Anxiety Disorders Among Women: A Female Lifespan Approach

Abstract

Anxiety symptoms may present differently between women and men as well as at different points in the female lifespan. The female lifespan includes distinct epochs of hormonal function, including puberty, the premenstruum, pregnancy or postpartum (in some women), and the menopausal transition. These stages give rise to important treatment considerations when treating anxiety among women. When making assessments, the clinician should consider reproductive events and hormonal status as well as sex differences in anxiety presentation. This review is structured to define major epochs of the female lifespan; provide a brief summary of the major anxiety disorders, with a focus on prevalence and presentation in the context of sex differences and at points in the female lifespan; describe potential biopsychosocial underpinnings of anxiety disorders among women; provide guidelines for assessment and differential diagnosis; and describe treatment options with attention to reproductive events such as pregnancy.

Women experience markedly greater prevalence of anxiety disorders than men, including generalized anxiety disorder (GAD), panic disorder, and specific phobias. In a study of more than 20,000 individuals in the United States, the Collaborative Psychiatric Epidemiology Studies found higher rates of lifetime diagnosis for nearly all anxiety disorders among women (1). Core features of the anxiety disorders include subjective anxiety or fear experience, physiologic reactivity, and, often, avoidance behavior (2). Anxiety disorders are characterized by anxious apprehension or fear in response to a perceived threat (3, 4). Anxiety or anxious apprehension is a future-oriented state in which one is concerned about potential threats, whereas fear occurs in response to an immediate threat (5). Anxiety disorders among women often precipitate or worsen at times of hormonal fluctuation, including puberty, the premenstruum, pregnancy or postpartum, and the menopausal transition (Figure 1). In this review, we address each of these pivotal timepoints in the female lifespan and how anxiety disorders present and may be treated at each timepoint.

FIGURE 1.

Anxiety Disorders Across the Female Lifespan^a

^a The female lifespan includes distinct hormonal and psychosocial epochs, including puberty, the premenstruum, pregnancy or postpartum, and the menopausal transition. These stages give rise to important assessment and treatment considerations. The clinician should consider reproductive events, hormonal status, and potential sex differences in anxiety presentation.

The Female Lifespan

Puberty

Puberty represents a vulnerable developmental window for the occurrence of anxiety disorders (6). It is at puberty that sex differences in anxiety begin to emerge, with a higher rate among girls as they enter adolescence independent of age (7, 8). A recent cohort study of nearly one million Danish youths found that prior to puberty, boys had higher rates of anxiety disorders; however, this trend reversed with puberty, showing female adolescents with higher rates of anxiety disorders (9). In the United States, a longitudinal community study annually assessed 1,420 youths from age nine through 16 years and found higher three-month percentage prevalence of most anxiety disorders among female youths (10). Among German youths age 14–24 years, panic disorder was more prevalent among female adolescents than male adolescents, with an earlier age of onset (11).

Female puberty is marked by the initiation of menses (menarche) and its accompanying monthly fluctuations in ovarian steroid hormones, described below. Ovarian steroids (e.g., estradiol, progesterone) and their metabolites are neuroactive, affecting neurotransmission at several receptors in the central nervous system. Hormonal fluctuations initiating with puberty may play a role in the etiology of anxiety disorders; however, further research in this area is needed (7). Puberty may also be a time of increased psychosocial stress, also described below, which may contribute to occurrence of anxiety symptoms.

The Menstrual Cycle

The simplest characterization of the menstrual cycle is that it includes two stages: the follicular and luteal phases. However, ovarian hormone levels are quite variable within as well as between phases. At onset of menstruation, typically considered day 1 of the cycle, ovarian hormones estradiol and progesterone are both low. However, under conditions of rising follicle stimulating hormone secreted from the pituitary, ovarian estradiol levels increase and peak at ovulation, approximately day 14. With ovulation, progesterone levels begin to rise, increasing more than 20-fold over the course of the luteal phase. Estradiol levels decline immediately after ovulation but then rise again in concert with midluteal phase progesterone. Finally, during the late luteal phase, progesterone and estradiol levels precipitously drop if pregnancy has not occurred, followed by menses and the next cycle (12).

