects, starting with the specific identification of X chromosome genes that differentially influence pain processing, is ongoing.

Research has uncovered the differential expression of 40 X chromosome genes following a traumatic event (e.g., car collision) in women who later developed chronic musculoskeletal pain and/or posttraumatic stress versus women who recovered from the accident (Yu et al., 2019). Many genes upregulated in the women experiencing longer recovery times with chronic pain were linked to immune and neuronal function (Yu et al., 2019). The investigators also observed increased expression of other X-linked genes that tend to escape X inactivation (Yu et al., 2019). Collectively, these findings suggest the possible involvement of certain X chromosome genes in triggering and perpetuating mechanisms of pain in women.

The Effects of Estrogens

Ample evidence exists that supports the role of gonadal hormones as critical mediators of the sex differences observed for chronic pain. Although the male hormone testosterone is generally associated with pain suppression (Lenert et al., 2021; Lesnak et al., 2020), estrogens have dual, opposing effects, promoting both analgesia and an increased sensitivity to pain, depending on the experimental context (Hernandez-Leon et al., 2018;

___________________

⁷ XX which developed ovaries or XX+SRY which developed testes.

⁸ XY which developed testes or XY-SRY which developed ovaries.

Linher-Melville et al., 2020). Research has linked low levels of estrogens to a greater inflammatory response that aggravates pain signaling (Lenert et al., 2021). However, high levels of estrogen also correlated with enhanced pain sensitivity (Hellman et al., 2021; Lenert et al., 2021), indicating the complexity of estrogen-mediated regulation of pain mechanisms. Although it is difficult to reconcile these seemingly contradictory results, numerous studies have provided evidence that hormonal fluctuations lead to increased pain, but stable hormone levels—after menopause, for example—suppress pain (Lenert et al., 2021).

Numerous studies have shown that gonadal hormones exert sex-specific effects on neural pain pathways and the immune response, which is an important component that triggers and modulates the sensation of pain (Lenert et al., 2021). Estrogens modulate pain transmission in females by binding to estrogen receptors that are widely expressed in the pain processing areas of the peripheral and central nervous systems (Maurer et al., 2016). Research has also shown that estrogens enhance anti-inflammatory responses (Athnaiel et al., 2023). In contrast, testosterone suppresses proinflammatory signaling molecules (Lenert et al., 2021) and decreases signaling mediated by the pain-causing chemical capsaicin, thereby supporting its analgesic effect in males (Athnaiel et al., 2023). Testosterone has similar analgesic effects in women, as evidenced by studies showing that higher levels of testosterone were associated with less neck and shoulder pain (Kaergaard et al., 2000), and that it eased pain associated with vulvodynia upon topical application along with estrogens (Burrows and Goldstein, 2013) (see Chapter 5 on female-specific and gynecologic conditions), ameliorated pain responses after formalin exposure in female rats (Aloisi et al., 2004), and reduced hypersensitivity to muscle pain in female mice by acting on the serotonin transporter (Lesnak et al., 2020).

Sex Differences in Pain Sensitivity

Research has identified sex differences in pain sensitivity, response to analgesics, and propensity to develop chronic pain after injury or disease. A literature search regarding 77 painful disorders revealed female prevalence for 38 of them and male prevalence in 15 (Berkley, 1997). Meta-analyses of epidemiological surveys (Fillingim et al., 2009), analyses of electronic medical record data (Ruau et al., 2012), and meta-analysis of clinical trial data (Li et al., 2023) all confirm that women report more frequent and greater levels of pain than men. In addition, evidence from international collaborations assessing sex and gender inequalities related to chronic pain indicates that a greater proportion of women are affected than men (Bimpong et al., 2022; Tsang et al., 2008). The Global Burden of Diseases, Injuries, and Risk Factors Study 2016 reported that musculoskeletal disorders, which included

lower back and neck pain, were the largest contributors to higher rates of years lived with disability (YLD) in women globally (GBD 2016 Disease and Injury Incidence and Prevalence Collaborators, 2017). In another study, pooled results from 19 European countries showed more women than men reporting chronic pain, with the greatest inequalities in back and neck pain; however, when data was disaggregated by country, there were variations in the extent of sex differences (Bimpong et al., 2022). While the influence of the female sex on chronic pain was consistently observed, the differences in pain inequalities among the studied nations suggest that gender, shaped by social and cultural influences unique to each country, may also play an important role in the expression of pain, and consequently the development of chronic pain in women (Ayres et al., 2023; Bimpong et al., 2022). Because systematic investigation into gender differences in pain is in its infancy, it remains unclear whether sex differences in pain perception result from biological factors or from gender differences in the reporting of chronic pain arising from sociocultural factors (Boerner et al., 2018). Studies on sex differences conducted over the last 10–20 years aimed to determine the mechanisms that link chronic pain and developing other pain-related conditions (chronic overlapping pain conditions), identify the factors that lead to chronic pain, and discern how they differ by sex (Hoffmann et al., 2022).

To clarify sex differences in pain sensitivity from sex and gender differences in disease susceptibility and pain reporting, researchers have conducted hundreds of controlled laboratory studies comparing pain thresholds, tolerances, and ratings between men and women. The variability in the resulting literature has encouraged some to conclude that sex differences in pain are inconsistent and possibly illusory (Racine et al., 2012) or at least more complex than a simple matter of differential “sensitivity” (Hashmi and Davis, 2014). However, when ignoring statistical significance of a study’s results and simply tallying the number of studies in which women were more sensitive to pain than men and vice versa, dramatic overall evidence appears that women are more sensitive to pain, irrespective of measure or modality (Mogil, 2012).

Many potential explanations exist for this sex difference. Theories supported by evidence include those at the level of genetics, neurochemistry, cell- and systems-level processes, organizational and activational effects of gonadal hormones, experiential factors, and gender-related sociocultural factors (Mogil, 2012).

Pain Processing in Female Versus Male Individuals

Regardless of the true explanation(s) of quantitative sex differences in pain (i.e., sex differences in the amount of pain or analgesia), increasing evidence (such as the null mutant phenotypes) suggests robust qualitative

sex differences in the operation of the biological mechanisms underlying pain processing and modulation (Mogil, 2020).

The now much-more-common use of both sexes in preclinical research has led to new revelations regarding sex-specific biological processing of pain. Perhaps the most celebrated and replicated of these findings concerns the role of spinal cord microglia⁹ in developing chronic pain (Ghazisaeidi et al., 2023; Gregus et al., 2021; Halievski et al., 2020; Midavaine et al., 2021; Rosen et al., 2017; Sorge et al., 2015). One of the main advances in preclinical pain research over the past 20 years has been gradually realizing the critical role of the innate immune system, especially spinal cord microglia, in pain processing (Beggs et al., 2012; De Leo et al., 2006; Ji et al., 2019; Marchand et al., 2005; Scholz and Woolf, 2007; Watkins and Maier, 2003). Investigators conducted virtually all research in this subfield on male mice and rats, as was typical at the time, but in a seminal sex difference finding, one team demonstrated that the microglial dependence of chronic pain was entirely a male-specific phenomenon (Sorge et al., 2015). This same group demonstrated that in female mice and male mice with lowered testosterone levels, T cells of the adaptive immune system were mediating pain signal processing, although the details of female-specific T cell signaling have yet to be fully elucidated. There have been other demonstrations of female-specific pain circuitry, including work demonstrating the role of prolactin—a protein that enables mammals to produce milk—and its receptor in pain sensitization and pain signaling (Chen et al., 2020).

Qualitative sex differences relevant to pain are not only seen in animal studies. A growing proportion of mechanistic pain research is being performed by RNA sequencing of human tissues. The sets of genes such studies have identified tend to depend significantly on the sex of the tissue donor (Ray et al., 2019, 2023; Tavares-Ferreira et al., 2022).

Specific Risk Factors in Women

Researchers have typically viewed chronic pain from a biopsychosocial perspective to identify the factors leading to it and discern the interaction among these factors (Hoffmann et al., 2022). General risk factors include demographics; lifestyle factors, such as physical activity, occupation, and nutrition; clinical factors, and history of injury, abuse, or interpersonal violence (Mills et al., 2019). Lower pain self-efficacy¹⁰ and a higher tendency

___________________

⁹ Microglia are cells in the nervous system that share many if not all the properties of immune system macrophages found in other tissues.

¹⁰ Pain self-efficacy is described as an individual’s belief in their ability to effectively control their pain (Hoffmann et al., 2022).

for pain catastrophizing¹¹ are gender related behaviors associated with chronic pain in women (Hoffmann et al., 2022), although these factors are generally not well established.

ACEs

Research has linked ACEs to an increased risk for several adverse health conditions, including chronic pain (Anda et al., 2010; Felitti et al., 1998; Kerker et al., 2015; Nelson et al., 2017, 2018). Emerging research suggests that exposure to certain ACEs may be more predictive of poor health outcomes, including chronic pain (Atzl et al., 2019; Beal et al., 2020). Evidence suggests that rates of certain ACEs, particularly maltreatment, are higher among parents of youth experiencing chronic pain, suggesting they may confer risk for child pain (Beveridge et al., 2020). Studies have shown that exposure to maltreatment was the driving factor in the relationship between parental ACEs and a gene encoding a specific dopamine receptor in young women. Decreased expression of this gene in young women of parents who were maltreated may result in reduced dopamine signaling, which might explain the increased susceptibility of female chronic pain patients to mental disorders, such as anxiety and depression that are known to have compounding effects on chronic pain (Altemus, 2006; Breslau, 2009; Leadbeater et al., 1999; Noble, 2005; Noel et al., 2016; Piccinelli and Wilkinson, 2000).

Reproductive Milestones

Limited epidemiological data support the idea that chronic pain is likely to develop at certain stages of life in women, such as adolescence and old age; it is thought to be influenced by certain sex-linked factors, such as hormones and reproductive factors (Casale et al., 2021; Osborne and Davis, 2022). How the menstrual cycle influences the perception of pain is not well understood (Piroli et al., 2019), nor is the etiology of underlying chronic pain risk among women in midlife. Research has shown that estrogens and other reproductive hormones have complex interactions with pain modulation and pain sensitivity, with hormonal fluctuations—particularly the decline in estrogens in perimenopause and into post-menopause—potentially influencing vulnerability to developing and exacerbating pain conditions during midlife (Pavlovic and Derby, 2022).

Evidence that the stressors experienced by midlife women affect symptom sensitivity and pain tolerance (Lachman et al., 2015). They face the

___________________

¹¹ Pain catastrophizing refers to the continuous, exaggerated, and negative thoughts and feelings occurring during painful experiences (Leung, 2012).

physical effects of aging linked to musculoskeletal pain. A systematic review and meta-analysis of studies on musculoskeletal pain during midlife demonstrated that women in perimenopause were particularly vulnerable to developing it compared to premenopausal women and that postmenopausal women were vulnerable to develop moderate to severe musculoskeletal pain compared to premenopausal and perimenopausal women (Lu et al., 2020).

Weight gain and decreased physical activity at midlife may further contribute to chronic pain morbidity (Gibson et al., 2019). An analysis of data on middle-aged adults from the biannual Health and Retirement Study found that being overweight or obese accounted for 32.1 percent of the increases in severe or limiting pain among women, compared to 19.0 percent in men and as much as 20 percent of the increase in mild or moderate pain in women compared to 10 percent in men (Stokes et al., 2020). Obesity and chronic knee pain are often related to the onset of osteoarthritis in women, with greater risks of developing osteoarthritis as a function of weight gain in women compared to men (Batushansky et al., 2022). Aside from the biomechanical stress elicited by excess physical weight, obesity also alters metabolism and promotes inflammation, which increases the risk of osteoarthritis and chronic pain (Ait Eldjoudi et al., 2022; Batushansky et al., 2022).

In a national sample of midlife women U.S. veterans, those with menopausal symptoms had nearly twofold higher odds of diagnosed chronic pain and chronic pain multimorbidity, accounting for age and other known risk factors (Gibson et al., 2019). These findings may suggest that midlife is a vulnerable and critical period for women to experience chronic pain.

Gender-related factors that influence chronic pain in women

Factors associated with gender (see Chapter 2) also influence pain experiences and related behaviors in women and men (Keogh, 2022). Certain gender-related constructs, such as traditional masculinity, which is often associated with stoicism and independence, and femininity, typically characterized by sensitivity, sociability, compassion and expressiveness, can shape the perceptions and experiences of pain (Keogh, 2022). Greater tolerance to pain in experimental settings was observed in men identifying with traditional male roles compared with women who identified with traditional female roles (Hoffmann et al., 2022; Pool et al., 2007). Associations between high masculine/low feminine traits and pain tolerance have been noted in several studies, regardless of biological sex (Alabas et al., 2012; Applegate et al., 2005; Keogh, 2022). In addition, compared to men, women report their pain more often, demonstrate a greater likelihood for seeking help and for health care utilization (Keogh, 2021, 2022; Mills et al., 2019), and tend to exhibit pain-related catastrophizing and

lower self-efficacy, both of which can aggravate pain (Keogh, 2022). Other evidence has shown that women and men use different coping strategies to manage chronic pain, with women relying more on social support, positive self-affirmation, and self-care, while men resort to avoidance, distraction, and/or problem-solving techniques (Hoffmann et al., 2022; Keogh, 2022). Although gender-based behaviors in response to pain likely originate in early childhood through one’s social environment, pain-related behaviors are also informed by existing socially accepted stereotypes (Keogh, 2021). A better understanding of the impact of gender constructs on the perception and experience of pain is needed to formulate effective treatment strategies for both women and men.

Gender factors influencing chronic pain in women are also shaped by culture (Ayres et al., 2023). For example, a comparison of men and women in the U.S. versus Romania showed that despite receiving the same types and dosages of pain medications, women reported higher pain scores than men in the U.S., but not in Romania (Ayres et al., 2023). These differences in the behavior of reporting pain—reflecting related gender aspects such as coping mechanisms and adherence to social norms—may depend on the cultural context (Ayres et al., 2023). Overall, evidence to date supports a female bias in developing chronic pain, but variability in sex differences across countries are likely influenced by social and cultural differences (Ayres et al., 2023).

Condition Presentation and Diagnosis

Diagnosing pain in women is challenging given its complexity, and studies have investigated differences in symptoms associated with chronic pain. One important point relevant to sex differences in humans is that women are more likely than men to be diagnosed with a musculoskeletal condition leading to pain and have comorbidities or chronic overlapping pain conditions that may affect the risk of chronic pain, but it is not known if these comorbidities contribute to or are a result of heightened sensitivity to noxious stimuli (Templeton, 2020).

Much is known about sex differences in behavioral responses to noxious stimuli that shape pain perception. However, little is known about how to extrapolate the developments evident in epidemiological and preclinical data to clinical settings, resulting in the slow progress of diagnosing and identifying chronic pain in women. For example, data from a meta-analysis of studies that investigated high- and low-intensity pain response among healthy controls and those with depression showed that overall pain tolerance was similar across depressed and control groups, but low-intensity stimulation had small but significantly higher average sensory and pain thresholds in depressed participants (Thompson et al., 2016).

