In 1990, the Global Burden of Disease Studyranked depression as the fourth leading disease burden worldwide; by the year 2020, it is projected to be the second leading disease burden. Depression is a leading cause of life-years lost to disabilityand it also has an impact on family, community, and employers. The risk of depression is approximately twice as high in women as in men, with greatest risk during the reproductive years. As women are frequently caregivers, the impact of depression has a carry-over effect on those receiving care. Research presented at the 159th Annual Meeting of the American Psychiatric Association highlighted points of vulnerability during the reproductive life cycle, examined risk factors, and reviewed strategies to enhance treatment during these specific times.
Ethnicity, Age, and Gender as Factors in the Management of Anxiety and Depression
Dr. Diana O. Perkins, University of North Carolina School of Medicine, Chapel Hill, North Carolina, opined that the disease burden for women suffering from depression is reflected in a number of consequences, including poor self-care. This increases the risk of other illness, impairs social function and vocational function (including missed days at work or not working at all), and increases the risk of suicide. In addition, she noted that women who are depressed and in a caregiver role may also adversely affect their children, due to dysphoric mood and depressive behaviors. She underscored the fact that women are almost twice as likely as men to suffer from depression during their lifetime, with a 13% incidence among men and a 21% incidence among women.
In reviewing the reproductive life cycle in women, she highlighted those times of increased risk, specifically menarche, pregnancy, and the perimenopause. She discussed the association between depression and the hormonal milieu, specifically the hypothalamic-pituitary-thyroid (HPT) axis, in addition to hormones such as estrogen and progesterone, which are likely to influence the human brain in its response to the environment.
Because of the importance of relationships in the lives of women, the loss of support networks and disruptions in close relationships (especially following a traumatic loss), may play a role in women's greater propensity toward depression. Dr. Perkins discussed a population study in Virginia that evaluated levels of social support and depression at baseline and at 1-year follow-up among 10,000 twin subjects (heterogeneous male/female pairs). The authors found that in instances where the level of social support was low at baseline, women had greater risk of developing depression than did women with higher levels of support. A second study cited by Dr. Perkins examined the role of stressful life events and found that women were 3 times more likely than men to develop depression following the experience of a traumatic event. The stressors determined to be significant included the death of a relative, illness, or change in residence. One stressor that was found to be significant in men, but not in women, was financial stressors.
The impact of sexual hormones on the risk for depression is unclear, but menarche is considered the first time a young woman is at increased risk. Before the onset of puberty, the risk for depression is equal and perhaps slightly higher for males than for females, but this changes as a female reaches puberty. One study evaluated 10,000 British citizens for the risk of developing ICD-10 major depression. The study found that by age 16, the risk for depression was much higher in women and this difference continued to increase as a woman entered young adulthood. The study also found that, during menopause and into late adulthood, the rate of depression in males and females tended to be more alike, but after menopause, women had a lower risk of developing new-onset depression.
Other studies have shown that as a girl enters puberty and experiences hormonal changes, she may become more vulnerable to the impact of relationships and stressful life events. One study evaluated the impact of life events on depression risk among 1400 males and 1700 females and found that the correlation between 1 or more life events and the risk for depression increased with age for both genders, but at a greater rate for females.
The reciprocal relationship of social and biological development challenges researchers as they examine risks and vulnerabilities. Research in this area raises a question of whether our guidelines for treatment should differ according to gender. The standard treatment for both men and women involves antidepressant medication and psychotherapy. Psychotherapy focused on relationship issues, for example, may influence outcome for women more.
With respect to pregnancy and the postpartum period, Dr. Perkins reminded the audience that a pregnant woman undergoes dramatic hormonal changes. For example, estradiol may increase 50-fold and then return to follicular phase levels by Day 3 postpartum; progesterone increases by 10-fold and then normalizes by Day 3, and oxytocin levels may surge, activated by mother/infant attachment. Such rapid changes are highly likely to influence postpartum symptoms and increase the risk of depression. On the other hand, no evidence implicates a causative role for any specific hormone. This lack of evidence may reflect a difference in peripheral, as opposed to brain, levels and their impact on neurotransmitter systems. More likely, though, is that marked fluctuations in these levels reflect uniquely individual sensitivities. Ongoing research is examining neuroactive steroids, precursors to progesterone, testosterone, cortisol, prolactin, oxytocin, and other hormones. Most likely, it is the fluctuations in the interactions between a combination of these hormones and their effect on the central neurotransmitters, including serotonin, norepinephrine, and dopamine, that are important.
Dr. Perkins also referred to research that is examining the association between low cholesterol in women and the increased risk of suicide. However, studies that have looked at cholesterol levels postpartum have been inconsistent. Omega-3 fatty acids, essential to cellular membranes but also critical to the developing fetus, may be decreased in pregnant women. Two studies have found that women with postpartum depression have had lower levels of omega-3. If a woman avoids fatty fishes (due to concerns about mercury content), taking omega-3 fatty acid supplements during pregnancy may be a safe hedge, although there is no current evidence for any benefit.