As many as 80% of reproductive-age women experience at least one physical, mood, or anxiety symptom in the luteal phase (13–16). Approximately 20% experience significant premenstrual symptoms (13, 14, 16). A smaller subset of women, 5%−8%, experience premenstrual dysphoric disorder (PMDD), a mood disorder that often has a strong anxiety symptom component (13, 17–19). In fact, Nillni and colleagues suggested common biopsychosocial pathways between anxiety disorders and PMDD (20). Increased anxiety sensitivity and perceived control over anxiety-related events are potential contributors (21). Women with anxiety disorders may also experience premenstrual exacerbation of symptoms, described below.

Pregnancy and Postpartum

During pregnancy, estrogen and progesterone levels increase exponentially, then drop rapidly within hours of delivery. Stress or depression during pregnancy can have profound effects on both the woman and fetus, with implications for development through childhood and into adulthood (22–24). Literature on the impact of prenatal anxiety on offspring is relatively small. In a large longitudinal study, antenatal anxiety affected offspring behavioral and emotional concerns, independent of depression (25). Women with the most severe anxiety during pregnancy had children with a twofold increased risk for a behavioral or mental disorder after analyses controlled for confounders, including postnatal maternal mood and paternal pre- and postnatal mood, which would be more consistent with genetic effects (26). Given this potential for intergenerational transmission of stress, clinicians should be aware of the impact of untreated anxiety on both mother and offspring, and clinicians should also be attentive to emergence of anxiety symptoms during pregnancy. A 2014 systematic review revealed a high prevalence of anxiety disorders during pregnancy, with GAD, phobias, and panic disorder being among the most prevalent (27). A meta-analysis suggested that there may be an association between anxiety disorders and preterm birth or low birth weight; however, anxiety assessments were heterogeneous (28).

Menopausal Transition

As menopause approaches, ovarian hormones fluctuate erratically, and periods become irregular. The PENN-5 staging system divides the menopause transition into five stages: premenopausal (regular menstrual cycles of 21–35 days), late premenopausal (one observed change in cycle length of at least seven days), early transition (at least two cycles with cycle length changes of at least seven days), late transition (greater than or equal to three months of amenorrhea), and postmenopausal (greater than or equal to 12 months of amenorrhea) (29). Anxiety may heighten with the menopausal transition and then decrease. Indeed, the Study of Women’s Health Across the Nation, a multisite longitudinal epidemiologic study, found that even women with low levels of anxiety were likely to experience an increase in anxiety symptoms across the menopausal transition (30). The menopausal transition may be a time of increased stress because women experience a life transition accompanied by physical symptoms, such as hot flashes and insomnia. In fact, research strongly has suggested a link between perimenopausal hot flashes and anxiety symptoms (31–33). For women in the Study of Women’s Health Across the Nation who did not have high levels of baseline anxiety, vasomotor symptoms such as hot flashes increased the odds of having high anxiety across the stages of menopause transition (30).

Epidemiology of Anxiety Disorders Across the Female Lifespan

In this section, we describe each of the major anxiety disorders, focusing on sex differences and symptom presentation across the female lifespan.

GAD

GAD is characterized by constant, nonspecific, difficult-to-control worry (17). Lifetime prevalence of GAD is greater among women (about 6%) than among men (about 3%) (34). Perinatally, GAD prevalence ranged from 0% to 10% in a systematic review (27). Compared with men, women with GAD often present with somatic complaints such as fatigue, muscle tension, or gastrointestinal symptoms. Women often have comorbid mood disorders, whereas men often have comorbid substance use disorders (35, 36).