This example led to another development in chronic pain research, “phenotyping,” which recognizes that even people with the same condition have tremendous variability in signs, symptoms, and features (Hoffmann et al., 2022), reflecting the influence of biopsychosocial factors, such as sex, gender, and age (Ocay et al., 2022). Thus, placing an emphasis on clustering individuals with chronic pain in research studies is warranted so that they can identify those whose pain may be driven by different underlying mechanisms (Hoffmann et al., 2022). Fibromyalgia is another example; it presents with variable symptoms, including mood disturbances, fatigue, pain, painful bladder, gastrointestinal issues, sleep issues, and restless legs (Hoffmann et al., 2022). This variability has led researchers to identify subcategories, including pain, physical involvement, psychological function, and social support, that could lead to a more comprehensive assessment of an individual’s symptoms (Yim et al., 2017). These results suggest that phenotyping may be a strategy to better understand how to manage chronic pain in women.

Research has begun to focus on chronic overlapping pain conditions, where one develops with others (Hoffmann et al., 2022), such as chronic lower back pain, chronic migraine and headache, fibromyalgia, irritable bowel syndrome, painful bladder syndrome, endometriosis, chronic tension-type headache, temporomandibular disorders, vulvodynia, and myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS), all of which are more common in women (Hoffmann et al., 2022).

Women face a unique challenge in the health care system when it comes to diagnoses of multiple overlapping pain conditions such as fibromyalgia and migraines, both of which are more common in women and perceived to be poorly understood by medical providers (Driscoll et al., 2018). Given the complexity in diagnoses and management of common pain disorders among women, in addition to the lack of multidisciplinary treatment teams and the stigma and bias that serve as complicating factors in pain management, women are left feeling dismissed, disrespected, and frustrated by the health system when it comes to seeking treatment for pain.

A study identified a bias toward underestimation of pain in patients from the perspective of caregivers’ and clinicians’ pain stereotypes. The study demonstrated that caregivers and clinicians estimated women to be in less pain than men at the same level of pain expressiveness and self-report, specifically underestimating women’s pain relative to actual self-reported pain while overestimating men’s (Zhang et al., 2021). They also indicated that women would benefit more than men from psychotherapy versus pharmacotherapy when choosing between the two (Zhang et al., 2021). These findings support the reality that clinicians underestimate women’s pain and see it as less intense than men’s, which leads to ineffective or suboptimal pain treatment recommendations.

Treatment and Management

Treating and managing chronic pain includes pharmacological interventions, such as opioid analgesics and nonopioid analgesics, and nonpharmacological interventions, such as physical therapy, acupuncture, and meditation (Qaseem et al., 2017). Few pharmacological options have been developed specifically for women, and treatment is largely with drugs that were developed for other diseases and have limited effectiveness and problematic tolerability (Casale et al., 2021). Although studies have shown that women are less likely to receive an opioid analgesic, meta-analyses and clinical experimental studies suggest that opioids may provide greater relief from pain for women than for men, although that difference depends on factors such as age and comorbid mental disorders (Pisanu et al., 2019).

In addition, the lack of knowledge regarding how to properly treat pain has led to a knowledge gap when it comes to treating pain in women. Women have reported that they were less informed about the availability of certain treatment modalities, including complementary and integrative health, which is an approach to health care that a majority of non-complementary and integrative health care providers are not familiar with, which means they lack the knowledge needed to provide appropriate counseling (Washington et al., 2011). There are advantages to creating an integrated health care system that prioritizes lifestyle, exercises, self-care, complementary and integrative health, and interdisciplinary modalities as effective and critical treatment options to handle the unmet need in addressing, evaluating, and managing chronic pain disorders among women.

Midlife women experience increasing pain symptoms related to generalized body aches, headaches, and genitourinary pain as reported by a multiethnic cohort of midlife women that identified clear patterns of symptom clustering (Harlow et al., 2017). However, the scarcity of studies regarding pain symptoms have contributed to underdiagnosis of specific pain disorders in the setting of common symptoms endured by midlife women. Thus, poor management of their chronic pain remains a growing national problem, with a critical need to fill this research gap and address pain symptoms; for example, the highest increase in opioid prescribing and opioid-related overdose mortality has been among women in their 40s to 60s (Gibson et al., 2019).

Disparities

Studies suggest that the prevalence of chronic pain disproportionally affects certain racial and ethnic groups, resulting in health disparities (Anderson et al., 2009; Janevic et al., 2017). Some studies have sought to understand how differential exposures to psychosocial and environmental

factors influence these differences (Aroke et al., 2019; Overstreet et al., 2023). Early exposure to psychological and environmental risk factors is associated with chronic pain and contributes to disparities in chronic pain prevalence among people of color (Overstreet et al., 2023). Few studies have focused on identifying the psychological factors that lead to higher levels of pain sensitivity among people of color. For example, although the Hispanic and Latino/Latina population had the most population growth in the U.S. between 2010 and 2020 (Census Bureau, 2021), limited research has focused on their experiences with pain (Zajacova et al., 2022). Population-based studies indicate higher prevalence and greater pain sensitivity among Black, Asian, and Hispanic participants compared to White participants (Zajacova et al., 2022). A higher prevalence of pain was also observed in American Indians and Alaska Natives; however, unlike the other ethnic groups, greater pain sensitivity compared to White individuals was not apparent (Zajacova et al., 2022). These findings suggest that current experimental pain research may not reflect the extent of pain disparities in the general population (Zajacova et al., 2022).

Lesbian, Gay, Bisexual, Transgender, Queer, Intersex, Asexual, and Other (LGBTQIA+) Disparities

Compared to the general population, LGBTQIA+ individuals experience more chronic physical and mental health problems than heterosexual individuals (Moagi et al., 2021; Pinnamaneni et al., 2022). Although poor physical and mental health are known correlates, (Edwards et al., 2016), PRIDE¹² is the first cohort study that assessed chronic pain among U.S. sexual and gender minority adults (Chadwick et al., 2024). Compared to cisgender men, transgender women had 1.46 greater odds of widespread chronic pain (Chadwick et al., 2024). Research has also shown that the LGBTQIA+ population is at greater risk for chronic pain than their cisgender, straight counterparts. One study, for example, found that pain prevalence among sexual minority adults was 4.5–11.4 percent higher than among straight adults (Zajacova et al., 2023).

Research has shown that many disparities in chronic conditions, disability, and mental distress observed in younger lesbian, gay, and bisexual adults persist into adulthood compared to other populations (Fredriksen-Goldsen et al., 2017). Studies have also found that LGBTQIA+ adults are more likely to experience more frequent functional limitations from pain and pain from multiple sites compared to heterosexual adults (Case et al., 2004; Chadwick et al., 2024; Fredriksen-Goldsen et al., 2017). Additional studies examining the association between sexual orientation

___________________

¹² PRIDE is a community-engaged research study conducted in the LGBTQIA+ community.

and pain among a range of age groups and residencies found that sexually minoritized adults have comparable to higher prevalence of pain or painful conditions (Cochran and Mays, 2007; Fredriksen-Goldsen et al., 2017; Katz-Wise et al., 2015; Nagata et al., 2020; Roberts et al., 2013).

Research Gaps

The National Institutes of Health’s Sex as a Biological Variable policy in 2016 forced investigators to account for sex as a biological variable in research study designs and should address the problem that many of the biological mechanisms underlying pain are relevant only to the male subjects that had dominated pain research. It seems likely that sex differences in pain mechanisms (Nemmani et al., 2004; Sorge et al., 2015) may have been responsible for the failure of certain clinical trials of analgesics (Galer et al., 2005). The plethora of sex-dependent pain mechanisms suggests that some therapies will be effective in only one sex. That day may have already arrived. An analysis of publicly available clinical reviews of the New Drug Applications for ubrogepant (Ubrelvy^®) and rimegapant (Nurtec ODT^®) revealed that they were only effective as migraine treatments in women (Porreca and Dodick, 2023).

Although human studies now typically enroll both men and women, disaggregation of data by sex continues to be a problem, with less than 50 percent of the papers in Pain 2015–2019 featuring any mention of sex differences in the results or discussion sections (Mogil, 2020).

The influence of hormonal factors on chronic pain is nuanced, and more studies are needed. Although studies have investigated sex differences in chronic pain, the primary biological and hormonal mechanisms that explain these differences in prevalence and expression continue to be understudied (Pavlovic and Derby, 2022). In addition, the female-specific risk factors are not well established. The literature suggests that midlife is a critical period for women during which the frequency and intensity of pain symptoms increase, but only a few studies have addressed the epidemiology and mechanisms of pain in midlife women.

Chronic pain diagnosis is a significant research gap, partly due to its similarity with other overlapping pain conditions, with a pressing need for more research to unravel the connections between chronic pain and overlapping pain conditions and understand gender biases in diagnosing pain in women. Studies for establishing consistent definitions of chronic pain are needed; the absence of a clear diagnostic framework for women complicates the ability to synthesize findings across various studies, hindering progress in understanding and treatment.

Moreover, the understanding of effective treatment and management strategies for chronic pain in women is insufficient, posing severe challenges

for women navigating the health care system, with a notable gap in research on how chronic pain medications specifically affect women and a need for more studies on self-care and nonpharmacological interventions. Additionally, the disparities in chronic pain experiences among women of diverse racial and ethnic backgrounds and within the LGBTQIA+ community remain poorly understood. This includes a lack of clarity on the reasons behind the mismatch between measures of pain sensitivity and reported prevalence rates, and a scarcity of research investigating the factors contributing to chronic pain in these groups.

FIBROMYALGIA

Fibromyalgia is a syndrome characterized by chronic musculoskeletal pain and symptoms such as fatigue, sleep disturbances, and mood disorders. It is believed to be caused by CNS dysfunction (Siracusa et al., 2021). Disagreement and ambiguity regarding diagnostic criteria and presence of bias were themes highlighting the historical context of fibromyalgia (Arout et al., 2018; Galvez-Sánchez and Reyes Del Paso, 2020).

A review of the diagnostic criteria for fibromyalgia highlighted the evolution from a singular to a multi-symptom diagnosis (Galvez-Sánchez and Reyes Del Paso, 2020). First described in 1642 as “muscular rheumatism” and later in 1904 as “fibrositis,” it was labeled “fibromyalgia” in 1976. The concept was centered around complaints of pain in at least three areas, no other disorders that may contribute to the symptoms, and at least five consistent tender points (Yunus et al., 1981). It was not until 1990 that the American College of Rheumatology (ACR) established official diagnostic criteria (Wolfe et al., 1990). However, defining and diagnosing fibromyalgia has remained a challenge, and research aimed at understanding how to prevent, diagnose, and treat it has lagged, as has research on disparities among women.

Biological Factors

The etiology of fibromyalgia is largely unknown and may be similar to mechanisms described in chronic pain for central sensitization¹³ and endogenous pain inhibitory mechanisms (Galvez-Sánchez and Reyes Del Paso, 2020). Greater perception of pain has been a common thread in the diagnosis, perhaps from central sensitization that may result from persistent painful stimuli associated with injured tissue that triggers persistent

___________________

¹³ Central sensitization occurs when the central nervous system becomes hypersensitive to pain-inducing stimuli, including injury, pressure, temperature, and/or movement (Hoffmann et al., 2022).

peripheral sensitization. Some studies have shown peripheral nerve changes in fibromyalgia, including a reduced number of epidermal nerve fibers in skin biopsies and altered heat, cold, and pain thresholds. Researchers have postulated that blocking peripheral sensitization may stop the chronic pain. Other principal processes underlying fibromyalgia include dysfunction of neural circuits and inflammatory, immune, endocrine, genetic, and psychosocial factors (Siracusa et al., 2021). Although the etiology is not well understood, numerous studies have demonstrated that more women have fibromyalgia than men, implicating biological sex as a driving factor (Martínez-Lavín, 2021).

Sex Chromosome Effects

No direct evidence exists for sex chromosome effects on fibromyalgia. Some researchers have noted its co-occurrence with several autoimmune conditions, including autoimmune thyroiditis and Sjogren’s syndrome (Applbaum and Lichtbroun, 2019; Martínez-Lavín, 2021; Tagoe et al., 2019), which hints at an autoimmune-based etiology (Meester et al., 2020). Considering that many autoimmune conditions have recently been attributed to incomplete X chromosome inactivation, leading to higher expression of immune-system-related genes (Pallier et al., 2022), an influence mediated by the X chromosome may exist in comorbidities that include fibromyalgia.

The Effects of Estrogens

Estrogens may exert analgesic effects that depend on a woman’s hormonal milieu. For example, menopausal women with erratic hormonal fluctuations before the final period may be more likely to experience fibromyalgia, and estrogen therapy has been shown to reduce pain (Athnaiel et al., 2023). Estrogen’s analgesic effects were also seen in an experimental animal model of fibromyalgia in which surgically removing the ovaries aggravated pain symptoms in female rats but administering estrogen reduced them (Hernandez-Leon et al., 2018). Together, these preclinical and clinical results suggest that steady levels of estrogens may reduce pain perception, but fluctuations or a sudden drop in levels of estrogen exacerbate pain in women with fibromyalgia (Athnaiel et al., 2023).

In contrast to these findings, other studies suggest that estrogens increase pain symptoms in fibromyalgia. For example, animal model studies demonstrated that estradiol upregulates the expression of an ion channel protein, Acid-Sensing Ion Channel-3 in dorsal root ganglia—a neural population implicated in conducting fibromyalgia pain (Martinez-Lavin and Solano, 2009; Ren et al., 2018) Estradiol also increases the expression

of the sodium channel Nav1.7, in trigeminal ganglion neurons, aggravating the pain response in animal models (Bi et al., 2017). It is interesting to note that a polymorphism in the gene encoding Nav1.7, SCN9A, has been associated with a severe form of fibromyalgia in a small cohort of Mexican women (Vargas-Alarcon et al., 2012), suggesting that estrogen-dependent regulation of ion channels may influence symptoms of fibromyalgia, and other chronic pain disorders predominantly affecting women.

Other studies suggest that variations in ERs may mediate the effects of estrogens on the preponderance of fibromyalgia among women. A study of Turkish women found that variants of estrogen receptor-alpha (ER-α) were more prevalent in fibromyalgia patients compared to controls. The investigators noted that specific genotypes were associated with certain symptoms, such as headache and menstrual cramps (Arslan et al., 2021). Although estrogens appears to modulate the fibromyalgia pain mechanisms, whether it promotes or suppresses pain perception and the mechanisms underlying these dual effects remain unresolved.

Other Biological Factors and Mechanisms

Despite gaps in the understanding of the etiology of fibromyalgia, a growing consensus supports a genetic contribution to it (Ablin and Buskila, 2015). Familial inheritance patterns have been observed, with the odds ratio of fibromyalgia higher in relatives of individuals diagnosed with fibromyalgia versus rheumatoid arthritis (Arnold et al., 2004). Research has shown that genetic polymorphisms in serotonergic, dopaminergic, and catecholaminergic neural transmitter systems involved in pain processing are present in fibromyalgia. One specific genotype in a serotonin transporter gene was associated with fibromyalgia and symptoms of depression and anxiety (Cohen et al., 2002; Offenbaecher et al., 1999). In contrast, the frequency of a polymorphism in a dopamine receptor gene was negatively correlated with fibromyalgia (Buskila et al., 2004). Genome-wide analyses have identified additional genes involved in neurotransmission (Arnold et al., 2013; Docampo et al., 2014). These studies suggest that the genetic vulnerability may account for as much as half of the risk of fibromyalgia (Siracusa et al., 2021).