Pregnancy is also a time of increased stress. Whether or not stress is an additional risk factor for postpartum depression is a topic of research. For women who develop depression during pregnancy, the decision to treat depression requires thoughtful consideration of risks and benefits and treatment options.
The third major period of risk for new onset or relapse of depression is the perimenopause. A recent study looked at new cases of first-onset depression as measured by the Center for Epidemiologic Studies Depression Scale (CES-D) and followed women who were perimenopausal at the time of enrollment in the study for up to 8 years. Of the 231 women, 59 developed major depression during the follow-up period and, of those, half presented with significant depressive symptoms. The study concluded that the transition between perimenopause and menopause was associated with a 5-fold increase in the risk of depression. Although absolute estradiol levels did not predict depression, the fluctuation at this level was a predictor, and the higher the standard deviation, the greater the risk. Of interest, hot flushes and vasomotor events, signs of a transition to menopause, are also associated with fluctuations in estradiol levels. Hormones that regulate ovarian reserve (eg, inhibin, LH, and FSH, all of which impact on the ovary) have been shown in studies to be associated with the risk of developing depressive symptoms. As ovarian reserve decreases, the risk of depression appears to increase.
With respect to identifying those women who are most vulnerable, susceptibility is increased in women with the following:
- Past depression history;
- Significant premenstrual symptoms;
- Early menopause;
- Vasomotor symptoms;
- Stressful life events;
- Unhealthy lifestyles;
- Marital discord;
- Obesity; and
Rates do appear to be somewhat lower in Asian women, for unclear reasons.
The role of hormone replacement therapy (HRT) in managing mood disorders in the peri- and postmenopause has been confusing. To date, most studies have been poorly designed and have yielded inconsistent findings. The Women's Health Initiative indicated that HRT did not offer a health benefit and did pose some risks; however, the study has been criticized as inadequate in being able to answer the risk-benefit question for younger women who use HRT in the perimenopause. A 12-week, double-blind, placebo-controlled study evaluated subjects with major depressive disorder (MDD), dysthymia, or minor depression who were on either an HRT patch or placebo. There was significant improvement in depressive symptoms in women taking HRT. The remission rates for those on HRT were significantly higher than remission rates of those on placebo, suggesting that modulating estrogen during perimenopause may be helpful in reducing risk of depression during this period.
Dr. Perkins stressed the need for gender-specific research for both men and women to more thoroughly examine differences between the sexes with regard to treatment efficacy in general and psychotherapeutic approaches in particular.
New research studies were presented at the APA meeting bearing on these issues.
At the Emory Women's Mental Health Program, a shortened version of the Beck Depression Index (BDI) termed the "Baby Beck," a 5- item scale, was validated and may be beneficial in busy obstetrical practices as a quick and reliable measure for screening, but not diagnosing, depression. Scales were scored using the standard procedures designated for each measure, and statistical analyses were conducted to validate the BDI for use during pregnancy and postpartum to create a brief clinical screening instrument that could be used during the perinatal period. Berg and colleagues evaluated the "Baby Beck" as a method of assessing depression and emphasized that this scale can be used as a basis for providers to begin a discussion of depression with their patients and can assist in identifying symptoms during pregnancy and the postpartum period that are most often linked to clinically significant depression. The scale was validated in a sample of 502 women during their pregnancies and up to 1 year postpartum. Participants were grouped by trimester (trimester 1 [0-13.9 weeks estimated gestational age [EGA]), trimester 2 (14.0-27.9 weeks EGA), trimester 3 (28 weeks - birth); postpartum 1 (0-6 weeks), and postpartum 2 (6.1-52.0 weeks) and given the BDI, Edinburgh Postnatal Depression Scale (EPDS), Clinical Global Impression (CGI) scale, and Hamilton Rating Scale for Depression (HRSD)-17.
Statistical analyses show the BDI itself to be reliable during pregnancy and an effective tool for clinical care or research.
Sharma and colleagues also examined depression diagnosis during the postpartum period, screening for major depression at 3 months following delivery. In this study, 38 women aged 17-37 years (mean age, 30.1) who were seen consecutively at a mood disorders clinic were asked to participate and were interviewed using the Structured Clinical Interview for DSM-IV (SCID) as the primary measure. Primary diagnoses for these women were bipolar II (24%), bipolar I (24%), MDD (24%), and bipolar not otherwise specified (18%). Current comorbidity included anxiety disorders (86%), substance use disorders (3%), eating disorders (3%), and no disorders (8%). Lifetime comorbidities included substance use disorders (42%), anxiety disorders (38%), eating disorders (5%), somatoform disorder (5%), and none (10%). Though this study was conducted on a very small population, the results suggest that there is the opportunity for early detection of bipolar disorder through the use of reliable and valid assessments, so that subsequent implementation of appropriate prevention and treatment strategies can occur.