The Avon Longitudinal Study of Parents and Children followed a population-based cohort in the United Kingdom from age seven to 13 years. The peak frequency of “worry cognitions” occurred at age 10 years for boys and girls. Among girls, the highest level of interference with daily life occurred at age 13 years, whereas it decreased among boys (37). Although this study did not assess GAD specifically, worry cognitions are a key facet of GAD. Literature on premenstrual exacerbation of GAD is limited. One study of women with GAD found that more than half experienced worsening of anxiety symptoms premenstrually, on the basis of clinical interview (38). The literature on GAD during pregnancy and postpartum is much richer. Goodman (27) and Misri (39) both published recent reviews of perinatal GAD. Briefly, GAD is fairly prevalent perinatally (27), but symptoms can be difficult to discern from normal aspects of pregnancy and postpartum (e.g., insomnia, fatigue, difficulty concentrating) (39). Women with a history of GAD are more likely to have a recurrence perinatally; in one study, women with four or more previous GAD episodes were significantly more likely to experience GAD in pregnancy (40). Women with postpartum GAD were at higher risk for later major depressive disorder (41). A study from the Menopause Strategies: Finding Lasting Answers for Symptoms and Health trials found that 16.6% of women experiencing perimenopause had mild anxiety symptoms, and 4.4% had moderate or severe symptoms, on the basis of the GAD-7 scale (42).

Panic Disorder

Panic disorder’s core feature is an unexpected but brief burst of fear accompanied by physical “fight or flight” symptoms (e.g., tachycardia, chest pain, nausea) and followed by excessive concern about additional attacks or their consequences (17). Panic disorder is at least twice as common for women compared with men (34). Women often present with fear associated with the physical symptoms of panic, whereas men may focus more on the social consequences of a panic attack (43).

As girls progress through puberty, the likelihood of panic attack increases. For each increase in Tanner stage, there was a twofold increase in rates of panic attack in a sample of nearly 1,000 peripubertal girls (44). Those with high levels of anxiety sensitivity may be particularly vulnerable to panic attacks. Adolescents self-reported higher anxiety sensitivity in the early stages of puberty than in the later stages, but they did not report sex differences (45). In a study of healthy adolescents age 12–17 years, those with more advanced pubertal status and increased emotional reactivity were most likely to experience panic symptoms in a hyperventilation challenge, after analyses controlled for age and gender (46).

Premenstrually, women with panic disorder in one study reported increased state anxiety, anxiety sensitivity, and anxiety concerning bodily sensations or illness, compared with a control group (47). Women are also physiologically more reactive to anxiety, showing greater increases in skin conductance in the luteal phase compared with the follicular phase (47). In a small sample of women with panic disorder, patients retrospectively reported increased panic symptoms premenstrually, but this finding was not confirmed by prospective symptom ratings (48). In fact, findings indicate that panic symptoms may not commonly be exacerbated premenstrually among women diagnosed as having panic disorder (49). However, some researchers have proposed commonalities between panic disorder and PMDD because those with either diagnosis are more prone to panic attacks in laboratory challenge studies (50).

First onset of panic disorder occurs perinatally in 3%−11% of women with the disorder (51, 52). New-onset panic disorder in pregnancy is associated with a higher risk of relapse in subsequent pregnancies (53). Women with a previous diagnosis of panic disorder or another anxiety disorder are at increased risk for panic disorder perinatally (54). Women who experienced panic attacks during pregnancy had a 2.29-fold increased risk of having a small-for-gestational-age infant and a 2.54-fold (adjusted odds ratio) increased risk for preterm birth compared with a control group of healthy women (55). Women with panic disorder who did not have a panic attack during pregnancy were at similar risk for preterm birth compared with women without panic disorder (55). Interestingly, some women with panic disorder experienced improvement of symptoms during pregnancy (56). In the longitudinal study of more than 2,500 pregnant women, panic disorder was not associated with increased risk for preterm birth in analyses adjusted for factors such as other psychiatric disorders, nicotine or alcohol use, race and ethnicity, and previous pregnancy history, to name a few (57).