Studies on the role of several proteins in fibromyalgia pain, including Substance P, nerve growth factor, brain-derived neurotrophic factor, and pentraxin-3 have been inconclusive, with only some studies showing an association (Favretti et al., 2023).

Research is exploring the role of vitamin D and fibromyalgia. Despite some studies suggesting a potential relationship between vitamin D deficiency and fibromyalgia, other studies suggest that these findings may be related to limited exercise and sun exposure, and the effectiveness of

vitamin D supplementation in addressing the pain is unclear. Research is needed on the relationship between changes in the gut microbiome and fibromyalgia severity (Favretti et al., 2023).

Specific Risk Factors in Women

Factors associated with fibromyalgia in women are largely understudied and lack evidence based on studies in U.S. populations. Potential factors are younger age and mental disorders such as depression and anxiety, but these are mainly based on cross-sectional studies (Arout et al., 2018). A history of ACEs, IPV, and physical abuse have been identified as potential factors affecting women, but these findings are based on limited cross-sectional studies (Bacon and White, 2023; Chalder et al., 2021; Chandan et al., 2021). Women veterans are identified as a population at risk of fibromyalgia (D’Aoust et al., 2017).

The role of female gonadal hormones in fibromyalgia is evident from studies conducted at different stages of the life-span. For example, higher incidence of fibromyalgia was observed in girls compared to boys at the onset of puberty (Kashikar-Zuck and Ting, 2014; McLeod, 2014). Another study correlated hormone levels during the menstrual cycle with changes in pain symptoms in women with fibromyalgia and found that high levels of progesterone and testosterone were associated with lower pain severity (Schertzinger et al., 2018). In contrast, estrogen and cortisol levels did not influence pain severity, though a post-hoc analysis showed that the association of high progesterone with low pain severity was significant only when cortisol levels were high (Schertzinger et al., 2018). This study along with separate findings showing how treatment with testosterone gel improved pain symptoms in a small cohort of fibromyalgia patients (White et al., 2015) is in line with testosterone’s purported analgesic effects, previously described in the Chronic Pain section of this report.

Condition Presentation and Diagnosis

Gender differences and potential gender biases in diagnosis are a topic of debate, and study results depend on the use of criteria that are appropriate in research settings versus clinically based fibromyalgia measures (Wolfe et al., 1990). The latter are subject to confirmation and selection bias and present major limitations in validity and reliability in a clinical context. Research studies show less unbiased estimates in research contexts, especially among women, and it has been stated that this condition predominantly affects middle-aged women (Häuser and Fitzcharles, 2018; Wolfe et al., 2018). This situation can lead to challenges of misdiagnosis and overdiagnosis and overtreatment (Galvez-Sánchez and Reyes Del Paso, 2020).

Given these biases and that the diagnosis is primarily clinical, identifying biomarkers would be helpful for both diagnosis and monitoring the benefit of treatment. Despite 25 years of research, no specific biomarker has been identified (Favretti et al., 2023).

The typical age of onset is 30–35 (Siracusa et al., 2021). ACR criteria have provided essential guidance for clinical assessment for decades and were recently updated. The key diagnostic criteria include complaints of pain above and below the waist and generalized pain for at least 3 months (Siracusa et al., 2021). To facilitate diagnosis in the clinical setting, researchers developed survey instruments, such as the Polysymptomatic Distress Scale, which incorporates the Widespread Pain Index. These instruments also provided a means to assess the severity of the disease in any given patient.

Fibromyalgia has other associated characteristics and symptoms. Depression is the most common chronic condition associated with it, with a weighted prevalence of up to 63 percent of patients. Other comorbidities include migraine/headache, irritable bowel syndrome, myofascial pain syndrome, and temporomandibular disorders (Kleykamp et al., 2021). Overactive bladder syndrome is also common in women with fibromyalgia, with as many as 40 percent experiencing it (Salaffi et al., 2022). Sexual dysfunction affects as many as 86.9 percent of women with fibromyalgia (Salaffi et al., 2022). Despite the significant prevalence of these two conditions, they are understudied, as instruments such as the Fibromyalgia Impact Questionnaire used to estimate fibromyalgia in epidemiologic studies do not include items for them (Salaffi et al., 2022).

Treatment and Management

Many studies exist on therapeutic interventions for fibromyalgia. For example, researchers have used animal models of non-inflammatory pain to investigate the efficacy of specific medications in reducing fibromyalgia; antidepressants, opioids, glutamate receptor antagonists were effective. However, nonsteroidal anti-inflammatory agents were not effective (Siracusa et al., 2021).

A systematic review of clinical trials assessed the efficacy of two antidepressants commonly prescribed for treatment. Amitriptyline had a greater effect on quality of life but little evidence that it had a beneficial effect on remedying pain (de Farias et al., 2020).

The literature highlights exercise, massage therapy, and diet as potentially beneficial. Investigation continues into the best medical interventions of associated conditions, such as depression. Given the effects of fibromyalgia over the life course, it is important to appropriately adjust treatment based on not only severity but also the stage of life.

The effect of aquatic training and detraining on the physiology and the quality of life of women with fibromyalgia was assessed in a randomized trial. Aquatic training was associated with improvements in oxygen intake, an increase in pain threshold, and improved quality of life. However, these benefits declined when activity was reduced, suggesting that persistent activity is needed to maintain the benefits of exercise (Andrade et al., 2019).

One study assessed the effects of a tryptophan and magnesium-enriched Mediterranean diet in a randomized trial of 22 women with fibromyalgia over 16 weeks. The participants experienced improvements in sleep quality and psychological metrics compared to controls (Martínez-Rodríguez et al., 2020).

Considering that musculoskeletal pain is a common presenting symptom of fibromyalgia, investigators conducted an RCT to determine if massage therapy is effective in addressing pain and the psychological effects of fibromyalgia. Improvements were noted in the intervention group versus the control group across key variables: fatigue, pain, sleep, stress/anxiety, depression, anger, vigor, and fatigue. This study suggests benefits from massage therapy for fibromyalgia patients (Nadal-Nicolás et al., 2020).

A meta-analysis of RCTs evaluated the effectiveness of virtual-reality therapy in mitigating the impact of fibromyalgia on quality of life. The key findings were that exercise therapy showed significant improvements in functional movement, pain, dynamic balance, aerobic capacity, fatigue, quality of life, anxiety, and depression in women aged 38–55 with a BMI of 25–28. However, few studies and participants, lack of blinding, and potential publication bias limit the generalization and quality of the evidence (Cortés-Pérez et al., 2021).

Disparities

Limited research is available on racial and ethnic disparities in women with fibromyalgia, including how symptoms may vary, pain severity, and differential response to treatment. Findings from a clinical study found that Black women reported higher pain severity and pain interferences with daily functioning compared to White women (Jacobs et al., 2023). The authors mention that factors that may contribute to these disparities observed may be linked to structural and social determinants of health (see Chapter 7). A scoping review to describe the literature on fibromyalgia and other neurological and somatic conditions suggests that the prevalence of fibromyalgia is greater in sexual and gender minority individuals, but this was not broken down by group and was based on limited studies outside of the U.S. (Lerario et al., 2023).

Research Gaps

Research gaps can be identified in every category of the translational spectrum, ranging from the basic sciences to population health. In the basic sciences, developing an animal model that captures the key clinical findings will be important for identifying biomarkers that are predictive of the disease process. Furthermore, the close link between fibromyalgia and co-occurrence of autoimmune diseases suggests the role of incomplete X chromosome inactivation and should be explored further. Greater diversity is needed in publicly available databases used to assess the genetic predisposition to fibromyalgia.

Research to improve the diagnostic characterization of fibromyalgia, that is objective and validated across populations, is needed along with studies of diagnostic bias and strategies to mitigate bias. Moreover, as shown by the omission of overactive bladder in surveys used to diagnose and assess fibromyalgia severity, it is important to include all conditions associated with fibromyalgia in such surveys. Further examination of the role of lifestyle and environmental factors that shape the severity of the disease over the life-span is necessary.

Research is needed on an array of medical therapeutic, complementary and alternative therapies, and lifestyle options. Given that fibromyalgia often emerges in the mid-30s and can persist into postmenopause, a comprehensive understanding of treatment and support through these life stages is essential. Additionally, considering the impact on mobility, research should focus on identifying the environmental and social supports that could enhance quality of life.

MYALGIC ENCEPHALOMYELITIS/CHRONIC FATIGUE SYNDROME

As with fibromyalgia, diagnosing myalgic encephalomyelitis/chronic fatigue syndrome (ME/CFS) can be challenging. The definition has evolved over several decades and most commonly uses ME/CFS, a descriptive label that captures the most specific symptoms. Patients experience significant adverse effect on their quality of life, reporting high levels of unemployment, permanent inability to work, and significant dependence upon others to carry out activities of daily living (Bested and Marshall, 2015).

Biological Factors

The etiology of ME/CFS remains unknown, despite observations that it may be initiated by an infection (Lacerda et al., 2019), stressful experiences, such as ACEs (Heim et al., 2006), or other events, including immunization,

anesthesia, and exposure to environmental pollutants, chemicals, and heavy metals. Physical trauma, such as a motor vehicle accident, fall, or surgery, may also trigger it (Carruthers et al., 2003). Researchers have estimated that a flu-like illness is the precursor in 50–80 percent of cases (Bested and Marshall, 2015). Some studies have implicated viral illness with the idea that it may be the result of a chronic Epstein-Barr viral infection or infection by other viruses, such as human herpes-viruses 6 and 7, cytomegalovirus, and enterovirus, but the evidence is inconclusive (Grach et al., 2023; Yancey and Thomas, 2012).

Nonetheless, since the December 2019 global emergence of coronavirus disease 2019 (COVID-19), interest has been renewed in the potential link with viruses. In a recent meta-analysis of publications that assessed the prevalence of chronic fatigue following COVID-19 infection, chronic fatigue syndrome (CFS) developed 4 weeks after the initial infection in 45.2 percent of 127,711 patients (Salari et al., 2022). Some studies have shown increased prevalence of chronic fatigue or other long COVID symptoms for women (Huang et al., 2021), including a study that has been retracted because of inconsistencies in data reporting, but others have found no gender associations (Carfì et al., 2020; Chopra et al., 2021).

Despite ongoing research addressing the influence of biological sex on ME/CFS, efforts are hindered by the lack of consistency related to definitions and diagnostic criteria that make it difficult to conduct clinical studies and estimate prevalence in both women and men (Lim et al., 2020). In general, being female appears to be a predictive risk factor (Lacerda et al., 2019; Lim et al., 2020; Reyes et al., 2003; Thomas et al., 2022), and results support a prominent role for sex gonadal hormones underlying this sex bias.

Sex Chromosome Effects

The contribution of sex chromosomal effects to ME/CFS is unknown. Similar to fibromyalgia, indications suggest that an autoimmune component may contribute to the development and progression of ME/CFS (Sotzny et al., 2018), but research is required to concretely establish immune mechanisms and whether the X chromosome is involved.

The few published studies attempting to identify genes associated with ME/CFS have focused mainly on autosomes, not sex chromosomes (Das et al., 2022; Gamer et al., 2023). One exception was a GWAS that identified 35 single nucleotide polymorphisms (SNPs) located on the X chromosome, along with 407 autosomal SNPs, as potential gene candidates for association with ME/CFS (Schlauch et al., 2016).

The Effects of Estrogens

Lines of evidence supporting a role for female gonadal hormones in the pathophysiology of ME/CFS include (1) lower mRNA expression of ER-β, but not ER-α, in ME/CFS patients (Gräns et al., 2007); (2) certain SNPs in CYP2D6, a gene involved in androgen and estrogen biosynthesis and metabolism, at a higher frequency in ME/CFS patients (Perez et al., 2019); and (3) metabolomics analysis using mass spectrometry showing reductions in androgens and progestins in ME/CFS patients compared to controls (Germain et al., 2020).

Specific Risk Factors in Women

Risk factors for ME/CFS in women are not well understood, and very few studies have examined those unique to women. The Nurses Health cohort identified specific risk factors for ME/CFS versus severe fatigue and found that a history of mononucleosis was not associated of ME/CFS. Various factors, such as smoking, caffeine and alcohol intake, were not associated with ME/CFS (Palacios et al., 2023). BMI was marginally associated with ME/CFS, and physical activity, hormone therapy, and menopause status had no association (Palacios et al., 2023). A literature review on how ME/CFS and other related conditions, such as long COVID, develop noted that reproductive milestones, such as pregnancy, and endocrine events, such as early age of menopause, were associated with ME/CFS (Pollack et al., 2023).

The effects of female gonadal hormones likely underlie the preponderance of ME/CFS in women. Hormonal changes during the menstrual cycle and menopause have been reported to worsen symptoms, and similar detrimental effects were observed when women were administered hormone-based birth control and hormone therapy (Allen, 2008; Thomas et al., 2022). Together, these clinical studies suggest that gonadal hormones affect how women experience symptoms. However, it is unclear if pregnancy positively or negatively affects symptoms, or if ME/CFS adversely impacts pregnancy and/or the postpartum period, as findings are limited and have produced contradictory results (Allen, 2008).

No consensus exists yet on whether hormone levels during the menstrual cycle differ between women with ME/CFS and controls. Smaller studies of women with ME/CFS reported higher odds of gynecological conditions, including endometriosis, menstrual abnormalities, pelvic pain unrelated to menstruation, early hysterectomy, and gynecological surgeries, suggesting a relationship with sex hormones, but further research is needed (Boneva et al., 2011, 2015; Pollack et al., 2023). Two small, independent studies have demonstrated differences in estradiol levels (Boneva et al., 2015; Studd and Panay, 1996) and levels of progesterone and progesterone

metabolites (Murphy et al., 2004), but a third study found no notable differences in these hormones, nor in follicle stimulating hormone and luteinizing hormone (Cevik et al., 2004). To resolve this question, researchers should conduct follow-up studies in larger cohorts.

Condition Presentation and Diagnosis

The literature has 25 clinical definitions for this complex chronic condition, indicating a lack of consensus across disciplines (Lim and Son, 2020). These definitions are impacted by etiology, pathophysiology, and exclusion of other conditions (Lim and Son, 2020). Some investigators have favored the Canadian Consensus Criteria, which have been validated using online tools to assess symptoms and function (Bested and Marshall, 2015) and demonstrated to more effectively distinguish ME/CFS from other causes of fatigue, such as depression. Following the exclusion of other diagnoses associated with fatigue and muscle tenderness, patients are assessed based on the following criteria: fatigue considered pathological for a period of 3–6 months, malaise, sleep disorder, pain, and cognitive impairment. Moreover, two symptoms must fall into at least two of the following domains: immune, neuroendocrine, or autonomic (Bested and Marshall, 2015). Given the reliance on clinical symptoms without specific biomarkers, it is not surprising that patients are often not diagnosed and regularly face disbelief that their symptoms are valid.

