Depressive Symptoms, Menopausal Status, and Climacteric Symptoms in Women at Midlife
Bosworth, Hayden B. PhD; Bastian, Lori A. MD, MPH; Kuchibhatla, Maggie N. PhD; Steffens, David C. MD; McBride, Colleen M. PhD; Sugg Skinner, Celette PhD; Rimer, Barbara K. DrPH; Siegler, Ilene C. PhD, MPH
From the Department of Medicine, Division of General Internal Medicine, (H.B.B. and L.A.B.); Department of Psychiatry and Behavioral Sciences, (H.B.B., D.C.S., and I.C.S.); Center for Aging and Human Development (H.B.B., L.A.B., and I.C.S.); and Cancer Prevention, Detection, and Control Research Program, Duke Comprehensive Cancer Center and Department of Community and Family Medicine (M.N.K., C.M.M., C.S.S., and B.K.R.), Duke University; Health Services Research and Development, Durham VAMC (H.B.B. and L.A.B.), Durham, North Carolina; and Division of Cancer Control and Population Studies, National Cancer Institute (B.K.R.), Bethesda, Maryland.
Address reprint requests to: Hayden B. Bosworth, VAMC (152), 508 Fulton St., Durham NC 27707. Email: hboswort@acpub. duke.edu
Received for publication June 16, 2000; revision received November 15, 2000.
Objective: Previous studies have found increased rates of depression in women aged 45 to 54 years, but the factors that influence these rates are not understood. It was assessed whether higher rates of depressive symptoms were associated with menopausal status, climacteric symptoms, and use of hormone replacement therapy.
Design: Cross-sectional survey.
Setting: Community sample.
Methods: Data are from 581 women ages 45 to 54 years who were interviewed by telephone between October 1998 and February 1999.
Measures: Depression was measured with the abbreviated CES-D, a depressive symptoms screening measure. Women’s reported perception of menopausal stage, frequency of periods in the preceding 12 months, and history of oophorectomy were used to classify their menopausal status into four categories: (1) no indication of menopause; (2) close to menopause; (3) had begun menopause; and (4) had completed menopause.
Results: There were 168 women (28.9%) who reported a high level (≥10) of depressive symptoms when the abbreviated CES-D was used. In a logistic-regression analysis, significant factors associated with increased depressive symptoms included physical inactivity, inadequate income, use of estrogen/progesterone combination, and presence of climacteric symptoms (trouble sleeping, mood swings, or memory problems). Menopausal status was not associated with depressive symptoms.
Conclusions: In this sample of women age 45 to 54 years, climacteric symptoms but not menopausal status were associated with higher rates of depressive symptoms.
HRT = hormonal replacement therapy;, CES-D = Center for Epidemiological Studies Depression Scale;, OR = odds ratio;, CI = confidence interval;, SD = standard deviation;, DSM-III-R =Diagnostic and Statistical Manual of Mental Disorders, third edition, revised.
The National Comorbidity Study found rates of recurrent depression to be highest among women between the ages of 45 and 54 years when compared with those of older women (1). For the majority of women, the process of menopause occurs between the ages of 45 to 54 years. However, several studies have shown no association between menopause and increased depressive symptoms (2–6). Less clear is the association between depression and perimenopause, the time before cessation of menses that is characterized by increased occurrence of symptoms such as hot flashes and night sweats that may lead to insomnia and depressed mood (7–9). Two studies have suggested modest increases in depressive symptoms at perimenopause (7, 10). However, it is unclear whether hormonal shifts during perimenopause increased depressive symptoms or that climacteric symptoms were being interpreted as depressive symptoms.
Because depression symptoms may be related to decreasing levels of endogenous estrogen associated with perimenopause, researchers have explored whether HRT has an antidepressant role. Findings are equivocal. Although some observational studies have found HRT to be efficacious in alleviating both mood and somatic symptoms during perimenopause (11–13), randomized placebo-controlled trials have not found that use of an estrogen/progesterone combination significantly improves mood (14, 15). The frequent administration of progesterone with estrogen confounds exploration of the antidepressant role of HRT. There is some evidence that the addition of progesterone—necessary for women with an intact uterus—may negate or attenuate the overall well-being attained with estrogen alone (16–18).
This study addressed two questions concerning depressive symptoms in this age cohort of women: (1) are menopausal status and climacteric symptoms associated with depression, and (2) is the association among menopausal status, climacteric symptoms, and depression moderated by the use of HRT?