During perimenopause, panic disorder may arise or worsen (58). Intriguingly, there may be a link between hot flashes and panic attacks (32). Hot flashes are experienced as a feeling of flushing or heat, often with perspiration; the onset is sudden and peaks rapidly, as in panic attacks. Both may include sweating, heart palpitations, dyspnea, and nausea; both can be triggered experimentally by yohimbine (59, 60); and both respond to treatment with clonidine, gabapentin, selective serotonin reuptake inhibitors (SSRIs), and selective norepinephrine reuptake inhibitors (SNRIs).

Specific Phobia

Specific phobia is an excessive, unreasonable fear of a specific object or situation (17). Lifetime prevalence of specific phobia is about twice the rate for women (12%−27%) as for men (6%−12%) (34, 61, 62). Women tend to have more animal and environmental phobias than men. Onset is often prior to puberty (63) and is not associated with reproductive events in the female lifespan, suggesting that the higher prevalence among women is not due to ovarian steroids (64). There is little research on pubertal status and specific phobias. Among adults, women had four times greater prevalence of a specific phobia (emetophobia [fear of vomiting]), with onset most frequently in late puberty (65). A large study of Korean women found that those with PMDD had comorbid specific phobia relatively frequently (16.9%) (66). Case studies have described women who experience onset or worsening of phobias during pregnancy, but there is little literature on this topic. Pregnancy-related phobias have been described, including emetophobia, tokophobia (fear of pregnancy or childbirth itself) (67–69), and blood and injection phobia (70–72). There is similarly little research on specific phobias in perimenopause or menopause specifically.

Social Anxiety Disorder

Social anxiety disorder is characterized by excessive fear of social or performance situations in which one might be judged by others (17). Lifetime prevalence is slightly higher for women (about 5%−15%) than men (about 4%−11%) (34, 73). The National Epidemiologic Survey on Alcohol and Related Conditions assessed more than 43,000 adults in the United States by using DSM-IV criteria and found a prevalence of 5.67% for women and 4.20% for men (74). Women’s anxieties often focus on performance (e.g., scrutiny from authority figures, eating and drinking in front of others) (75), and women often present with comorbid mood disorders (74).

Social anxiety disorder typically has onset prior to age 18 years; one study found that nearly 80% of individuals with a diagnosis of the disorder experienced onset prior to age 18 years (76). However, whether social anxiety disorder onset occurs more frequently with puberty or earlier in development has been debated (77). A longitudinal study with more than 1,000 peripubertal girls and a time-varying pubertal timing model found that less advanced pubertal status was associated with greater self-reported social anxiety symptoms (78). Although this is supported by other studies (79), earlier puberty has also been associated with increased social anxiety symptomatology (80). In a longitudinal study that followed boys and girls from age nine to 16 years, the transition to adolescence brought an increase in social anxiety rates for girls but not for boys (10). Among reproductive-age women with social anxiety disorder, 45% experienced more severe anxiety symptoms in the premenstrual phase in a small sample (81). Symptoms decreased during pregnancy in one sample, returning to prepregnancy levels postpartum (81). Little is known about the course of social anxiety during perimenopause.

Anxiety in the Context of Menstrual Cycle Disorders

Women experience normal monthly fluctuations in neuroactive steroids with their menstrual cycles. A subset of women, about 5%–8%, appear to be more sensitive to these fluctuations than others and present with PMDD. PMDD itself has a strong anxiety component (18), and women with PMDD often have a comorbid anxiety disorder. In a review of the literature, women with PMDD had comorbid GAD (up to 40%), panic disorder (25%), and social anxiety disorder (approximately 20%) (82). Thus, when assessing women who present with premenstrual syndrome or other premenstrual mood symptoms (e.g., PMDD), clinicians should consider whether there is an accompanying anxiety component.