It has been estimated that as many as 80 percent of patients who are symptomatic have struggled to be diagnosed, largely because providers have not been appropriately educated. ME/CFS has not been as well represented in the medical curriculum compared to other conditions, and targeted continuing education options may close this gap and improve earlier diagnosis (Bested and Marshall, 2015; Brimmer et al., 2009). A number of other conditions may also be reported, such as fibromyalgia, multiple chemical sensitivity, orthostatic intolerance, irritable bowel syndrome, irritable bladder syndrome, interstitial cystitis, sicca syndrome, temporomandibular joint syndrome, migraine headache, allergies, thyroiditis, Raynaud’s phenomenon, and prolapsed mitral valve (Bested and Marshall, 2015).

Treatment and Management

Given the complexity of the symptoms of ME/CFS, treatment requires a customized, integrated approach. Attention must first be given to any comorbid conditions, such as depression, sleep dysfunction, and pain. Research has demonstrated that cognitive behavior therapy and exercise are effective (Turner-Stokes and Wade, 2020). However, according to guidelines published by the National Institute for Health and Care Excellence

in the United Kingdom in 2020, these therapies may not be helpful despite published evidence to the contrary (Turner-Stokes and Wade, 2020). These guidelines were based on quality criteria that assessed the evidence for these therapies, which may not fit well when assessing complex conditions such as ME/CFS. Long-term observational trials would be better suited to assess the efficacy of interventions for ME/CFS (Turner-Stokes and Wade, 2020).

The effect on the lives of patients with ME/CFS is significant. They often feel stigmatized when presenting with an array of symptoms and not believed when sharing their symptoms with providers. Somatization, the conflation of physical and psychological explanation for symptoms, continues to require greater recognition in their care (Wessely, 1997). Providers are not well educated regarding diagnostic criteria, which are often overwhelmed by other comorbidities (Bateman et al., 2021; Bested and Marshall, 2015). Treatable active diseases that contribute to symptoms such as fatigue, pain, sleep dysfunction, and cognitive dysfunction must be ruled out before making a diagnosis of ME/CFS (Bested and Marshall, 2015). Moreover, treatment is embedded in controversy, although those individuals who are guided through a well-structured rehabilitation program may improve (Chambers, 2006; Wearden et al., 2010). The prognosis is variable and dependent upon the severity of the condition at onset and the stage of diagnosis (Bested and Marshall, 2015).

Earlier diagnosis and appropriate management should lead to a better outcome. On average, 5 percent of patients recover, 40 percent improve, 8–30 percent return to work, and 5–20 percent worsen (Bested and Marshall, 2015). Ongoing research exploring mechanisms such as redox imbalance (Paul et al., 2021), genetics, and immune system dysfunction may offer diagnostic and therapeutic paths (Bested and Marshall, 2015).

Disparities

The research on racial and ethnic disparities in ME/CFS is scant. An older meta-analysis of CFS reported on racial and ethnic differences in prevalence, symptom severity, and coping and illness behaviors. Three studies suggested that prevalence was greater in Black and African American and Alaska Native groups. The findings were mixed; greater symptom severity was reported in Latinas compared to Black and African American and White women, but another study indicated that Black and African American individuals reported more severe fatigue than White individuals. However, these findings are based on limited studies available (Dinos et al., 2009).

Research Gaps

Studies are needed to determine the underlying etiological mechanisms of ME/CFS in women and whether sex chromosome and hormonal effects

play a role. The multiple consensus criteria make it difficult to define this condition. A report from the Institute of Medicine¹⁴ recommended a new label, but this recommendation does not appear to have had much traction based on a review of the literature. One recommendation was to review the diagnostic criteria 5 years after publication, and the committee suggested that funding may be provided by “nonconflicted sources such as the Agency for Healthcare Research and Quality through its Evidenced-based Practice Centers process and foundations” (IOM, 2015). It is unclear if such a review process has been undertaken. The lack of established diagnostic tools and biomarkers that can rule out other similar chronic conditions present major challenges in properly diagnosing individuals who are experiencing clinical symptoms associated with ME/CFS. Other research needs include addressing the lack of relevant education in physicians and other providers, developing a better understanding of the basic mechanisms specific to it, and conducting long-term observational studies to better understand which patient-centered interventions are the most useful.

ALZHEIMER’S DISEASE

Alzheimer’s disease (AD) causes neurocognitive deficits. Despite advancements in the field characterizing genetic associations, delineating the disease process, identifying diagnostic and prognostic biomarkers, and assessing possible treatments, it still has no cure (Ferretti et al., 2018). Research into disease mechanisms has helped identify pathological biomarkers to gauge disease risk and monitor progression. Although AD is largely a heterogeneous disease, with high variability in timing, type of symptom manifestation, and progression stemming from multiple etiological sources, women have notable differences in risk factors, symptoms, and progression versus men.

Biological Factors

The three main underlying factors for increased AD risk are age, variation in the apolipoprotein E (ApoE) gene, and sex, with age being the most significant (Valencia-Olvera et al., 2023). The contribution of the aging process to AD is beyond the scope of this report. Even though the age-specific incidence rates are similar in women and men, the mechanisms underlying the incidence often vary by sex. Both the frequency and the strength of

___________________

¹⁴ As of March 2016, the Health and Medicine Division of the National Academies of Sciences, Engineering, and Medicine (the National Academies) continues the consensus studies and convening activities previously carried out by the Institute of Medicine (IOM). The IOM name is used to refer to reports issued prior to July 2015.

the effect of a given risk or protective factor may be different in men and women (Rocca et al., 2014).

Sex Chromosome Effects

Limited data exist to support a direct association of sex chromosome effects with AD, but a few studies indicate an influence of X chromosome number and X chromosomal genes. For example, X chromosome loss may be an underlying factor. One research group demonstrated that the likelihood of X chromosome loss in somatic cells increases 100-fold in women over age 65 compared to girls under age 16 (Russell et al., 2007), which may influence neurodegeneration and aging in the brain (Guo et al., 2022). Specific to AD, the rate of X chromosome loss increased by twofold in the hippocampus and cerebral brain regions of patients (Yurov et al., 2014). Conversely, an extra X chromosome may be neuroprotective, as illustrated by a preclinical study showing that a second X chromosome in a transgenic mouse model for AD reduced mortality and improved cognitive deficits (Davis et al., 2020).

Research has linked several genes on the X chromosome to AD. For example, KDM6A, which encodes a histone demethylase, is associated with cognitive function, as loss-of-function mutations in humans are linked to intellectual disability (Miyake et al., 2013), and a knockout of the same gene in mouse models leads to memory impairments (Tang et al., 2017). Studies have tied a SNP in the PCDH11X gene, known to play a role in CNS function, to higher risks of AD in women (Carrasquillo et al., 2009). GWAS did not replicate this finding (Beecham et al., 2010; Lescai et al., 2010; Miar et al., 2011; Wu et al., 2010), but more recent GWAS confirmed the association (Jiao et al., 2015; Zeng et al., 2021). These inconsistencies may result from differences in cohort size, ethnicity of the study population, environmental factors, and computational method (Bajic et al., 2019). Analysis of data from the Alzheimer’s Disease Neuroimaging Initiative study identified 15 additional SNPs, located within or near genes on the X chromosome—some of which other studies have implicated in neuropsychiatric disorders, brain development, or intellectual disability—as significantly associated with quantitative biomarkers of AD (Wang et al., 2023b). Collectively, the highlighted studies support contributions of sex chromosome effects on the susceptibility of women to AD, although research is needed to discern the individual contributions of sex chromosomes and sex hormones.

The Effects of Estrogens

Estrogens exert neuroprotective effects on the brain, and many studies have shown that reductions in estrogens are associated with a higher

risk of AD in women (Pike, 2017). For example, surgically induced menopause and the resulting abrupt drop in estrogens significantly increase the risk of Alzheimer dementia (Bove et al., 2014; Phung et al., 2010; Rocca et al., 2007). A diagnosis of AD in older women has been correlated with low levels of estrogens in the brain (Rosario et al., 2011). The association between AD and estrogens is also evident in mouse models. Removing the ovaries caused significant increases in beta-amyloid deposits (Jayaraman et al., 2012; Petanceska et al., 2000), and impaired behavioral performance (Carroll et al., 2007) in female animals.

Evidence also suggests that estrogens interact with ApoE, influencing dementia risk in women, especially those carrying the AD-predisposing ApoE4 gene variant that some studies have estimated occurs in approximately 55 percent of the U.S. population (Valencia-Olvera et al., 2023; Ward et al., 2012). Taken together, these findings point to a role of estrogens as a mediator of AD in women.

Studies have shown that ApoE variants appear to modify the role of estrogens in women’s risk of developing AD. ApoE2, the least prevalent variant, is associated with reduced risk (Liu et al., 2013; Serrano-Pozo et al., 2021), but ApoE4 increases the risk by 15-fold in individuals with two copies (Husain et al., 2021; Troutwine et al., 2022; Valencia-Olvera et al., 2023). Several studies have shown that the negative consequences of ApoE4 on brain structure and function before developing dementia are more severe in women compared to men (Altmann et al., 2014; Holland et al., 2013; Liu et al., 2010).

Some evidence suggests that the effect of sex hormones on AD risk may begin in utero. Some studies suggest that women who had an increased exposure to male sex hormones in utero may develop lower risks for dementia later in life (Jiang et al., 2020; Luo et al., 2020). In mice, feminization of males in utero leads to an increase in beta-amyloid, and masculinization of females in utero leads to a reduction in beta-amyloid accumulation (Carroll et al., 2010).

Other Biological Factors and Mechanisms

In the 1990s, research identified mutations in three genes involved in beta-amyloid formation that are causative factors for familial, autosomal dominant, early-onset forms of AD: amyloid precursor protein (APP), presenilin-1, and presenilin-2 (Goate et al., 1991; Lambert et al., 2023; Levy-Lahad et al., 1995; Migliore and Coppedè, 2022; Sherrington et al., 1995). Familial AD affects approximately 1 percent of individuals with the disease, with the remaining 99 percent of cases tied to susceptibility genes and environmental factors (Migliore and Coppedè, 2022; Reitz et al., 2020). Although several transgenic mouse models for familial AD have

demonstrated that females have greater deficits than males (Dubal et al., 2012; Ferretti et al., 2018), it is unknown whether similar differences are present in human familial AD studies.

ApoE4, one of the first characterized genetic susceptibility factors, occurred at a three- to fourfold higher frequency in patients with AD (Strittmatter et al., 1993) that is sex dependent (Neu et al., 2017). Clinical studies have demonstrated that among ApoE4 carriers, women are more likely than men to develop AD (Valencia-Olvera et al., 2023). Specifically, female ApoE4 carriers have increased risks for mild cognitive impairment at ages 55–70 and Alzheimer dementia aged 65–75 (Ferretti et al., 2018; Neu et al., 2017). Clear sex differences were also observed in an ApoE4 transgenic mouse model, with females exhibiting more severe cognitive impairments and faster disease progression than males (Bour et al., 2008; Ferretti et al., 2018; Reverte et al., 2012).

The strong sex dependence associated with ApoE4 suggests that its physiological effect is estrogen dependent and only emerges during menopause, when levels of estrogen fall. The cellular and physiological changes that ApoE4 modulates accumulate over many years before cognitive and behavioral alterations become detectable. ApoE4 is also linked to other conditions and diseases, including frailty, higher levels of LDL, and CVD, with specific risks depending on sex (Bennet et al., 2007; Deelen et al., 2019; Valencia-Olvera et al., 2023); the ApoE2 variant is linked to greater longevity in both sexes (Gerdes et al., 2000; Schächter et al., 1994) and lower CVD risk in women compared to men (Lahoz et al., 2001). As ApoE is the major lipoprotein component expressed in the CNS (Holtzman et al., 2012; Liu et al., 2013; Valencia-Olvera et al., 2023) and involved in cholesterol uptake and sex hormone production, it is not surprising that dysregulation of this protein will have downstream consequences on both cardiovascular and neuronal health.

GWAS have identified over 90 gene locations as significantly associated with AD risk (Lambert et al., 2023). Some of these genetic variants are highly expressed in microglia (Guerreiro et al., 2012; Jonsson et al., 2013; Lambert et al., 2023; Sims et al., 2017), supporting the idea that microglial immune activation contributes to the disease process. GWAS has revealed other genetic variants restricted to certain ancestries. For example, one variant is associated with higher AD in people with African ancestry (Kunkle et al., 2021; Lambert et al., 2023; Logue et al., 2011; Reitz et al., 2013). Sex stratification of GWAS data has identified novel sex-specific associations with specific chromosomal locations. For example, a region on chromosome seven is associated with preserved hippocampal volume, better cognitive function, and later age of onset of AD in men, although research has not identified the genes and mechanisms related to neuroprotection (Dumitrescu et al., 2019). As AD is complex and multifactorial,

genetic variants associated with risk identified from GWAS could be used to generate polygenic risk scores that are sex and ethnicity specific and may serve as a diagnostic predictor and prognostic indicator. This idea is illustrated in a study showing that a higher polygenic risk score based on 85 SNPs is linked to faster rates of tau accumulation—another hallmark of AD—in non-Hispanic White individuals (Rubinski et al., 2023).

Biomarkers for AD

Sex can affect measuring or interpreting of potential AD biomarkers via several mechanisms. For some markers, a measure such as the concentration of a protein in a bio-fluid or volume of a specific brain region, may vary by sex. For others, it is the interpretation that may vary by sex, depending on whether the biomarker is used for diagnosis or prognosis. When interpreting sex differences in biofluid biomarkers, it is important to consider sex differences in the permeability of the blood–brain barrier and the binding of biomarkers to blood proteins (Mielke, 2020). Unfortunately, many studies have considered sex a confounding variable and adjusted for it in their analyses rather than studying it as a major effect modifier (Mielke et al., 2022a).

The cerebrospinal fluid concentration of neurofilament light chain is higher in males than in female individuals across the entire life. It is a marker of large-caliber, subcortical axonal degeneration that is not specific to AD. Therefore, it has been suggested that the reference intervals should be sex specific. In contrast, female individuals have higher concentrations of total tau in the cerebrospinal fluid. Both biomarkers indicate greater neurodegeneration, but the explanation for these sex differences remains unknown (Bridel et al., 2019; Mielke et al., 2022b).

Another example is the higher concentration of blood phosphorylated tau in men, although excluding individuals with chronic kidney disease from the analysis attenuated this difference. Chronic kidney disease was more common in men in that study and may have increased the plasma levels of phosphorylated tau (Mielke et al., 2022b). This example emphasizes the importance of considering comorbidities when interpreting blood biomarkers.

As an example of sex differences in imaging biomarkers, a 2023 study in cognitively unimpaired individuals suggested that women may have higher tau measured by PET in seven brain regions in the temporal, parietal, and occipital lobes compared to age-matched men (Coughlan et al., 2023). However, the findings have not been replicated.

As AD occurs more frequently in women, several large research initiatives have explored whether sex differences related to biomarkers exist (Buckley et al., 2019; Cavedo et al., 2018; Deming et al., 2018). Such information could be useful in determining differential susceptibility of women

to AD, and sex-specific prognostic features and therapeutic responses (Toro et al., 2019).