Data are from telephone interviews conducted between October 1998 and February 1999 among a random sample of women aged 45 to 54 years residing in Durham County, NC. Women who were “willing to consider the pros and cons of HRT” and did not have a history of breast cancer were eligible to participate. Of 858 women contacted, 581 (68%) met inclusion criteria (eg, no breast cancer and age criteria) and were willing to participate.
The telephone interview assessed sociodemographics (ie, age, race, education, marital status, adequacy of household income, and regular health care provider), exercise, smoking status, menopause status, gynecological surgery history (hysterectomy and/or oophorectomy), and use of HRT (ever and current use). Depression was measured by the abbreviated CES-D.
Women’s perceived menopausal stage (19), as well as their reported frequency of periods during the preceding 12 months and history of oophorectomy, was used to classify four categories of menopausal status: (1) no indication of menopause; (2) close to menopause; (3) had begun menopause; and (4) had completed menopause.
Climacteric symptoms, including whether women were experiencing vasomotor symptoms such as night sweats, hot flashes, and insomnia, as well as memory loss/forgetfulness and mood swings, were assessed. These five symptoms were included in the analyses because they were the most commonly endorsed climacteric symptoms in our study.
The abbreviated CES-D, a 10-item depression-screening questionnaire, assessed depressive symptoms. Efficacy has been examined in previous studies that used receiver operating characteristics analysis to compare the CES-D with psychiatrist-administered semistructured interviews (20), the CES-D had an area under the curve (AUC) of 0.74, which is considered to be moderately accurate (21). In fact, its accuracy is comparable to such routinely used medical tests as the mean red cell volume to screen for iron-deficiency anemia (AUC = 0.76) and the fasting blood glucose to detect diabetes mellitus (AUC = 0.83) (22).
The abbreviated CES-D has been validated against the full 20-item CES-D in a sample of >4000 community-residing elderly adults (23). The full CES-D has high levels of reliability and validity to detect both clinical (by the criteria of the DSM-IIIR) and nonclinical symptoms of depressed mood for a wide range of study populations (24–29). In addition, a cutoff score of 16 on the full CES-D has been validated with DSM-IIIR criteria for clinical depression (24, 26–29). The abbreviated CES-D showed good predictive accuracy when compared with the full-length 20-item CES-D (κ = 0.97, p < .001) (20). Test-retest correlations for the CES-D were relatively stable at 3 (r = .71) and 12 (r = .59) (23). A previous study found that use of a cutoff of 10 or more yielded a prevalence estimate of 11.7% in a large sample of older adult (M = 72.8 years) Health Maintenance Organization members (23).
The abbreviated CES-D measures depressed mood and affect (3 items), somatic complaints (4 items), well-being (2 items), and irritability (1 item). Somatic complaints include “sleep problems,” “everything required effort,” “lack of concentration,” and “not being able to get going.”
Frequencies are presented across the whole sample and are stratified by depressive status. Chi-square statistics were used to determine whether menopausal status differed in amount of depressive symptoms, and logistic-regression models were used to examine whether climacteric symptoms, menopausal status, and HRT use (estrogen only or estrogen/progesterone combination) were independently associated with depressive symptoms.
Among this cohort of women age 45 to 54 years, 70 (12%) had no indication of menopause, 173 (30%) were close to reaching menopause, 187 (33%) had begun menopause, and 145 (25%) had completed menopause. The most common climacteric symptoms were hot flashes (65%), night sweats (56%), trouble sleeping (45%), mood swings (49%), and memory problems (44%). One hundred sixty-four women (28%) reported current use of hormonal replacement medication; 236 (41%) reported ever using a hormone replacement medication. Almost one third of the participants (29%) had abbreviated CES-D scores that indicated significant depressive symptoms (eg, ≥10). Among these 168 women with high depressive symptoms, the mean depressive score was 14.5 (range 10–30).
Overall, menopausal status—(1) no indication of, (2) close to reaching, (3) had begun, or (4) had completed menopause—was not strongly associated with depressive symptoms (p < .08). Examination of menopausal status indicated no significant difference in proportion of depressive symptoms among women in categories 2 to 4 (eg, close to, had begun, or had completed menopause). However, of women with no depressive symptoms, 14% were in group 1 (no indication of menopause), whereas, among women who did indicate depressive symptoms, only 7% were in this premenopausal group (Table 1).