Women with polycystic ovarian syndrome (PCOS), which is characterized by increased androgen levels and irregular menstrual cycles, are at risk for mood and anxiety disorders (83–85). Women with PCOS and irregular menstrual cycles with hirsutism had higher anxiety symptoms but not depressive symptoms than women in a control group (86). Among women with PCOS who lost weight during a 16-week intervention with oral contraceptive pills or a lifestyle modification program, anxiety symptoms decreased significantly (87). Clinicians should be aware when treating ovarian-hormone-related disorders, such as PCOS, that anxiety or affective symptoms may also occur.

Biopsychosocial Underpinnings

A combination of biological and social factors influences the sex differences and course of anxiety disorders over the female lifespan. Women may be more likely to experience stressors contributing to anxiety disorders (88), have coping or cognitive styles more prone to rumination and worry (89), or have biological predispositions such as anxiety sensitivity and hormonal fluctuations that propagate anxiety disorders.

Stressors

Childhood sexual abuse occurs twice as often to girls as to boys, and girls experience multiple forms of childhood abuse, neglect, or household dysfunction at a higher rate than boys (88, 90). The Nurses’ Health Study II, which included 68,505 women, found that 57% reported some form of physical or sexual abuse in childhood (91). The pubertal window may represent a period of increased vulnerability to the programming of lifetime risk for stress-related psychiatric disorders. Although pubertal timing itself may play a role in development of anxiety disorders (92), the occurrence of stress relative to puberty may also have an influence (93). In a sample of 2,899 girls from the National Comorbidity Survey Replication—Adolescent Supplement, traumatic stress during the pubertal window (the three years prior to menarche) increased risk for anxiety disorders, compared with stress during preadolescence, which increased risk for mood disorders (93).

In adulthood, women experience trauma at a similar rate as men but experience sexual assault at 10 times the rate of men (94). Women may also be more sensitive to interpersonal stress than men, potentially contributing to the higher prevalence of depression among girls after puberty (95). As described above, women with PMDD are more likely than others to have a trauma history. Finally, women may be more physiologically reactive to stressors, contributing to increased risk of anxiety or other neuropsychiatric disorders (96–98).

Coping and Cognitive Styles

During childhood and adolescence, girls exhibit more anxious coping and more emotional distress in response to stressors compared with boys (95, 99, 100). In a prospective multiwave study of adolescents, a negative cognitive style or rumination in combination with a stressor accounted for an increasing trajectory of anxious arousal symptoms over the course of adolescence for girls (101). In adulthood, women exhibit a ruminative coping style (102) and often seek support from others in times of stress (103). Women exhibit greater anxiety sensitivity than men, defined by catastrophic interpretations of bodily sensations such as heart pounding or dizziness. Women’s susceptibility to, for example, panic disorder is likely due to an interaction of anxiety sensitivity, cognitive factors, and neuroactive steroid fluctuations (20).

Biological Factors

Although all women experience the hormonal fluctuations of puberty, the menstrual cycle, and menopause (and some pregnancy), not all women develop anxiety disorders in response to these fluctuations. Thus, determining the biological contributors that make certain women vulnerable to hormonal fluctuations is critical. Women with reproductive mood or anxiety disorders do not differ in peripheral hormone levels from women without these disorders. Thus, it is not hormone levels per se but their interaction with the central nervous system that are implicated (104).

At puberty, ovarian hormone cycling initiates with monthly fluctuations of neuroactive hormones, including estradiol and progesterone. Significant brain remodeling also occurs at puberty, including areas implicated in anxiety disorders, such as the amygdala (105–107). This remodeling may program the brain’s responsivity to stress hormones and neuroactive steroids, persisting through adulthood (108, 109). Puberty engenders both organizational and activational effects of neuroactive hormones that may be implicated in the development of anxiety disorders among female adolescents (108, 110).