The use of medication during pregnancy remains a controversial and complicated topic. One study, presented by Dubin and colleagues, reported on a retrospective chart review of 80 women admitted to a crisis response center with HCG-positive urine. Within this population, 39% of the women received psychotropic medication, 24% received a benzodiazepine, 47% received an antipsychotic only, and 35% were administered a combination antipsychotic and benzodiazepine during admission. Haloperidol alone or in combination with a benzodiazepine was the most frequently administered agent. No adverse effects were observed in patients for whom records of delivery were obtained. However, the sample size was impossibly small and therefore precluded evaluation of any increased risk of treatment over nontreatment.
King and colleagues examined 155 pregnant women (< 32 weeks gestation) with a history of MDD as determined by the Structured Clinical Interview for DSM-IV Axis I Disorders (SCID) (HRSD-17 ≥ 15) who were enrolled in a prospective study and followed for 6 months postpartum. Depression rating scales, including the BDI, HRSD, and SCID mood modules, were utilized. Of the 141 patients who completed the study, 107 were taking antidepressant medication near delivery. Women with HRSD ≥ 15 at 32-36 weeks vs in late pregnancy (n = 25) had significantly higher rates of postpartum depression than women with an HRSD < 15 in late pregnancy (n = 116) (60% vs 28%, respectively). This study was neither designed nor powered adequately to address the possible protective effects of treatment, but did show protection of antidepressants for these women. Women not taking antidepressants in the more severely depressed group (n = 7) relapsed at a rate of 57%, while those with less severe depression had a 22% recurrence of postpartum depression. More severely depressed women on antidepressant medication during pregnancy (n = 107) had a recurrence rate of 61%, while those less depressed but on medication (n = 18) had a recurrence rate of 29%.
Bothmer and colleagues evaluated 158 patients with premenstrual dysphoric disorder (PMDD) in a randomized, double-blind, placebo-controlled study with 3 arms: escitalopram 10 mg; escitalopram 20 mg; and placebo. Participants self-rated their symptoms using a visual analog scale (VAS) across 3 menstrual cycles. Those receiving escitalopram 10 mg showed an 86% (P < .01) decrease in symptoms, patients in the 20-mg group had a 94% (P < .001) decrease, and patients in the placebo group demonstrated a 69% decrease (P < .01). Reduction in irritability was observed in 86%, 92%, and 56%, respectively, and the percentage of subjects achieving remissions (as defined by a ≥ 80% reduction in the irritability score) was 60%, 80%, and 30% respectively. The most common adverse event was nausea, which declined following the first cycle; study discontinuation rates were 13%, 6%, and 6%, respectively.
Reports involving mood in the perimenopause address symptoms such as sleep disturbances and hot flushes.
In a double-blind, placebo-controlled study, Soares and colleagues treated patients with eszopiclone 3 mg or placebo nightly for 4 weeks to evaluate whether treatment of insomnia improved mood, menopause-related symptoms, and quality of life (QOL). A total of 410 women met the Staging System for Reproductive Aging in Women (STRAW) criteria stages -2, -1, or 1a, reporting sleep latency ≥ 30 minutes and total sleep time (TST) ≤ 6 hours a night. Participants gave a daily report of sleep satisfaction. Physician global evaluations (PGE) of menopause, menopause-specific QOL questionnaire (MENQOL), Greene Climacteric Scale (GCS), the Montgomery Asberg Depression Rating Scale (MADRS), and the Sheehan Disability Scale (SDS) were collected at baseline and at the end of treatment.
Results showed that subjects receiving eszopiclone reported significantly greater improvements in sleep latency, sleep maintenance (awakenings and time awake at night), TST, sleep quality (all P values < .0001 vs placebo), and awakenings due to hot flushes (P = .001). No differences in severity or number of daytime hot flushes were found. Those treated with eszopiclone had significantly greater improvements in MADRS scores (P < .03) and PGEs (P < .0001); total GCS score and the vasomotor and physical domains of the MENQOL (P < .05 vs placebo); and family life-home disability domain using the SDS (P < .05). The most common adverse event was unpleasant taste in those receiving eszopiclone (17.6% vs 0.5%). Other adverse events, such as headache, were similar for the 2 groups.
Finally, Joffe and colleagues examined hot flushes in postmenopausal women with major depression (N = 29) in an open-label clinical trial. At study entry, all subjects were off hormonal therapy and had Mini International Neuropsychiatric Interview (MINI)-rated major depression, MADRS score > 20, and significant menopausal symptoms (GCS total score > 20, GCS vasomotor subscale score > 3, or > 14 hot flushes/week). After a 2-week single-blind placebo run in, all subjects not responding to placebo were treated for 8 weeks with flexible dosing of duloxetine 60 mg/day to 120 mg/day. Changes in mood and hot flushes were assessed using the MADRS and GCS scores, respectively. To date, 17 women have enrolled (mean age = 52) and 9 have been eligible for treatment after the placebo run in. MADRS scores in this group improved significantly from 24.5 + 2.8 to 5.0 + 4.0 (P < .001) with duloxetine therapy (final dose 81.4 + 22.7 mg/day). In addition, menopausal symptoms showed significant improvement, with GCS total and vasomotor subscale scores decreasing from 24.8 + 4.4 to 9.2 + 3.4 (P = .001) and 4.7 + 1.5 to 2.2 + 0.98 (P = .004),respectively.