Two common, highly validated biomarkers indicative of the pathophysiological process of AD are beta-amyloid and tau proteins. In cognitively unimpaired individuals, beta-amyloid formation precedes and modulates the phosphorylation of tau protein (Jack et al., 2013; Milà-Alomà et al., 2022), which is linked to the initial stages of cognitive decline (Hanseeuw et al., 2019; Ossenkoppele et al., 2021, 2022). However, certain individuals with high levels of beta-amyloid do not develop markers for phosphorylated tau and experience cognitive decline (Josephs et al., 2022; Nordberg, 2015; Roberts et al., 2018). A study revealed a possible mechanistic explanation, implicating astrocyte activation as a modulator of beta-amyloid-mediated phosphorylation of tau. Higher plasma levels of a surrogate marker for astrocyte activation correlated with beta-amyloid-dependent tau phosphorylation in cognitively unimpaired individuals, with a more robust association in men than women (Bellaver et al., 2023). Clinical studies have also shown that, compared to men, women exhibit higher tau-related brain pathology (Buckley et al., 2019; Palta et al., 2021) and plasma levels of phosphorylated tau that correlated with higher rates of converting to dementia in beta-amyloid-positive individuals (Tsiknia et al., 2022). Collectively, these findings support that in the early stages of the disease process, nuanced, sex-specific differences are related to mechanisms upstream of beta-amyloid’s effects on tau phosphorylation.

In the advanced stages of AD, postmortem studies have not found sex differences in beta-amyloid in affected brain regions. Measuring it by PET brain imaging and biochemical protein assays of cerebrospinal fluid samples in living patients yielded the same results (Ferretti et al., 2018). Similarly, studies using the same types of samples and methods found no consistent differences by sex in levels of neurofilament tangles—another physiologic hallmark of AD—and tau phosphorylation. However, some studies have shown that women have more neurofilament tangles in certain brain regions compared to men (Barnes et al., 2005; Salehi et al., 1998). It remains unclear why these brain regions are specifically affected in women.

Studies have also observed sex differences in atrophy rates of different brain regions at all AD stages, with faster atrophy rates of 1–1.5 percent per year in women during the early stages of AD and when full-blown AD dementia develops (Ardekani et al., 2016; Ferretti et al., 2018; Hua et al., 2010; Lin et al., 2015; Skup et al., 2011). In addition, a more complex and widespread pathological profile, measured as diffuse plaques and neurofilament tangles across four brain regions, was associated with greater than 20-fold increase in the odds of developing dementia in women compared to a threefold increase in men (Barnes et al., 2005). Other studies have shown that both beta-amyloid and total tau in cerebrospinal fluid are associated

with a greater extent of hippocampal degeneration and rapid cognitive decline in women (Ferretti et al., 2018).

Overall, multiple pathological markers of AD may correlate with disease risk and parallel the accelerated cognitive decline in women. With no way to reverse AD, a possible management approach would be to implement early detection by monitoring blood and cerebrospinal fluid biomarkers and imaging the brain for gross pathological changes. This could change the disease course by creating a larger window of opportunity for therapeutic and lifestyle interventions that could be initiated before irreversible brain pathological changes occur (van der Flier et al., 2023).

Specific Risk Factors in Women

Reproductive Milestones

Cognitive decline in women seems to occur after menopause, with marked dysfunction most evident after premature or early menopause. Studies have shown that the drastic change in cyclical levels of estrogen resulting from the surgical removal of the ovaries and the resulting early menopause makes women more vulnerable to cognitive impairments. In addition, reductions in progesterone and its metabolite, allopregnanolone, during menopause, have been linked to neuroinflammation and risk of AD (Yilmaz et al., 2019), while replacement with exogenous allopregnanolone improves pathological features (Irwin and Brinton, 2014). It is thus important to understand the neuroprotective effects of cyclical estrogens and other female sex hormones; if they are helpful in preserving cognitive function, hormone therapy should be used in women who experience premature or early menopause (Kaunitz et al., 2021).

The rate of decline is also faster in women compared to men at older ages. Several factors may account for this, with one being the propensity for older women to be socially isolated. Several groups have proposed that social and mental activities stimulate neural activity and increase synaptic connections, which help preserve cognitive function as women age.

Women are more susceptible to various AD risk factors compared to men, with differences in both the frequency and the strength of the effect of a given risk or protective factor (Rocca et al., 2014). Certain conditions occurring specifically in women, such as gestational diabetes and preeclampsia during pregnancy (Buhimschi et al., 2014; Carey et al., 2024; Fields et al., 2017; Zhang et al., 2024) and premature or early menopause (Bove et al., 2014; Karamitrou et al., 2023; Rocca et al., 2007), are also associated with higher risks of developing AD and other dementias. One study found that women with hypertension in midlife had an increased risk of dementia compared to men, despite higher prevalence of hypertension

in men, indicating that vascular and metabolic pathways contribute to this disease (Gilsanz et al., 2017; Lorius et al., 2015).

Research has established that sex hormones play a neuroprotective effect in the brain throughout a woman’s lifespan. Analysis of several studies in a recent scoping review suggest that hormonal contraceptive use at younger ages was linked to a lower risk of developing AD later in life (Gregory et al., 2023); however, more data from young women using hormonal contraceptives and longitudinal follow-up are needed. Conversely, hormonal fluctuations and the decline in estrogens in women during the menopause transition may increase the risk of brain changes associated with AD (Rahman et al., 2019). Brain imaging findings indicate a progressively increased risk in women undergoing the menopause transition and suggest that the effect of estrogens outweigh the effects of chronological aging in the female brain by several years, if not decades, before possible clinical symptoms emerge (Scheyer et al., 2018). One imaging study found that menopausal transition has significant effects on brain structure—particularly in regions involved in higher-order cognitive processes—connectivity, and energy metabolism that were not observed in age-matched men and may increase AD risk (Mosconi et al., 2021). Compared to spontaneous menopause, early and surgical menopause are associated with more severe cognitive impairment and higher risk of dementia (Jett et al., 2022). Furthermore, fluctuations in estrogen during perimenopause have been proposed to trigger detrimental changes in metabolic and inflammatory pathways that contribute to AD (Lopez-Lee et al., 2024). Additional studies are needed to understand how fluctuations in other hormones during this reproductive stage affect AD risk.

Data from the Mayo Clinic Study of Aging show that sex-specific comorbidities can predict conversion from normal cognition to mild cognitive impairment. Susceptibility factors included midlife elevated circulating lipid levels, diabetes, and hypertension in women and obesity in men (Pankratz et al., 2015). Gender also plays important roles in AD etiology. For example, lower levels of income and education were more common in women and may increase the risk of AD (Ferretti et al., 2018; Petersen et al., 2010; Rocca et al., 2014; Sharp and Gatz, 2011; Vemuri et al., 2016). Similarly, the social isolation faced by women as they age (Stahl et al., 2017) may contribute to AD, suggesting that mental and social activities that trigger new or maintain established synaptic connections are important for maintaining healthy brain function (Daly, 2023; Sommerlad et al., 2023).

Symptom Presentation and Diagnosis

Depending on the stage of the disease, women and men exhibit sex-specific tendencies related to cognitive performance and behavior.

For example, in preclinical, asymptomatic AD, women scored higher on verbal tasks, whereas men scored better in visuospatial and motor coordination tasks (Ferretti et al., 2018). Among AD patients, women scored lower than men in verbal memory, fluency, and naming tasks, even after controlling for confounding factors such as education, age, and depression (Ferretti et al., 2018). Generally, before menopause, women appear to be protected from AD compared to men, but they display higher rates of cognitive decline and brain atrophy after menopause (Ferretti et al., 2018).

Disease progression also appears to be faster in women. In the longitudinal Alzheimer’s Disease Neuroimaging Initiative study, women with prodromal AD exhibited faster cognitive decline (Petersen et al., 2010). Limited studies with small sample sizes have noted sex-specific behavioral symptoms, with women exhibiting depression, isolation, mood swings, delusions, and affective and manic symptoms, whereas men tended to display apathy, agitation, and abusive and socially inappropriate behavior (Ferretti et al., 2018; Hollingworth et al., 2006; Ott et al., 1996).

Treatment and Management

Despite better representation of both sexes in clinical trials testing drug therapies for AD, published reports typically do not include sex-based data stratification and correlations with outcomes (Ferretti et al., 2018). Applying sex-based analyses to clinical trial data could identify sex differences in drug safety and efficacy and lead to better interventions tailored to men and women (Legato et al., 2016). Currently, there is no therapy that can reverse or cure AD, although three main drug classes are commonly administered to treat AD symptoms, with several studies indicating variable responses between men and women.

Cholinesterase Inhibitors

Cholinesterase inhibitors prevent the breakdown of the neurotransmitter acetylcholine at neuronal synapses, thereby prolonging neural activity. Most clinical trials assessing adverse effects for the cholinesterase inhibitors, donepezil, rivastigmine, and galantamine, did not analyze data disaggregated by sex (Mehta et al., 2017). The two studies found no clear differences in safety and efficacy (Canevelli et al., 2017; Haywood and Mukaetova-Ladinska, 2006). In contrast, studies with another cholinesterase inhibitor, rivastigmine, did find sex differences in clinical outcomes—it delayed progression from the initial appearance of symptoms to later stages of AD in women (Ferris et al., 2009). However, its benefits were less evident in women with advanced dementia (Haywood and Mukaetova-Ladinska, 2006; MacGowan et al., 1998).

Other studies have suggested that in women, the ApoE genotype modulates the response to certain cholinesterase inhibitors, such as tacrine (Farlow et al., 1998). These differences may arise from physiological differences in the cholinergic system in men and women (Counts et al., 2011; Ferretti et al., 2018; Rhodes and Rubin, 1999) or drug interactions with sex hormones, ER 1, and other sex steroid regulatory factors (Ferretti et al., 2018). Cholinesterase inhibitors are the first-line treatment for cognitive deficits, but as efficacy wanes with the progressive neuronal loss characteristic of AD, therapies preserving neuronal function need to be developed.

Intranasal Insulin

Insulin has emerged as a promising treatment because of its pleiotropic effects on multiple organ systems involved in AD pathogenesis, including the vasculature via modulation of glucose and lipid metabolism and inflammation and the nervous system via promotion of neural connections (Kellar and Craft, 2020). Its mechanism of action in AD may also involve modulating beta-amyloid regulation and tau phosphorylation (Kellar and Craft, 2020). Some clinical trials have noted sex differences in intranasal insulin administration. In a secondary analysis of a Phase II clinical trial in patients with mild cognitive impairment, women performed better on activities of daily living, whereas men showed improvements in the delayed story recall task (Claxton et al., 2013). However, the AD-predisposing ApoE4 variant eliminated the benefits of intranasal insulin in women carriers, who exhibited worse outcomes, whereas men carriers showed improvements (Claxton et al., 2013).

Anti-Amyloid Antibody Therapy

In July 2023, FDA approved the first drug to slow AD progression. Lecanemab (Leqembi^®) is an antiamyloid monoclonal antibody that reduces amyloid plaques in brain imaging measurements (van Dyck et al., 2022). However, it remains unclear what its effect will be in the general population. It has several disadvantages, including some important contraindications, side effects, high cost, and the requirement of supervised administration and monitoring of side effects via brain imaging (Pittock et al., 2023). In addition, concern arose that the therapeutic effect may be weaker in women than men (Kurkinen, 2023; Valenzuela and Pascual-Leone, 2023).

Hormone Therapy

Hormone therapy for women, involving estrogen alone after hysterectomy or estrogen combined with a progestogen otherwise, is currently

not recommended to treat AD or mild cognitive impairment or prevent dementia in women who have experienced natural menopause (North American Menopause Society, 2022). However, it is currently recommended for women who experienced premature or early menopause. These recommendations are based on observational studies of the risk of AD or dementia in women who underwent premenopausal removal of their ovaries in the premature or early age range (Kaunitz et al., 2021).

It is important to note that different types of estrogens, as well as different routes of administration, may result in variable outcomes related to cognitive function. For example, estradiol and estrone differentially alter cognitive function in post-menopausal women (Hogervorst et al., 2000). Other studies indicate that oral estradiol is converted to estrone, while transdermal estradiol is not (Lobo and Cassidenti, 1992; Tebbens et al., 2022). Consideration of how estrogens are given as treatments has important ramifications and requires further study, especially since compared to oral estrogen formulations, transdermal estradiol has been shown to improve cognition and increase brain white matter in postmenopausal women (Kantarci et al., 2016; Wharton et al., 2011).

Disparities

Comprehensive data are lacking on the impact of AD that consider both sex and their ancestry and ethnicity. However, an analysis of administrative Medicare claims data from a large population-based cohort, which deliberately included a significant number of women and racial and ethnic groups that had been less studied, revealed a 17 percent higher risk of late-onset AD in women than men, after adjusting for age. Incidence rates for both AD and related dementias was provided for women by race and ethnicity. The age-standardized incidence rate for AD per 1,000 person-years was highest among African American (9.54), followed by Latina (7.76), Native Hawaiian (7.65), White (6.85), and Japanese American women (6.60) and lowest among Filipino women (4.58). With respect to AD and related dementias, the age-standardized diagnostic incidence rate per 1,000 person-years was highest for African American (22.92), followed by Native Hawaiian (19.33), Latina (16.79), White (16.38), Japanese American (14.76), and lowest among Filipino women (12.53) (Lim et al., 2022).

The 2020 U.S. Census indicates that the population is more racially and ethnically diverse than a decade ago. Disparities in dementia have been observed in African American, American Indian and Alaska Native, and Latino/Latina individuals. In addition, disparities affecting caregivers and families of patients with dementia are evident (Aranda et al., 2021). Caregivers of those groups have found differences in overall well-being, and studies suggest that they also had poorer health (Aranda et al., 2021).

The racial and ethnic group differences in the incidence and prevalence of AD and related dementias among women is an important research gap that needs to be addressed.

Research Gaps

Much has been learned about AD, but a number of gaps remain. Although studies have pointed to the role of sex chromosome effects on the susceptibility of women to AD, additional research is needed to discern the individual contributions of sex chromosomes and sex hormones. Identifying sex differences in a range of AD biomarkers and their interpretation is another research gap that deserves attention. As AD is no longer viewed as a disease solely associated with aging, it is important to identify and develop preventive measures to address changes in brain biochemistry throughout the life course. By the time AD symptoms are evident, it may be too late to reverse the damage that has occurred in the brain. The only possibility to help patients is to manage the disease. Therefore, it is important to catch the early manifestations by assessing social determinants of health and lifestyle factors and using biomarkers and imaging. Research on how AD differentially affects the lives of racially and ethnically minoritized women is also needed.

OSTEOPOROSIS

Osteoporosis is a chronic condition that affects women more than men and is associated with bone fragility and repeated fracture risk (Weaver et al., 2017). Hip fractures account for a large proportion of the morbidity and mortality associated with osteoporotic fractures. Biological factors, prevention, diagnosis, treatment, and disparities associated with osteoporosis in women are described next.