Among the neuroactive steroid hormones, particularly relevant to anxiety disorders is allopregnanolone (ALLO), a metabolite of progesterone and a strong agonist of the GABA_A receptor, with anxiolytic properties similar to a benzodiazepine. Some researchers have proposed that altered interactions between ALLO and the GABA_A receptor across the menstrual cycle may contribute to anxiety and affective disorders (111, 112). For instance, some researchers have proposed that panic disorder and PMDD share common biological pathways, with ALLO as a key component (20, 113). Indeed, patients with panic disorders had higher levels of ALLO during the follicular phase compared with a control group. Brambilla and colleagues proposed that ALLO was hypersecreted to offset hypothalamic-pituitary-adrenal axis activity (114). Estrogens may also play a role in anxiety disorders. Lebrón-Milad and colleagues proposed that estradiol modulates fear learning and fear extinction, giving it a potential role in anxiety- or trauma-related disorders (115, 116).

Assessment and Differential Diagnosis

There are several tools specialized to identify anxiety symptoms in women across the lifespan. In addition, a thorough clinical interview should be conducted, assessing the patient’s family history; reproductive history, including querying about affective or anxiety symptoms (e.g., past episodes of postpartum depression); and medication history, including psychiatric medications and hormonal contraceptives or hormone therapy (HT).

In terms of diagnostic criteria, the DSM-5 provides the most up-to-date guidelines for diagnosing anxiety disorders. Accordingly, the Anxiety Disorders Workgroup developed several brief self-report scales to assess the major anxiety disorders with focus on the core components of anxiety disorders: subjective fear experiences, physiological reactivity, and avoidance behavior. These scales are available free of charge on the American Psychiatric Association’s website.

Premenstrual Anxiety

To identify a pattern of premenstrual worsening of anxiety, it is ideal to obtain prospective daily ratings from the patient for at least two menstrual cycles. The Daily Record of Severity of Problems is the gold standard for prospective tracking (117). It includes anxiety symptoms of feeling tense or on edge, difficulty concentrating, fatigue, and some physical symptoms. The patient rates each symptom daily on a scale of “none” to “extreme”; the provider can then compare the follicular and luteal phase to see whether there is a luteal increase in symptoms. The provider may ask the patient to also record specific anxiety symptoms that she experiences, such as panic attacks.

The Carolina Premenstrual Assessment Scoring System is a new instrument that provides a means of scoring Daily Record of Severity of Problems data (118). It is available for download as a worksheet or statistical package macro. Although it is designed to assess DSM-5 criteria for PMDD diagnoses, and not anxiety, it may provide a general impression of symptoms across the cycle. Mobile applications may also provide a convenient means for patients to track daily symptoms; however, many of these are not empirically supported and have not been validated for clinical accuracy, that is, compared with a diagnosis provided by a professional.

Perinatal Anxiety

Anxiety symptoms are fairly prevalent perinatally (27) but can be difficult to discern from normal aspects of pregnancy and postpartum (e.g., insomnia, fatigue, difficulty concentrating) (39, 119). Thus, a careful diagnostic interview should be conducted. A promising new scale is the Postpartum Specific Anxiety Scale, which shows good validity and properly identified women with a current clinical diagnosis of anxiety (120). Another recently developed tool is the Perinatal Anxiety Screening Scale (121), which reliably identified, using a clinical cutoff score, women at risk of anxiety (122). Scales not designed specifically for the perinatal population may be an option; one study found that Beck Anxiety Inventory items on panicky feelings, restlessness, and sleep difficulty estimated GAD prevalence at 20.2%, compared with the clinician-diagnosed 15.7% prevalence among women with postpartum symptoms and had high (83%) specificity (123). Factors such as lack of partner support, low social support, abuse history, an unplanned or unwanted pregnancy, and pregnancy complications or loss were factors associated with prenatal anxiety and should be considered in the clinical assessment (124).