All five climacteric symptoms were related to depressive symptoms. Women who had significant depressive symptoms, compared with women who did not, reported more hot flashes (72% vs. 62%), night sweats (66% vs. 52%), trouble sleeping (61% vs. 39%), mood swings (68% vs. 41%), and memory problems (60% vs. 37%).
There were no differences in proportion of depressed vs. nondepressed women by age, race, marital status, and whether women had a regular health care source. Women who reported greater depressive symptoms were significantly less likely than those who reported fewer depressive symptoms to have a college education (69% vs. 78%) or to report not having adequate income (85% vs. 94%), work for pay (79% vs. 87%), and exercise (48% vs. 68%). Women with depressive symptoms also were more likely to smoke (19% vs. 9%) than women who reported fewer depressive symptoms.
There were no differences between depressed and nondepressed women by whether they ever used HRT, were currently using HRT in general, or were currently using estrogen alone. However, women with fewer depressive symptoms were more likely to be using an estrogen/progesterone combination than women with more depressive symptoms (30% vs. 16%).
After adjusting for demographic factors (see Table 2), women in the close-to-menopause and begun-menopause categories were more likely to report significant depressive symptoms than those with no indication of menopause. In the final model, after adjusting for age, education, income, exercise, smoking status, climacteric symptoms (presence of hot flashes, night sweats, trouble sleeping, mood swings, and memory problems), and HRT use, menopausal status was no longer significantly related to depressive symptoms. Covariates significantly associated with increased depressive symptoms included physical inactivity, having inadequate income, being a smoker, the use of estrogen alone rather than in combination with progesterone, and experiencing climacteric symptoms (trouble sleeping, mood swings, and memory loss).
Perimenopausal status was significantly associated with greater depressive symptoms only when climacteric symptoms were not considered. We conclude that, although perimenopause is associated with an increase in depressive symptoms, it is climacteric symptoms rather than depressive symptoms that account for this relationship. In particular, we note the overlap between symptoms of depression and climacteric symptoms such as disturbances of mood, sleep, and concentration.
Almost 30% of our sample of women age 45 to 54 years of age reported high levels of depressive symptoms. Lower rates of depression had been previously reported in a similarly aged community sample (20–25% of women reported symptoms of irritability or depression) (30). The relatively higher prevalence of depressive symptoms noted in our study may be explained in part by our methods. That is, women were recruited on the basis of their interest in discussing HRT, and those more interested in HRT may have been experiencing more climacteric symptoms.
The high prevalence of depression may also reflect differences in specificity of the abbreviated CES-D compared with other measures. For the full CES-D, a cutoff value of 16 is commonly used for community samples and 20 for hospital patients. Further examination into appropriate cutoff values for perimenopausal women experiencing climacteric symptoms needs to be explored.
Women with increased depressive symptoms, as indicated by the abbreviated CES-D, were less likely to have a college education, reported less adequate income, work for pay, less likely to exercise, and were more likely to smoke, have trouble sleeping, and report mood swings and memory problems. Depressed women were also less likely to use an estrogen/progesterone combination. In bivariate analyses, women who had reported no indication of menopause had fewer depressive symptoms.
Depressive symptoms were not associated with surgical menopause (oophorectomy) or use of unopposed estrogen (eg, not combined with progesterone). However, women who used an estrogen/progesterone combination had significantly fewer depressive symptoms than those who did not use estrogen alone. A previous randomized, placebo-controlled study found no evidence that estrogen/progesterone combination therapy was related to increased depressive symptoms (31). Although our findings might imply that estrogen/progesterone combination is more effective in relieving depressive symptoms than estrogen alone, it is important to consider that only 10 women categorized as being depressed were using an estrogen/progesterone combination, as opposed to 54 women who were using a similar combination and were not depressed. An alternative explanation for our finding is that physicians may have been more hesitant to prescribe progesterone to depressed women. Thus, given the small sample size and the possibility that a selection bias occurred, this result needs to be considered carefully.