Biological Factors

Genetic factors substantially influence an individual’s bone strength, microarchitectural properties, and risk of osteoporosis and fractures (Mäkitie et al., 2019). Bone mineral density (BMD) is highly heritable, as demonstrated by twin studies showing that one’s genetic makeup determines up to 80 percent of its variance (Mäkitie et al., 2019). Studies have identified specific genes associated with osteoporosis risk, such as those involved in bone metabolism and regulation of calcium absorption. A family history of osteoporosis or fractures is known to increase fracture risk, indicating a strong genetic component. The most common genetic cause is osteogenesis imperfecta, an inherited disorder also known as “brittle bone disease”

(Mäkitie et al., 2019). A meta-analysis has identified several important genes involved in osteoporosis (Hasan et al., 2021) and several genes and pathways that may be possible to target to develop new treatments for osteoporosis.

Sex Chromosome Effects

Recent findings have identified a gene on the X chromosome that is associated with osteoporosis, plastin 3 (PLS3). This gene codes for a protein involved in the formation of actin bundles and also in the maintenance of bone health. This is supported by human genetic studies demonstrating that variants in this gene caused osteoporosis and osteoporotic fractures in members of 10 different families (van Dijk et al., 2013). In vivo studies using PLS3 knockout mice also led to phenotypic features of osteoporosis, and in vitro cellular experiments indicated the involvement of this gene in osteoclast formation (Neugebauer et al., 2018). More recently, research has linked PLS3 polymorphisms with early-onset osteoporosis in young individuals (Costantini et al., 2020), and mutations in this gene correlate with osteoporosis in postmenopausal women (Wolff et al., 2021).

The Effects of Estrogens

Estrogens maintain bone homeostasis by regulating the metabolic activity of the two main cell types found in bone—osteoblasts, which release bone growth factors, and osteoclasts, which promote bone resorption (Biason-Lauber and Lang-Muritano, 2022). Estrogens preserve bone mass by interfering with the action of inflammatory cytokines that promote osteoclast development (D’Amelio et al., 2008; Jilka et al., 1992; Pietschmann et al., 2016; Weitzmann et al., 2002) and inducing apoptosis, or programmed cell death, of osteoclasts (Kameda et al., 1997). The gradual decline of estrogens during menopause may therefore increase osteoclast activity, leading to higher bone turnover and breakdown and potentially osteoporosis.

As explained in Chapter 2, estrogens act on three different receptors (ER-α, ER-β, and GPR30) to affect numerous cell and tissue types. Given the importance in mediating the effects of estrogens, researchers proposed that ERs could determine osteoporosis risk (Gennari, 2005). Studies have found that genetic variants in two of these receptors occur in women from different racial and ethnic backgrounds, with genetic variants in ERα associated with lower BMD and increased risk of osteoporosis (Gennari, 2005). Although the mechanism by which this occurs is unknown, one group proposed that these variants alter estrogen sensitivity (Khosla et al., 2004). Other studies support the idea that ERα is linked to osteoporosis,

including one in which a man with an ERα mutation had features of osteoporosis (Smith et al., 1994) and in an ERα knockout mouse model that exhibited impaired bone growth (Lindberg et al., 2001; Vidal et al., 2000).

Other Biological Factors and Mechanisms

Osteoclasts and osteoblasts are involved in bone remodeling; they resorb old bone and then form new bone through a well-coordinated mechanism regulated by different factors, such as systemic hormones, local growth factors, cytokines, chemokines, adhesion molecules, and extracellular matrix proteins (Kenkre and Bassett, 2018). These factors are secreted by bone surface cells and produced by osteocytes in the bone matrix, which are the cell type that, together with bone lining cells, is involved in bone remodeling (Robling and Bonewald, 2020).

In addition to hormonal changes, other factors involved in osteoporosis include aging, oxidative stress, and inflammation. Evidence is increasing that oxidative stress is elevated with age, and that the decline in estrogens during menopause is often accompanied by increases in reactive oxygen species and proinflammatory cytokines. The production of reactive oxygen species occurs in parallel with inflammation and is thought to mediate the inflammatory process. These factors influence the functional activity of bone cells and consequently, bone homeostasis. Both oxidative stress and inflammation contribute to the development of osteoporosis by preventing the differentiation of bone-forming osteoblasts, inducing the differentiation and activity of bone-degrading osteoclasts, and enhancing apoptosis of osteocytes. In addition, there are bidirectional influences between certain microRNAs(miRNA) with oxidative stress and inflammation. Reactive oxygen species can alter miRNA expression, and conversely, miRNAs can regulate reactive oxygen species production and inflammatory processes (Iantomasi et al., 2023).

Researchers have used various -omics technologies, including genomics, transcriptomics, epigenomics, proteomics, and metabolomics, to identify the biomolecules involved in the pathogenesis of osteoporosis. The use of multiple -omics approaches can yield a better and more integrated understanding of the complex pathogenesis that could potentially lead to advancements in personalized disease risk prediction, intervention, and treatment, as well as drug development or repurposing (Yang et al., 2020b).

Specific Risk Factors in Women

Reproductive Milestones

Bone mass accumulates progressively from birth, through childhood, and into adolescence, typically reaching 90 percent of peak by late adolescence (Loud and Gordon, 2006; Pitts, 2022). The greatest increase, around

26 percent, occurs during the 4-year pubertal period coinciding with peak growth rate in height (Loud and Gordon, 2006). Genetic factors are crucial, accounting for about 60–80 percent of achieving peak bone mass (Loud and Gordon, 2006; Pitts, 2022). Other factors include hormonal status, nutritional status, physical activity level, overall health, medication use, and illicit substance use (Loud and Gordon, 2006; Pitts, 2022).

Estrogens play a significant role during the pubertal growth spurt by rapidly increasing bone mass and during adolescence to attain maximal peak bone mass (Pitts, 2022). Estrogens, acting via ERs on bone, promotes osteoblast survival while exerting a pro-apoptotic effect on osteoclasts. This results in a positive net balance favoring bone growth. Other hormones influential in this process include insulin-like growth factor, growth hormone, and sex hormones. Adolescent girls with untreated hormonal deficiencies are at risk of low bone density (Pitts, 2022).

Adequate calcium intake is critical for optimal mineralization of bone and cartilage matrix. During the pubertal growth spurt, intestinal calcium absorption increases (NIH, 2024). Vitamin D is also essential, stimulating intestinal calcium absorption and regulating calcium homeostasis (Fleet, 2022; Loud and Gordon, 2006). Adolescents lacking adequate dietary vitamin D should take a supplement (Society for Adolescent Health and Medicine, 2013). Prolonged deficiency of calcium and vitamin D can lead to rickets, particularly concerning in cases of malabsorption where supplementation is advised (NIH, 2024).

Pregnancy constitutes a natural physiological process linked with bone loss (Scioscia and Zanchetta, 2023). Although the precise pathophysiology remains unclear, researchers have hypothesized that molecular factors released during lactation, excessive calcium loss in breast milk, diminished postpartum levels of estrogen, and genetic factors may contribute. Additional risk factors include a family history of the condition, reduced physical activity, smoking, poor nutritional status with low BMI, inadequate intake of calcium and vitamin D, pre-existing factors predisposing to low bone density, and medications known to induce bone loss. Pregnancy- and lactation-associated osteoporosis is rare, but it can lead to multiple vertebral compression fractures. Clinicians should be suspicious with women experiencing mid- to low-back pain during the third trimester or early postpartum phase. Evaluation typically involves thoracic and lumbar spine X-rays or MRI, followed by dual-energy X-ray absorptiometry (DXA). Management strategies include discontinuing breastfeeding, optimizing weight-bearing activities, ensuring adequate calcium and vitamin D intake, managing pain, and considering antiosteoporosis medications (Scioscia and Zanchetta, 2023).

Menopause plays a role in the loss of bone density. Data from SWAN showed that the rates and cumulative amounts of bone loss were greatest during perimenopause, from 1 year before through 2 years after the final menstrual period (Greendale et al., 2019). Bone loss 2–5 years after the

final menstrual period was lower than that observed during the transition. Diminishing levels of estrogens during perimenopause lead to accelerated bone resorption relative to bone formation, predisposing to bone loss and elevating the risk of osteoporosis and fractures in later years. An elevation in levels of the gonadotropin, follicular stimulating hormone, which occurs in response to the decline in estrogen levels during menopause, appears to contribute to bone loss (Liu et al., 2017). Other hormones secreted by the ovaries—inhibin, activin, and relaxin—may also play functional roles in bone formation and resorption (Mills et al., 2021).

BMD starts to decline swiftly approximately 1 year before the final menstrual period and continues for about two years, followed by a slight reduction in bone loss thereafter (Karlamangla et al., 2018). For 1 year before through 2 years after the final menstrual period, women lost 7 and 6 percent of their lumbar spine and femoral neck BMD, respectively (Greendale et al., 2012). Slower BMD loss was observed in African American women, whereas faster BMD loss was observed in Japanese and Chinese women (Greendale et al., 2012).

Lifestyle Behavioral Factors

Physical activity and dietary patterns are two well-recognized lifestyle behaviors that are associated with osteoporosis prevention (Mortensen et al., 2021; Pinheiro et al., 2020). With respect to diet, nutrients such as vitamin D, phosphorus, riboflavin, calcium, protein and fiber intake, and folate from fruits and vegetables are known to contribute to the prevention of osteoporosis and reduction in fractures in postmenopausal women (Ilesanmi-Oyelere and Kruger, 2020). A systematic review found that engaging in physical activity, specifically in higher doses and of multiple types, including resistance training, were beneficial for the prevention of osteoporosis in postmenopausal women (Pinheiro et al., 2020). Larger effects were seen for improvements in lumbar spine and hip BMD (Pinheiro et al., 2020). Sleep fragmentation has an adverse impact of bone health (Swanson, 2021). An increasing body of research indicates that abnormal sleep duration and disturbances in circadian rhythms may impact bone health, including the susceptibility to osteoporosis and fractures. The effects of disrupted sleep may be particularly pertinent for individuals with certain medical conditions predisposing them to both sleep disturbances and skeletal fragility, such as diabetes (Swanson, 2021).

Initial Fracture and Risk of Subsequent Fractures

Among postmenopausal women aged 50–75 at baseline in the WHI Study, every type of initial fracture, even non-hip and non-vertebral fracture, was associated with increased risk of every other type of subsequent fracture.

For example, initial knee fracture was associated with increased risk of subsequent hip fracture, and these associations are evident even in younger postmenopausal women in their 50s (Crandall et al., 2021). Postmenopausal women with incident wrist fracture had elevated risk of experiencing an additional clinical fracture within 10 years after wrist fracture (Crandall et al., 2015).

Symptom Presentation and Diagnosis

The operational definition of osteoporosis is based on BMD assessment. The most commonly used screening test used is lumbar spine and hip DXA, which major drug therapy studies have also used to determine enrollment eligibility (USPSTF et al., 2018). In clinical practice, osteoporosis is diagnosed in individuals with BMD T-score¹⁵ of less than or equal to −2.5 at either the lumbar spine or hip (USPSTF et al., 2018). For each 1–standard deviation decrease in BMD measurement, there is an estimated 1.6- to 2.6-fold higher fracture risk (Marshall et al., 1996; Silva et al., 2022). The exact risk depends on various factors including the anatomical location of fracture, prior occurrence of fracture, and number of osteoporotic sites (Fink et al., 2008).

Osteoporosis is significantly underdiagnosed in older U.S. women. Approximately 26 percent of women ages 65–79 years do not receive a DXA test (Gillespie and Morin, 2017). No general agreement exists on the appropriate time to begin DXA for early menopausal women (Jiang et al., 2016).

Currently, a diagnosis of osteoporosis is based on the occurrence of fragility fractures or low BMD scores as assessed by DXA (Lenchik et al., 2018). Numerous professional organizations, including the Office of the Surgeon General, the U.S. Preventive Services Task Force, ACR, American College of Radiology, International Society for Clinical Densitometry, and Bone Health and Osteoporosis Foundations recommend the use of DXA to diagnose osteoporosis (Lenchik et al., 2018; Warriner et al., 2014). However, in the U.S., only approximately one-third of women aged 65 years and older undergo DXA testing (Warriner et al., 2014).

Imaging Bone Density and Quality

Computed Tomography (CT) Scan–BMD Correlation

Novel approaches for identifying patients at risk for osteoporotic fracture who are candidates for pharmacologic therapy are needed.

___________________

¹⁵ A T-score is the difference between an individual’s BMD and 0, the BMD of a healthy young adult. The lower the T-score, the higher the risk of bone fracture. Health bone produces a T-score of −1 or higher.

“Opportunistic CT” is one such approach that utilizes results from routine CT scans to screen for additional body composition information, such as the attenuation of bone. Regardless of the primary indication, CT scans of the chest, abdomen, pelvis, and spine may thus yield useful information regarding a patient’s risk for osteoporosis, minimizing the burdens related to additional medical diagnoses and interventions, such as examination time and costs (Boutin and Lenchik, 2020; Lenchik et al., 2018).

Biomechanical computed tomography (BCT) is another method used to diagnose osteoporosis and assess fracture risk and measures both bone strength and BMD at the hip or spine. BCT may be helpful for patients without a recent DXA who are undergoing or have previously undergone CT scanning (Keaveny et al., 2020). Advancements in machine learning will provide future opportunities to make it easier to automate measurements and mine data from CT images and could potentially lead to substantial cost savings and improved care for patients at risk for osteoporosis (Lenchik et al., 2018).

Trabecular Bone Score (TBS)

Although DXA provides information about the bone density, TBS provides an estimate of the bone quality. TBS is a software program that integrates information from DXA images of lumbar spine and used as a tool to indirectly assess the microarchitecture and mechanical properties of bone (Goel et al., 2024; Shevroja et al., 2023). TBS measurements have been integrated into the FRAX^® tool¹⁶ to enhance prediction of fracture independent of BMD (Goel et al., 2024; Shevroja et al., 2023). This method may be useful in determining when to initiate pharmacologic treatments in patients as it can estimate 10-year fracture probability (Goel et al., 2024).

CT Scan–TBS Correlation

The World Health Organization recommends DXA as the gold standard technique for diagnosing osteoporosis. DXA is the most widely used modality for measuring BMD, which is a quantitative measurement of bone used to predict bone strength and fracture risk (Shevroja et al., 2021). However, DXA cannot differentiate trabecular bone from cortical BMD (Lee et al., 2023b).¹⁷

Given the limitations of DXA, researchers have investigated tools to assess bone architecture as well as bone density. One approach is to use DXA images of the lumbar spine to evaluate pixel gray-level texture to

___________________

¹⁶ “The FRAX^® tool has been developed to evaluate fracture risk of patients. It is based on individual patient models that integrate the risks associated with clinical risk factors as well as BMD (BMD) at the femoral neck” (https://www.fraxplus.org/).

¹⁷ Cortical bone is a dense, low-porosity, and less metabolically active tissue, and trabecular bone is a honeycomb-like network with a larger remodeling area and higher turnover rate (Clarke, 2008).

define a TBS, which is an indirect index and qualitative parameter of trabecular microarchitecture.