Perimenopausal Anxiety

The Women’s Health Questionnaire was developed in a sample of women age 45–65 years. The scale asks women to report on physical and emotional symptoms, some specific to menopause (125). Although not geared specifically toward assessing anxiety, it does include four items on anxiety symptoms. Clinicians should assess reproductive mood history because women with premenstrual symptoms are more likely to develop anxiety perimenopausally (126). The clinician should also assess vasomotor symptoms because there may be a relationship between anxiety and hot flashes. During the menopausal transition, women with high levels of somatic anxiety had more than three times the risk of hot flashes than those with low levels of somatic anxiety; these symptoms occurred in a time-lagged fashion, such that anxiety predicted hot flashes (31).

Treatment and Outcomes

Treating Anxiety in the Premenstruum

Continuous and luteal phase treatment with SSRIs have both shown efficacy in treating PMDD (127–129). In luteal phase treatment, the woman takes an SSRI during only her luteal phase, which the clinician may mark via time of ovulation, time predicted until next menses, or simply time of symptom onset. Treatment response is generally rapid, and often a low dose is sufficient (130, 131). However, some women do not respond to SSRIs, and side effects may occur. There is little research on SSRIs specifically for premenstrual anxiety, although one small study showed a significant 45.4% decrease in Hamilton Anxiety Scale score after four weeks of sertraline treatment among women with PMDD (132). Although hormonal contraceptives may influence aspects of anxiety response, there is limited evidence for the efficacy of hormonal treatment for premenstrual mood symptoms (133–135). Some women report worsening of premenstrual mood symptoms with hormonal contraceptives (136–138). In fact, a recent prospective cohort study of more than one million Danish women found that women who were prescribed a hormonal contraceptive had a greater risk ratio for subsequently being prescribed an antidepressant or being diagnosed as having depression compared with nonusers (139). However, anxiety diagnoses were not assessed in that study. Among women with premenstrual syndrome, continuous dosing of hormonal contraceptive may be effective over the typical 21 days on/7 days off dosing schedule, avoiding hormonal fluctuations. In one study, women initiated a 21/7 dosing schedule, then crossed over to a continuous regimen of 3 mg of drosperinone and 30 µg of ethinyl estradiol. Their premenstrual symptoms decreased more in the continuous phase than the 21/7 phase (140). Yaz (Bayer AG Pharmaceuticals), which contains 3 mg drosperinone and 30 µg ethinyl estradiol on a 24/4 dosing schedule, is approved by the Food and Drug Administration for treating PMDD (141). Clinicians should be aware of the potential risks and side effects of hormonal contraceptives. Gonadotropin-releasing hormone agonists and inhibitors are another hormonal treatment option, reducing levels of estradiol, progesterone, and ALLO to postmenopausal levels (142–144). Benzodiazepines are often prescribed for anxiety symptoms, but women with PMDD show decreased sensitivity to benzodiazepines (145); whether this is true for women with premenstrual exacerbation of anxiety is not known, and the potential for misuse is a consideration.

Perinatal Anxiety—Balancing the Risks

For women who are pregnant or postpartum, treatment considerations include potential impact of medication on the fetus (146) and transmission of medication via breastmilk (147, 148). For detailed information on placental and breastmilk transmission of medications, providers may refer to resources such as ReproTox.org, an online database that provides summaries of medication effects on pregnancy, reproduction, and development (149). Providers should consider the potential risks of medication versus the potential impact of untreated anxiety on the pregnancy. A recent review of treatments concluded that in terms of pharmacotherapy, SSRIs and tricyclic antidepressants had some evidence for effectiveness in treating anxiety disorders in the perinatal period (150). For women who do not want to use medication during pregnancy, cognitive-behavioral therapy may be a good alternative, particularly for women with panic disorder or specific phobias (150, 151).