Our findings seem to support the hypothesis that the association between depressive symptoms and menopause stems from a decline in estrogen that is directly associated with biochemical changes in the brain that lead to depression (32, 33). Our findings support this hypothesis in two ways (2). We found the highest rate of depressive symptoms among women who were close to or had begun menopause. In theory, women in these stages of menopausal transition experience the greatest decline in estrogen, whereas women who have completed menopause have adapted to low and stable levels of endogenous estrogen and are no longer experiencing drastic shifts in estrogen levels. Similarly, if estrogen decline caused depressive symptoms, one would expect women taking HRT to have lower levels of depressive symptoms (3). Although these differences were not statistically significant (p < .08), we found that fewer women who were currently taking HRT had depressive symptoms compared with those not taking HRT (29% vs. 22%). In support of this relationship, a recent randomized clinical trial reported that estrogen significantly boosted mood in 80% of the depressed women (34).
Exercise was associated with decreased depressive symptoms in this sample. Exercise may alleviate some climacteric symptoms. Research has indicated that exercisers’ moods are significantly more positive than sedentary women’s moods, regardless of menopausal state (35, 36). Slavin and Lee (35) found that exercising women scored lower on somatic symptoms and memory-concentration difficulties. Exercise may alleviate depressive symptoms indirectly by preventing hot flashes. In a controlled study of >1200 women aged between 52 and 54 years, moderate and severe hot flashes and sweats were only half as common among the physically active postmenopausal women (22%) than in the control group (44%) (37).
The significant relationship found in our study between smoking and depressive symptoms suggests that women who smoke may constitute a high-risk group that deserves more attention. This relationship between smoking and depression has been noted previously (38–43). In a community-based cohort of women aged 36 to 44 years, for example, smokers in the upper tertile of pack-years were 1.9 (95% CI 1.5–2.3) times more likely to have CES-D scores of 16 or more (38).
A lack of adequate income was associated with increased depressive symptoms. This result is consistent with other large community studies. Kaplan et al. (44), for example, found, in a random sample of approximately 7000 Alameda County adults, that inadequate income was associated with an increase risk of depression (RR = 1.42; CI = 1.24–1.72). It is likely that financial difficulties are associated with increased stress and subsequent depressive symptoms (45, 46).
The results of this study should be considered in light of a few limitations. Climacteric and depressive symptoms were both measured by self-report and may well be manifestations of the same underlying biological process. It may be difficult, even with a clinical interview, to clearly distinguish these variables. It is also important to recognize that this study was cross-sectional, and no causality can be inferred from the results. This study did not consider prior depression—a potent predictor of current depression status. Other potential factors that may be related to high level of depressive symptoms in this age group that were not examined include stress of daily living, chronic disease, and family problems (4).
Menopausal status does not appear to have a direct relationship with depressive symptoms, but rather increased climacteric symptoms explain the observed increased rates of depressive symptoms among women of this cohort. Even after considering hormone replacement medication, climacteric symptoms remained significantly related to increased depressive symptoms. We conclude that climacteric symptoms explain the relationship between menopause state and depression (eg, the menopausal state itself is associated with what appears to be depressive symptoms such as mood disturbances, sleep, and concentration problems). This has implications for our understanding of both menopause and depression. For example, should the nomenclature be changed to accommodate mood disorders that occur during perimenopause? Further studies are warranted to understand how hormonal shifts can produce depressive symptoms.
This research is supported by Grant 5U19-CA72099 from the National Cancer Institute.
1. Kessler RC, McGonagle KA, Swartz M, Blazer DG, Nelson CB. Sex and depression in the National Comorbidity Survey 1: lifetime prevalence, chronicity, and recurrence. J Affect Disord 1993; 29: 85–96.
2. Hallstrom T, Samuelsson S. Mental health in the climacteric: the longitudinal study of women in Gothenburg. Acta Obstet Gynecol Scand 1985; Suppl 130: 13–8.
3. Holte A, Mikkelsen A. Psychosocial determinants of climacteric complaints. Maturitas 1991; 13: 205–15.
4. Kaufert PA, Gilbert P, Tate R. The Manitoba Project. A re-examination of the link between menopause and depression. Maturitas 1992; 14: 143–55.
5. Mathews KA, Wing RR, Kuller LH, Meilahn EN, Kelsey SF, Costello EJ, Caggiula AW. Influences of natural menopause on psychological characteristics and symptoms of middle-aged healthy women. J Consult Clin Psych 1990; 58: 345–51.
6. Mitchell ES, Woods NF. Symptom experiences of midlife women: observations from the Seattle Midlife women’s health study. Maturitas 1996; 25: 1–10.
