Research report
Early life menstrual characteristics and pregnancy experiences among women with and without major depression: the Harvard study of moods and cycles

https://doi.org/10.1016/S0165-0327(02)00459-7Get rights and content

Abstract

Background: Other than premenstrual dysphoria, few studies have examined the extent to which basic characteristics of the menstrual cycle and the occurrence of other reproductive landmarks impact on the risk of major depression. Methods: From a population-based sample of 4161 premenopausal women 36–45 years of age, we identified 332 women who met DSM criteria for past or current major depression and a sample of 644 women with no such history. In person interviews included a detailed assessment of menstrual cycle characteristics from age at menarche through study enrollment as well as other reproductive landmarks. Results: Risk of depression increased significantly with decreasing age at menarche (P<0.001). The risk of depression was also higher in women with heavier menstrual flow and cycle irregularity during the first 5 years of menstruation. Women with a history of multiple abortions were 2–3-times more likely to develop major depression (95% CI 1.6–4.1). Increasing months of breastfeeding was associated with a decreased risk of depression after adjustment for education, marital status, and number of livebirths (P-value, test of trend=0.012). This association was largely confined to depression during the postpartum period. Limitations: Menstrual and pregnancy history exposures were self-reported and retrospectively assessed. However, women with and without a HISTORY of depression were subject to similar recall requirements that likely resulted in an underestimate of most risk estimates. Conclusions: Clinicians involved in routine obstetrical and gynecological care of women need to recognize that menstrual and pregnancy history events may serve as potential markers for subsequent psychiatric sequelae.

Introduction

Several national studies have documented the greater prevalence of major depression in women compared to men (Blazer et al., 1994, Regier et al., 1993, Weissman et al., 1996). It is also well documented that depressed women are more likely to have a history of premenstrual dysphoric disorder (PMDD) (Leibenluft et al., 1994, Endicott, 1993, Sherry et al., 1988, Graze et al., 1990, Halbreich, 1997). However, few studies have assessed how characteristics of the menstrual cycle during the early reproductive years, independent of late luteal phase symptoms, might impact on the development of depressive symptoms later on in life. For example, children with premature adrenarche display more psychological abnormalities and depressive symptoms (Dorn et al., 1999), and a recent study of depressed college students found them to be having longer, heavier and less frequent Menstrual periods (Harlow and Park, 1996). Similarly, there is an extensive body of literature pertaining to postpartum mood disorder (Nonacs and Cohen, 1998, Nielson Forman et al., 2000). However, it is unknown to what extent pregnancy outcomes and experiences influence the risk of developing new onset of mood disorder, not necessarily during the postpartum period. A better understanding of the relationship between early life menstrual and pregnancy events may help clinicians to identify sub-populations of women who should be more closely monitored for the development of depressive symptoms. The Harvard Study of Moods and Cycles, a population-based study of women with and without a lifetime history of major depression, provided the opportunity to examine antecedent menstrual cycle characteristics and pregnancy experiences in relation to the estimated risk of first onset of major depression.

Section snippets

Methods

The Harvard Study of Moods and Cycles was approved by the Brigham and Women’s Hospital Human Subjects Research Committee and is derived from a population-based cross-sectional sample of women between 36 and 45 years of age selected from seven Boston metropolitan area communities. Massachusetts Town Books (annual census publications that list residents by name, age, and address according to voter precincts), were used to identify all women in this age range with a verified address and telephone

Subjects with a lifetime history of major depression

From the 4,161 women that comprised the target population, all who scored 25 or greater on the CES-D were initially eligible for enrollment in the depressed cohort. In addition, those who reported a 2-week period of time when they were feeling depressed, and reported having been seen by a professional who diagnosed depression, and had received either psychotherapy or pharmacotherapy for their depression, were also eligible to be considered for enrollment. Women who met the above initial

Subjects with no lifetime history of major depression

From the same target population, all women who scored less than 16 on the CES-D and who did not indicate a 2-week period of time when they were feeling depressed and who had never sought diagnosis or treatment for depression, were initially eligible for enrollment in the nondepressed cohort. These women were also sent letters indicating their initial eligibility and were telephoned to confirm their premenopausal status and lack of current or past history of depression. Non-pregnant women

Assessment of menstrual and pregnancy history characteristics

In-person interviews were conducted with all study participants regarding their demographic and lifestyle characteristics, menstrual and reproductive history, past and current medical conditions, and use of hormonal and non-hormonal medications. Participants recalled characteristics of their menstrual cycle during the first 5 years of reproductive life, after their first livebirth delivery (when applicable), after their first use of oral contraceptives (when applicable), and at the time of