There are clear recommendations for the treatment of depression during pregnancy, with use of antidepressants for moderate to severe depression and psychotherapy as a first-line intervention for milder depression (152). Although anxiety is frequently comorbid with depression, no clear recommendations have been made for the treatment of specific anxiety disorders during pregnancy. Because 50% of pregnancies in the United States are unplanned, it behooves the patient and clinician to have a discussion early in the therapy regarding the risks and benefits of continuing to take psychotropic medications during pregnancy. For instance, benzodiazepines, which are often prescribed during the acute stages of managing anxiety or insomnia, fall into class D, which indicates evidence of human fetal risk; however, potential benefits may warrant use of the drug among pregnant women despite these potential risks. As discussed previously, active anxiety disorders during pregnancy are not without their risk even in the absence of psychotropic medication exposures. If the mother’s psychological stability would be jeopardized by tapering off the medication, a consultation with a reproductive psychiatrist would help the woman and her partner to better comprehend the literature regarding risks associated with specific medications and whether there are relatively safer alternatives. Fortunately, most antidepressants that are most frequently used to treat anxiety disorders are not considered teratogenic with first trimester exposure (152). Hence, the patient and her health care provider have time to consider options should the patient become pregnant unexpectedly.

Perimenopausal Anxiety—A Role for Hormones?

Among women experiencing perimenopause, treatment options for anxiety may include pharmacotherapy, HT, or psychotherapy (153). In terms of pharmacotherapy, most research has focused on depressive symptoms, not anxiety. In a chart review of more than 150,000 women experiencing perimenopause, those with an anxiety diagnosis were significantly more likely than those without anxiety to be prescribed HT (154). However, whether HT is effective in treating perimenopausal affective or anxiety symptoms is debated (155). In the Kronos Early Estrogen Prevention Study, women who recently experienced postmenopause and who were randomly assigned to receive 0.45 mg/day of oral conjugated equine estrogens showed improvement of depressive and anxiety symptoms compared with placebo over a four-year treatment period (156). Women with past episodes of perimenopausal depression who were treated with transdermal estradiol (100 µg/day) remained free of mood symptoms, whereas those who were withdrawn from estradiol developed depressive symptoms in a randomized, placebo-controlled trial (157). Women experiencing perimenopause with a diagnosed depressive disorder who used transdermal 17-beta-estradiol (100 µg) in a 12-week, double-blind, placebo-controlled study had significantly greater remission of depression than women with depression who received placebo (158). Again, these studies focused on depressive symptoms, not anxiety. Some studies have shown an advantage for SSRIs or SNRIs over HT for perimenopausal mood or vasomotor symptoms (159, 160). There is evidence for SSRIs reducing hot flashes (161–163), and paroxetine has been approved by the Food and Drug Administration to treat moderate to severe menopausal vasomotor symptoms (164). Gabapentin also reduces both anxiety and hot flashes and, thus, may be a treatment option for women experiencing both (165, 166).

Conclusions and Future Directions

In sum, the female lifespan encompasses several distinct epochs, each with a characteristic hormonal milieu and psychosocial features. Clinically, there is a need for better screening—although prenatal depression screening is becoming more common, only 20% of obstetricians and gynecologists surveyed reported actively screening for anxiety during pregnancy (167). When making assessments and treatment recommendations, the clinician should consider developmental stage, reproductive events and hormonal status, as well as differences in presentation between women and men. Important treatment considerations include assessing anxiety in the context of the menstrual cycle, differentiating anxiety symptoms from normal changes associated with pregnancy or postpartum, considering pregnancy and breastfeeding status when prescribing pharmacotherapy, and attending to HT and vasomotor symptoms when diagnosing and treating in perimenopause. To note, although the DSM-5 no longer classifies posttraumatic stress disorder and obsessive-compulsive disorder as anxiety disorders, there is a rich literature on these disorders among women. Although beyond the scope of this review, we encourage providers to explore this area of research. Finally, because the bulk of the literature on psychiatric sex differences focuses on depression, future research should address anxiety disorders specifically.