Quantitative computed tomography (QCT) has some diagnostic advantages over DXA. QCT allows true volumetric measurements of the lumbar spine independent of body size and identifies trabecular and cortical compartments (Link and Kazakia, 2020). Furthermore, QCT is known to be more sensitive than DXA for osteoporosis and has advantages over DXA for rheumatological disorders, such as ankylosing spondylitis (Link and Kazakia, 2020). Nevertheless, QCT has some limitations as a primary diagnostic tool for osteoporosis, given its relatively high cost and larger radiation exposure (Link and Kazakia, 2020). To overcome these limitations, investigators have attempted opportunistic screening for osteoporosis using the Hounsfield unit, a relative quantitative measurement of radio density used by radiologists to interpret CT images (DenOtter and Schubert, 2024). Its value is positively correlated with material density and compressive strength (Schreiber et al., 2011). Opportunistic screening has the advantages of not requiring additional imaging, radiation exposure, or appointments (Boutin and Lenchik, 2020; Lenchik et al., 2018; Schreiber et al., 2011).

MRI–TBS Correlation

Recent attempts have been made to diagnose osteoporosis using MRI-based measurements of the fat fraction of bone (Chang et al., 2020). In the bone marrow, osteoblasts and adipocytes—fat-storing cells—arise from common progenitor cells. Osteoporosis has predominance of differentiation toward adipocytes in progenitor cells, and differentiation toward osteoblasts decreases, leading to reduced bone formation. As a result, the increase in bone marrow fat becomes an indirect indicator of osteoporosis and a target for diagnosis. MRI-based measures of the fat fraction of bone are a reliable tool for evaluating vertebral bone marrow fat quantification without radiation exposure.

Despite attempts to measure BMD with various tools, none have been made to identify a relationship with TBS. Only one study in an Asian population has been published (Lee et al., 2023b), highlighting the need for evaluation in other racial and ethnic populations.

Vertebral Fracture Assessment

Vertebral fractures are commonly observed in postmenopausal women with osteoporosis (Yang et al., 2020a). Vertebral fractures cause substantial morbidity, with detrimental changes that include height loss, progressive and exaggerated forward rounding of the upper back, impaired movement and balance, and chronic pain. It is also associated with higher mortality, with estimates that one-third of patients will die within 5 years following hospitalization. Evidence has shown that prior experience of vertebral

fractures is a predictor of subsequent fractures. Most vertebral fractures are asymptomatic and only approximately one-third of vertebral fractures are identified on radiographs (Yang et al., 2020a).

Spine X-rays are primarily used to identify vertebral fractures, but lateral DXA of the entire spine for vertebral fracture assessment (VFA) is a convenient, low-radiation, and low-cost option to detect vertebral fractures (Lems et al., 2021). Studies have shown that VFA identifies vertebral fractures in one-third of asymptomatic women, although women with prevalent vertebral fractures often do not have osteoporosis (Yang et al., 2020a). VFA can also identify women at high risk of fracture, potentially leading to treatment.

Lateral spine imaging with standard radiography or densitometric VFA is indicated when a T-score is less than −1.0 and of one or more of the following is present (ISCD, 2019): women ages 70 and older and men ages 80 years and older; historical height loss greater than 4 centimeters or 1.5 inches, and self-reported but undocumented prior vertebral fracture.

Microindentation

Impact microindentation is a technique to assess bone material properties of cortical bone at the tibia in a transcutaneous, microinvasive manner. It is used increasingly in human studies evaluating the contribution of tissue material properties to bone fragility and approved for use in the clinic in the U.S. and Europe (Schoeb et al., 2023). Efforts to make this procedure less invasive may result in increased use in clinical settings.

Treatment and Management

Ensuring adequate calcium and vitamin D intake is essential for people with osteoporosis. Clinical trials of drug therapy demonstrated its efficacy in conjunction with calcium and vitamin D supplementation. Drug therapy decreases the risk of fracture and is effective for preventing both the initial fracture in persons with osteoporosis (as diagnosed using BMD criteria), subsequent fractures, and secondary fractures (among persons who already experienced a fracture).

FDA-approved prescription drug therapies include antiresorptive drugs, drugs that promote bone formation, and drugs that do both (Ensrud and Crandall, 2024). Antiresorptive drugs include bisphosphonates, such as alendronate, risedronate, ibandronate, and zoledronic acid; denosumab, an inhibitor of osteoclast development; and raloxifene, a selective ER modulator. Estrogens are a known antiresorptive agent approved to treat bone loss and prevent progression to osteoporosis (Eastell et al., 2019). Research is needed to better understand which patients could safely benefit from temporary cessation of bisphosphonate medication followed by reinitiation, the appropriate duration and monitoring of a bisphosphonate drug

holiday, and reasons for the higher risk of bisphosphonate-associated but rare serious harms in Asian versus White postmenopausal women.

Treatments that promote bone formation include recombinant parathyroid hormone and abaloparatide, a parathyroid-hormone-related peptide analog. Statins, approved for treating high cholesterol levels, may prove to be effective for increasing BMD and treating osteoporosis, with a dose-dependent therapeutic effect (Lin et al., 2018; Wang et al., 2016). However, some studies suggest that high-dose statin therapy can increase the risk of osteoporosis (Leutner et al., 2023).

People with osteoporosis are substantially undertreated (Gillespie and Morin, 2017; Jha et al., 2015; Kim et al., 2016; Singer et al., 2023). Bisphosphonate use within 6 months after hip fracture hospitalization has decreased from 15 percent in 2004 to 3 percent in 2013 (Kim et al., 2016). Among women enrolled in private commercial or Medicare Advantage plans who experience a new hip fracture, those aged 80 years and older are one-third less likely to use recommended services within 6 months compared with those ages 50–79. The most recent examination of commercial and Medicare medical claims data shows that even in 2023, 89 percent remain untreated for at least 1 year after fracture (Singer et al., 2023).

Treatment After Denosumab

Denosumab is a monoclonal antibody therapy approved in 2010 for patients with osteoporosis. It is a potent antiresorptive medication supported by 10 years of clinical trial data that demonstrate its safety and efficacy in reducing fracture risk (Bone et al., 2017; Kendler et al., 2022). Another type of monoclonal antibody therapy is romosozumab, which has also been used to treat postmenopausal women at high risk of fracture (Krupa et al., 2024). Compared to bisphosphonates, another drug class used in the treatment of osteoporosis, denosumab leads to continued gains in bone density (Kendler et al., 2022). Despite aging of the study population, nonvertebral fracture rates were lower after 4–10 years of treatment with denosumab compared to the initial 3 years of therapy (Kendler et al., 2022).

Treatment with denosumab for up to 10 years was associated with a favorable benefit risk profile when comparing fractures prevented per skeletal adverse event, such as osteonecrosis, a condition in which cells in the jawbone die, and atypical hip fracture. In addition, the frequency of adverse events, including infection and malignancy, remained low over time in the aging study population (Bone et al., 2017; Kendler et al., 2022).

If denosumab is discontinued, the gains in BMD may be lost and thus administration of a different class of anti-osteoporosis medication, such as bisphosphonate, can help prevent such bone loss, minimizing future risk of fracture (Kendler et al., 2022). Timing that switch is challenging,

and limited evidence is available. While transition to oral alendronate has been successful (Kendler et al., 2020), most postmenopausal women have contraindications to oral bisphosphonates and would benefit from intravenous infusions of bisphosphonates, such as zoledronic acid (Horikawa et al., 2015; Kendler et al., 2020). The timing of switching to other therapies is crucial for maintaining BMD gains from denosumab (Kendler et al., 2020).

Research is needed to understand how to treat osteoporosis, including the optimal duration of drug therapy, long-term risks and benefits of drug therapy, optimal frequency of BMD monitoring during drug therapy, anti-fracture benefits and harms of drug therapies in women aged 50–64, who are underrepresented in clinical trials compared to women 65 and older, and best means of selecting candidates for initial therapy with bone formation promoting agents instead of antiresorptive drug therapy.

Bone Turnover Markers

Biomolecules released during the bone remodeling cycle are used as bone turnover markers, which are typically detected in blood or urine samples. They are usually secreted by osteoblasts during bone formation or osteoclasts during bone resorption (Vasikaran et al., 2023) and are useful for monitoring the effects of oral therapy in patients with osteoporosis. Further work is needed in exploring their use to determine optimal duration of treatment with antiresorptive therapies, such as bisphosphonates and denosumab and that will yield the most benefit with minimal adverse effects. Large-scale fracture risk prediction studies assessing how bone turnover markers can contribute to risk prediction and consideration of its possible inclusion as a factor in risk calculators, such as FRAX, may help better predict risk. Improving commercial assays that detect bone turnover markers will enable the collation and comparison of data from different studies. Efforts to consolidate methods of detecting bone turnover biomarkers may lead to the uniform application of decision limits and treatment targets in clinical guidelines (Vasikaran et al., 2023).

Physician/Patient Education Tools

Educating patients about the importance of treating osteoporosis is essential for initiating treatment and improving compliance, leading to better outcomes in reducing fractures. However, given the constraints in medical practice, providing detailed education for a single condition is challenging. With the rapid advancement of digital technology, learning has shifted from verbal discussions and paper handouts to digital sources, offering accessibility anytime, anywhere, but the Internet may not always provide reliable information in a patient-friendly format, and social media

discussions often contain conflicting information based on personal experiences. It is also essential to provide culturally competent materials and information that are offered to patients in several languages (Lewiecki and Erb, 2022).

Disparities

Studies examining racial and ethnic disparities have focused on outcomes such as increased mortality (Wright et al., 2020). One recent study, for example, found that Black women experienced a greater 1-year mortality from osteoporosis compared to White women (Wright et al., 2020). After adjusting for age in Medicare beneficiaries, Black women had significantly higher risk of mortality following 1 year of encountering fractures of the femur, hip, humerus, or radius/ulna; debility based on new placement in long-term nursing 1 year following hip fracture; and becoming newly eligible for Medicaid because of limited finances for all fracture types compared to White women (Wright et al., 2020).

Equitable access to bone scans is a major determinant of outcomes associated with osteoporosis and has gender and racial bias (Courtney et al., 2023; Ruiz-Esteves et al., 2022). Black and African American individuals, for example, were more likely to be undiagnosed compared to White individuals based on data from the University of Michigan Health and Retirement Study (Courtney et al., 2023). Although Asian women have similar or lower BMD than White women, they are overrepresented in treatment for osteoporosis and bisphosphonate therapy, conferring a greater risk of atypical femur fractures (Lo et al., 2023). Furthermore, rural communities face inequitable access to DXA due to the reduction in insurance coverage for DXA and decreased availability of screening (Khosla et al., 2023).

BMD is an important risk factor for osteoporosis, and studies have found differences in Black, Asian, and Hispanic populations that vary based on anatomical site (Wu and Dai, 2023). SWAN measured bone density and found that Black and White women had similar rates of bone loss at the tibia, while Chinese and Japanese women had lower BMD values than White women (Cauley et al., 2021; Johannesdottir et al., 2022). There is variation among racial and ethnic groups based on NHANES data for women; non-Hispanic other (Asian) women had lower femoral neck BMD, followed by non-Hispanic White, Hispanic, and non-Hispanic Black women (Wu and Dai, 2023). Similarly, claims data from Kaiser Permanente found that femoral neck BMD was lower among women with Chinese, Filipino, or Japanese ethnicity compared to White women, particularly among older women (Lo et al., 2020); data for Hispanic women and non-Hispanic Black women were not included (Lo et al., 2020). Suggested mechanisms for lower BMD in Asian women have to do with genetic variants of vitamin D receptors

(Wu and Dai, 2023). While more recent NHANES and Kaiser Permanente datasets did not capture risk among American Indian and Alaska Native individuals, an earlier study based on NHANES data found that BMD among Navajo women was low, and they had a higher prevalence of osteoporosis (Miller et al., 2016). Lifestyle factors, including dietary patterns and physical activity, are important targeted areas for intervention to improve BMD in racial and ethnic groups.

Research Gaps

To address and understand disparities in osteoporosis and fracture outcomes, comprehensive research is needed. Disparities exist in postfracture outcomes influenced by social risk factors and fracture rates varying by self-reported race and ethnicity. This underscores the importance of multiethnic longitudinal studies to examine changes in BMD and identify risk factors particularly affecting racially and ethnically minoritized populations.

One critical area of research is determining the most effective risk assessment and screening strategies for young postmenopausal women aged 50–64. They may not have osteoporosis but are at an elevated risk of fractures. Questions about the frequency of osteoporosis screening and optimal intervals for BMD testing in postmenopausal women remain unanswered due to the absence of a standardized approach for assessing BMD risk in women (Gourlay et al., 2015). Moreover, strategies for treating BMD levels that are below normal but not yet in the osteoporotic range need to be developed. This includes determining the right settings and timing for initiating drug therapy in individuals with T-scores between −1 and −2.5, who do not meet the criteria for osteoporosis.

Additionally, the effectiveness of physical activity as a treatment requires further investigation to ascertain any ideal type, intensity, and frequency of exercise that can reduce fracture risk. The research must also explore the systemic barriers to accessing treatment post-fracture. These barriers may include systemic issues within health care, discrepancies among providers regarding who is responsible for initiating and monitoring drug therapy, and other factors at the health system or individual levels, including patient or caregiver characteristics. Addressing these gaps is essential for advancing care and outcomes for those at risk of or living with osteoporosis.

SARCOPENIA

As mentioned in Chapter 4, sarcopenia is primarily defined by a decline in muscle strength (Ooi and Welch, 2024). It is associated with detrimental health consequences: frailty, physical disability, risk of falls, loss of autonomy, poor quality of life, morbidity, and mortality (Santilli et al., 2014). Knowledge about it has evolved, and major insights include that it can

begin early in life, not just with aging; is a muscle disease; and is associated with low muscle quantity and quality (Cruz-Jentoft et al., 2019). In women, it contributes significantly to frailty.

Biological Factors

Sex Chromosome Effects

No evidence exists of sex chromosomal effects on the pathophysiology of sarcopenia. Although GWAS identified potential genes that are associated with this condition, these genetic polymorphisms are all located on autosomal chromosomes, not sex chromosomes (Jin et al., 2022; Jones et al., 2021).

The Effects of Estrogens

The most convincing evidence to support a role for estrogens in the muscular degenerative process of sarcopenia is found in studies of menopausal women. In perimenopause, the gradual decline in estrogens is thought to precipitate changes that lead to a decrease in the quality, mass, and function of muscle. These changes are attributed to several events that are normally regulated by estrogen prior to menopause, including reductions in muscle satellite cells (Collins et al., 2019) and a switch to a proinflammatory immune profile (Cioffi et al., 2002; Geraci et al., 2021).

Specific Risk Factors in Women

Reproductive Milestones

An acceleration of loss of muscle mass is observed with the menopausal transition and menopause (Kim and Kim, 2020). Studies have reported that lean body mass decreases by 0.2 percent a year and fat mass increases by 1.7 percent per year during the menopausal transition (Greendale et al., 2019). Hormonal changes during menopause represent a major risk factor for sarcopenia. For example, decline in estrogens in postmenopausal women is associated with a decline in muscle mass, impaired muscle function, and the development of sarcopenia (Buckinx and Aubertin-Leheudre, 2022). Women who experience premature ovarian insufficiency (<40 years) or early menopause (<45 years) demonstrate lower muscle mass and impaired muscle function compared to women experiencing menopause at normal age (>45 years), suggesting higher risks for sarcopenia in these groups during the post-menopausal years (Divaris et al., 2023).