Statistical analysis

When we assessed age at menarche and menstrual cycle characteristics during the first 5 years of menstruation, we excluded 19 depressed women whose age at first onset of depression preceded their age at menarche. Unconditional logistic regression models were used to estimate the odds ratios of subsequent depression in relation to early life characteristics and reproductive history. To assess the temporal relationship between depression and prior pregnancy exposures (including livebirths,

Results

Women with a history of depression were slightly older than nondepressed women (Table 1) and we therefore adjusted for age during all subsequent analyses. Women with and without depression history were similar by race, education, and BMI at age 18. Compared to currently married women with no history of marital disruption (divorced, loss of spouse, or extended separation), women never previously married were no more likely to become depressed. However, women who had experienced some form of

Discussion

Early onset of menarche appears to substantially increase the risk of depression later in life. An Australian study of nearly 1200 girls between 12 and 16 years of age reported greater psychiatric morbidity, particularly depression and comorbid anxiety disorders, in girls whose menarche occurred more distant than in the more recent past (Patton et al., 1996). In addition, early pubertal development may lead to low self-esteem and greater negative self-concept as a consequence of difficulties

Acknowledgements

This research was supported by grant R01-MH-50013 from the National Institute of Mental Health.

References (37)

  • J.W. Carnes

    Psychosocial disturbances during and after pregnancy

    Postgrad. Med.

    (1983)
  • E. Chouinard et al.

    Recall bias in case-control studies: an empirical analysis and theoretical framework

    J. Clin. Epidemiol.

    (1994)
  • Egret for Windows

    (1999)
  • L.D. Dorn et al.

    Biopsychological and cognitive differences in children with premature versus on-time adrenarche

    Arch. Pediatr. Adolesc. Med.

    (1999)
  • P.D. Duncan et al.

    The effects of pubertal timing on body image, school behavior, and deviance

    J. Youth Adolesc.

    (1985)
  • First, M.B., Gibbon, M., Spitzer, R.L., 1996. User’s guide for the SCID-I structured clinical interview for DSM-IV...
  • K.K. Graze et al.

    Premenstrual depression predicts future major depressive disorder

    Acta Psychiatr. Scand.

    (1990)
  • U. Halbreich

    Premenstrual dysphoric disorders: a diversified cluster of vulnerability traits to depression

    Acta Psychiatr. Scand.

    (1997)
  • Cited by (69)

    • When left-behind children become adults and parents: The long-term human capital consequences of parental absence in China

      2022, China Economic Review
      Citation Excerpt :

      Regions with a high prevalence of natural disasters in history were further excluded from the baseline regressions. Third, although previous studies have indicated that recall bias tends to occur non-differentially between treated and control groups (Cui et al., 2020; Harlow, Cohen, Otto, Spiegelman, & Cramer, 2004), we checked the robustness of our findings through sample adjustment by excluding observations with low performance on word recall tests administered in the 2012 CFPS. Table 2 presents the estimates of baseline regressions for the relationships between childhood LBE and adulthood human capital accumulation (full regression estimates are shown in Appendix Table B2 and Table B3).

    • Does premenstrual syndrome before pregnancy increase the risk of postpartum depression? Findings from the Australian Longitudinal Study on Women's Health

      2021, Journal of Affective Disorders
      Citation Excerpt :

      The following sociodemographic factors and common risk factors of PMS and PPD were included as covariates, which were repeatedly collected at each survey: age, highest qualification (university degree or higher, high school/trade/diploma and less than high school), marital status (married/de facto and not married/separated/widowed), area of residence (major cities, inner regional and outer regional/remote/very remote), ability to manage on income (easy, not too bad, difficult sometimes and difficult always/impossible), social support (all of the time, most of the time and some/a little/none of the time), body mass index (BMI, <18.5, 18.5 to <25, 25 to <30 and ≥30), physical activity (sedentary, low, moderate and high), smoking (never smoker, ex-smoker and current smoker), oral contraceptives (OCPs) use (not using and using), and history of depression (based on either self-report of doctor-diagnosed depression or use of antidepressants both asked at each survey; for both measures women were classified as having a history of depression if they answered yes at any preceding survey). Age at menarche (≤11, 12-14 and ≥15 years) (Harlow et al., 2004; Mendle et al., 2018), parity (primiparous/multiparous) (Di Florio et al., 2014; Iwata et al., 2016), gestational diabetes (yes/no) (Arafa and Dong, 2019; Azami et al., 2019) and preterm birth (yes/no) (de Paula Eduardo et al., 2019; Helle et al., 2015) were also included as covariates since previous literature has reported a positive association with depression or PPD. Baseline characteristics of the women and univariable associations with prospective development of PPD were described and compared using descriptive statistics, chi-square, or t-tests.

    View all citing articles on Scopus
    View full text