Population-based studies support the connection between declines in estrogens during menopause and muscle degeneration. One large population-based

cohort study of older English adults recruited from the English Longitudinal Study of Ageing showed that women had a 20 percent higher risk of sarcopenia than men (Yang et al., 2019). Another research group using data from the Estrogenic Regulation of Muscle Apoptosis study also demonstrated that menopausal status was a significant predictor of decreases in lean and muscle mass in women ages 47–55 years old (Juppi et al., 2020). As estrogens have been shown to promote satellite cell growth in muscle tissues by binding to its receptor (Collins et al., 2019) and curb muscle damage triggered by inflammatory signals (Tiidus et al., 2013), the decline in estrogens during menopause would thus contribute to sarcopenia. It is important to prevent and manage sarcopenia in postmenopausal women through nutrition, protein intake, vitamin D intake, and physical activity (Buckinx and Aubertin-Leheudre, 2022).

Nutrition and Physical Activity

The complex interconnection of age-related changes in physical activity and nutritional status between men and women contribute to gender differences in muscle quality and quantity, affecting the risk of sarcopenia (Ning et al., 2021). The complexity of these associations among physical activity, weight, and muscle quality/quantity was demonstrated in the Louisiana Osteoporosis Study (Du et al., 2022), which showed that in both older women and men, only those exercising 3 or more times a week were less likely to be obese and develop sarcopenic obesity (the combination of sarcopenia and obesity) compared to those who were not (Du et al., 2022). Exercising 1–2 times per week in middle-aged women and male adults and any regular exercise in older women and male adults were not related to less sarcopenia (Du et al., 2022). These findings illustrate the importance of nutrition and dietary interventions in sarcopenia prevention and management in middle-aged and older women. A systematic review reported the beneficial effects of protein, vitamin D, vitamin D and magnesium, and fish oil on muscle protein synthesis, muscle strength, and muscle function in addition to exercise (Thornton et al., 2023). Peri- and postmenopausal women can prevent and even combat the negative health outcomes of sarcopenia through healthy lifestyle and dietary habits, regular physical activity, and hormone therapy (Buckinx and Aubertin-Leheudre, 2022). Another study showed that food insecurity was associated with greater odds of sarcopenia based on NHANES data (Lynch et al., 2022).

Obesity

Obesity is a risk factor for sarcopenia. The relationship between insulin resistance and age-related muscle loss is an important factor in sarcopenic obesity (defined as low skeletal muscle mass with low muscle function)

(Donini et al., 2022). Sarcopenia and CVD also have a bidirectional relationship. Cardiac patients, for example, are in a state of chronic inflammation, malnutrition, and physical inactivity and at the highest risk of accelerated skeletal muscle loss and thus, sarcopenia (Damluji et al., 2023). This supports the understanding that multiple factors are involved in the development of sarcopenia (see Figure 6-4).

Sarcopenic obesity is of significant concern in the United States. In one study of community-dwelling older obese adults, 18 percent had sarcopenic obesity (Dondero et al., 2023). It is linked to geriatric vulnerabilities such as frailty—the same study found that individuals with sarcopenic obesity had a 6.4 times higher odds of frailty compared to those who were nonsarcopenic obese (Dondero et al., 2023).

Condition Presentation and Diagnosis

In 2019, the European Working Group on Sarcopenia in Older People (EWGSOP) developed a revised definition that focuses on muscle strength as the primary diagnostic hallmark for sarcopenia, and that assessments of grip strength or a chair stand test are best ways to assess muscle strength and physical performance and severity associated with sarcopenia (Cruz-Jentoft et al., 2019; Ooi and Welch, 2024). DXA is the most common approach for measuring muscle quantity because of its wide availability, but MRI and CT are considered the gold standards for noninvasive measurement of muscle quantity (Cruz-Jentoft et al., 2019). Gait speed is a quick, safe, and highly reliable test that can predict adverse outcomes related to sarcopenia, such as disability, cognitive impairment, institutionalization, falls, and mortality (Abellan van Kan et al., 2009; Guralnik et al., 2000).

According to EWGSOP, low muscle strength suggests that an individual probably has sarcopenia. Diagnosis is confirmed with documentation of low muscle quantity or quality; if those two criteria and low physical performance are apparent, sarcopenia is considered severe. The revised EWGSOP criteria recognizes that “techniques for evaluating muscle quantity are available in many but not all clinical settings. As instruments and methods to evaluate muscle quality are developed and refined in the future, this parameter is expected to grow in importance as a defining feature of sarcopenia” (Cruz-Jentoft et al., 2019). EWGSOP recommends the five-question SARC-F questionnaire (Malmstrom et al., 2016) to generate self-reports of signs that characterize sarcopenia. Sarcopenia is considered “primary,” or age related, when no other cause is evident but aging and “secondary” given one or more other causes (Cruz-Jentoft et al., 2019).

Treatment and Management

Especially in postmenopausal women, management of sarcopenia includes maintaining proper nutrition, including optimal dietary intake of protein and vitamin D, engaging in physical activity; and using hormone therapy dependent on the individual dose and level of physical activity (Buckinx and Aubertin-Leheudre, 2022). Whole body vibration has been shown to improve features of sarcopenia—specifically, impaired muscle strength and function in postmenopausal women, and has been proposed as an alternative to standard exercises in those with severely impaired functionality (Guedes De Aguiar et al., 2023; Rogan et al., 2015); however, further studies are needed for recommendation as an alternative treatment. Research efforts identifying new drug therapeutics for sarcopenia are ongoing and potential candidates include inhibitors of the myostatin signaling pathway, factors such as irisin released by muscle during exercise (Falsetti et al., 2024), and natural compounds enriched in certain fruits and vegetables, such as ursolic acid and tomatidine (Kwak and Kwon, 2019). No drugs are approved to treat or manage sarcopenia, and although some drugs have improved muscle mass or strength, they did not result in clinically relevant improvements on physical performance (Rolland et al., 2023).

Sarcopenia and Frailty

Frailty, like sarcopenia, increases with age. The two conditions overlap and diagnosing them involves the same physical measures, such as gait speed, grip strength, and muscle mass. Overlap exists between the classic frailty definition and the EWGSOP clinical diagnostic criteria. The Frailty Index reveals how frailty can be quantified based on five criteria: unintentional weight loss of 10 pounds or more in the past year, self-reported exhaustion,

weakness measured by grip strength, slow walking speed, and low physical activity (Fried et al., 2001; Rockwood and Mitnitski, 2007).

A major factor contributing to frailty in women is the deficiency in estrogens at menopause that leads to reductions in bone mass (Khadilkar, 2019). Aging-related hormonal changes also play a role. For example, the amplitude of normal pulsed release of pituitary growth hormone declines with aging (Bartke and Darcy, 2017), leading to decreased liver production of insulin growth factor-1 (Elmlinger et al., 2004; Juul et al., 1994), the hormone that stimulates bone and muscle production. Another phenomenon in older adults that contributes to frailty is aging-related appetite loss that results in reduced intake of protein, calcium, vitamin D, and other nutrients essential for bone and muscle formation and in weight maintenance (Landi et al., 2016; Molfino et al., 2021; Picca et al., 2022). Furthermore, an age-related, generalized decline in physical activity contributes to losses of bone and muscle. According to the 2020 NHIS, the adherence rate to exercise recommendations, was 22 percent in men and 17.2 percent in women aged 65 and older (CDC, 2020; HHS, 2018).

Sex differences in frailty prevalence or risk have been well studied, with robust evidence that supports that women are frailer than men at all ages, and this is evident in a range of populations with the Frailty Index (Kane and Howlett, 2021). A meta-analysis consistently showed higher scores in women compared to men at any age (Gordon et al., 2017). However, the underlying mechanisms of the higher prevalence in women despite being at lower risk for frailty mortality are poorly understood. Studies have suggested that sociobehavioral factors may play a role, as women are more likely than men to report health concerns and seek medical help (Kane and Howlett, 2021). Moreover, data from preclinical models indicate that, at least in some studies, female mice and dogs do have higher levels of frailty than males, suggesting a biological contribution to sex differences in prevalence (Bisset and Howlett, 2019).

Inflammation and immune cell changes, hormone dysregulation, cell senescence, mitochondrial dysfunction, genetic instability, and epigenetic changes have all been proposed as mechanisms of frailty (Bisset and Howlett, 2019). Evidence is still lacking on frailty in ethnic minority women, but it has been reported that this is likely influenced and exacerbated by biopsychosocial factors (Reid et al., 2021). Figure 6-5 describes the various factors that lead to frailty and the resulting outcomes.

Several clinical studies have indicated that the risk and prevalence and incidence of frailty are higher in women than men (Kane and Howlett, 2021). Major national databases that have looked at frailty and its association with many other chronic conditions include NHANES. In a recent study using NHANES from 1999–2018, the prevalence of frailty in women was 35.4 percent (Guo et al., 2024). Sex differences in frailty

may be attributed to the role of chronic inflammation, a key factor in the pathophysiology of frailty in women, and the role of estrogens (Gordon and Hubbard, 2019).

In general, a universal definition for frailty is lacking (Morley, 2011). The frailty phenotype developed using data from the Cardiovascular Health Study has been used to measure frailty (Fried et al., 2001; Morley, 2011). This assessment is characterized as having three out of five components: weight loss, exhaustion, weakness, slowness, and low physical activity (Fried et al., 2001).

Frailty is known to be at the intersection of various chronic conditions, including CVD, hypertension, osteoarthritis, obesity, and chronic pain (Chen et al., 2019a; Crow et al., 2019; Davis et al., 2021; Huang et al., 2023; Li et al., 2023b). Gender differences in frailty have also indicated that women with heart failure are at a greater risk (Davis et al., 2021). Sarcopenia is a major contributor to frailty.

Disparities

Data from the Louisiana Osteoporosis Study show that the prevalence of sarcopenia and sarcopenic obesity is significantly associated with social

determinants of health, such as income and education (Jeng et al., 2018). For women, age had the largest association with sarcopenic obesity, with the odds of developing sarcopenic obesity increasing from 2.3-fold for women in their 50s to fivefold for women older than 70, in comparison to women in their 40s (Jeng et al., 2018). Significant associations were also evident between race and ethnicity and sarcopenic obesity for Black compared to White adults. Furthermore, Asian adults had the highest rates of both sarcopenia and sarcopenic obesity compared to White, Black, and Hispanic adults (Jeng et al., 2018). In another U.S. study, Hispanic and Asian individuals had more severe sarcopenia as measured by grip strength and gait speed, but this study did not report gender differences (Bigman and As, 2021).

The literature shows that individuals with sarcopenia or who have evidence of a decline in muscle mass but do not yet meet the criteria for sarcopenia were more likely to have lower BMD (presarcopenia). These associations vary by race, ethnicity, and gender (Ning et al., 2021). In a study of non-Hispanic White, non-Hispanic Black, and non-Hispanic Asian men and women, older individuals with presarcopenia and sarcopenia were more likely to have lower BMD compared with those without presarcopenia and sarcopenia (Ning et al., 2021). The strongest association between sarcopenia and BMD was found in non-Hispanic Black followed by non-Hispanic White and non-Hispanic Asian individuals (Ning et al., 2021). In addition, significant differences of BMD across sarcopenia stages were found in more subgroups in women than in men after adjusting for age and BMI, with a negative linear relationship between sarcopenia and regional or whole-body BMD in non-Hispanic White, non-Hispanic Black, and non-Hispanic Asian men but not in non-Hispanic Asian women (Ning et al., 2021). The authors concluded that for outcomes that involve BMD, sarcopenia had a larger effect in women than in men.

Research Gaps

Many gaps remain in understanding sarcopenia, especially its initiation and progression, diagnostic tools and cutoff points, and outcomes and their applicability in different racial and ethnic groups and by gender (Cruz-Jentoft et al., 2019). An area lacking in research is the exploration of sex and gender differences in sarcopenia. Studies are also needed examining how structural and social determinants of health, which intersect with gender, are associated with sarcopenia. This includes investigating risk factors such as obesity, physical activity, and diet and nutrition. The link between sarcopenia and frailty needs to be explored further. In relation to frailty, areas of research opportunity include investigating the influence of inflammation, hormone dysregulation, cell senescence, and biopsychosocial factors.

SUMMARY

Biological and Mechanistic Factors

Further attention is required to understand how sex influences cellular and molecular processes in addition to physiology and pathophysiology to improve strategies in disease prevention, detection and screening, diagnosis, treatment, monitoring, and women’s specific response to drug therapies. Biological mechanisms are not well understood for certain conditions, and sex chromosome and gonadal hormone effects are not well understood for many conditions.

Understanding Risk Factors

Female-specific biological factors need to be included in both clinical and basic studies on topics such as the oral contraceptives, hormonal fluctuations, number of pregnancies, menstruation, age of menopause transition, and stage of menopause, as this will allow research to better reflect on human diseases that affect women. Understanding the influence of female-specific biological factors will facilitate developing criteria and diagnostic testing, risk assessment tools, and effective treatment strategies and options relevant to women. Sex- and gender-specific differences in pharmacokinetics play an important role in determining the safety of treatment options for women and their adverse drug reactions. According to FDA, women experience more adverse events than men, and those events are more serious in women; the reported reasons include that women are more frequently overdosed; are more sensitive; and take a greater number of medications. The strength of animal and cellular models is the ability to dissect the influence of potential biological factors by focusing on one potential sex-associated pathological driver to dissect disease mechanisms. When addressing the health conditions that plague women, researchers need to not only ensure data availability but also start evaluating the role of sex hormones and chromosomes.

Improving Diagnosis, Treatment, and Management

In terms of diagnostic tools, diagnoses are lacking for certain chronic conditions, such as fibromyalgia, ME/CFS, and chronic pain. Diagnosing sarcopenia is largely understudied compared to osteoporosis. Diagnosing mental disorders is a challenge among racial and ethnic groups.

Preclinical research for some diseases or conditions has inadequate biomarkers for diagnosis. For some biomarkers, the reference ranges have not been studied separately for men and women. Future studies should consider

sex as a major effect modifier of biomarkers rather than a confounder to be removed by adjustment. Preliminary findings on biomarkers obtained from convenience samples should be validated in population-based samples.

Comparisons of blood biomarkers in men and women should consider possible mechanisms for differences, such as sex differences in the binding of biomarkers to blood proteins, permeability of the blood–brain barrier, and effect of comorbidities. For some of the existing biomarkers, a major obstacle to clinical translation is the high cost and the invasiveness of the measure. Although research has advanced for CVD in women, diagnosis and treatment of INOCA, MINOCA, and SCAD are major gaps in research.

Gaps in research exist on how to improve treatment and management of several chronic conditions. New treatments, such as Ozempic for weight loss and Zurzuvae for PPD, have reached the market, but research on access to these medications is needed. Access to treatment is lacking for many conditions, such as HIV and SUD, and disparities are evident among different groups of women by race, ethnicity, sexual orientation, and incarceration.

Disparities

Despite advances in the last 2 decades to promote better representation of male and female sexes in basic, preclinical, and clinical research related to female-predominant conditions, disparities are still evident, and women continue to be underrepresented. Furthermore, glaring disparities related to access to care, prescribing practices, and social determinants of health are governed by a woman’s race, ethnicity, and gender. Future studies need to consider these differences to be able to identify the most efficacious ways to prevent, treat, and manage chronic conditions in women.

REFERENCES