7. Hunter MS. Somatic experience of the menopause: a prospective study. Psychosom Med 1990; 52: 357–67.
8. Dennerstein L, Burrows GD, Hyman GJ, Sharpe K. Hormone therapy and affect. Maturitas 1979; 1 (4): 247–59.
9. Palinkas LA, Barrett-Connor E. Estrogen use and depressive symptoms in menopausal women. Obstet Gynecol 1992; 80: 30–6.
10. Avis NE, Brambilla D, McKinlay SM, Vass K. A longitudinal analysis of the association between menopause and depression: results from the Massachusetts Women’s Health Study. Ann Epidemiol 1994; 4: 214–20.
11. Griffin S. Menopause and mood. Depression 1995; 3: 56–9.
12. Sherwin BB. Affective changes with estrogen and androgen replacement therapy in surgically menopausal women. J Affect Discord 1988; 14: 177–87.
13. Pearlstein T, Rosen K, Stone AB. Mood disorders and menopause. Endocrinol Metab Clin North Am 1997; 26: 279–94.
14. Myers LS, Dixen J, Morrissette D, Carmichael M, Davidson JM. Effects of estrogen, androgen, and progestin on sexual psychophysiology and behavior in postmenopausal women. J Clin Endocrinol Metab 1990; 70: 1124–31.
15. Iatrakis G, Haronis N, Sakellarapolous G, Kourkoubas A, Gallos M. Psychosomatic symptoms of post-menopausal women with or without hormonal treatment. Psychother Pschosom 1986; 46: 116–21.
16. Sherwin BB, Gelfand MM. A prospective one-year study of estrogen and progestin in postmenopausal women: effects on clinical symptoms and lipoprotein lipid. Obstet Gynecol 1989; 73: 759–66.
17. Holst J, Backstrom T, Hammarback S, von Shoultz B. Progeston addition during oestrogen replacement therapy-effects on vasomotor symptoms and mood. Maturitas 1989; 11: 13–20.
18. Dennerstein L, Burrows G. Psychological effects of progestogens in the post-menopausal years. Maturitas 1986; 8: 101–6.
19. Bastian LA, Couchman GM, Rimer BK, McBride CM, Feaganes JR, Siegler IC. Perceptions of menopausal stage and patterns of hormone replacement therapy. J Women Health 1997; 6: 467–75.
20. Furukawa T, Anraku K, Hiroe T, Takahashi K, Kitamura T, Hirai T, Takahashi K, Ida M. Screening for depression among first-visit psychiatric patients: comparison of different scoring methods for the Center for the Epidemiologic Studies Depression Scale using receiver operating characteristic analyses. Psychiatry Clin Neurosci 1997; 51: 71–8.
21. Swets JA. Measuring the accuracy of diagnostic system. Science 1988; 240: 1285–93.
22. Erdreich LS, Lee ET. Use of relative operating characteristic analysis in epidemiology: a method for dealing with subjective judgement. Am J Epidemiol 1981; 114: 649–62.
23. Andersen EM, Carter WB, Malmgren JA, Patrick DL. Screening for depression in well older adults: evaluation of a short form of the CES-D. Am J Prev Med 1994; 10 (2) 77–84.
24. Myers JK, Weissman MM. Use of self-report symptom scale to detect depression in a community sample. Am J Psychiatry 1980; 137: 1081–4.
25. Himmelfarb S, Murrell SA. Reliability and validity of five mental health scales in older persons. J Gerontol 1983; 38: 333–9.
26. Berkman LF, Berkman CS, Kasl SL, Freeman DH Jr, Leo L, Ostfeld AM, Cornoni-Huntley J, Brody JA. Depressive symptoms in relation to physical health and functioning in the elderly. Am J Epidemiol 1986; 124: 372–88.
27. Frerichs RR, Aneshensel CS, Clark VA. Prevalence of depression in Los Angeles County. Am J Epidemiol 1981; 113: 691–9.
28. Roberts RE, Vernon SW. The Center for Epidemiologic Studies Depression Scale: its use in a common sample. Am J Psychiatry 1983; 140: 41–6.
29. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385–401.
30. Goldani von Muhlen D, Kritz-Silverstein D, Barret-Connor E. A community based study of menopause symptoms and estrogen replacement in older women. Maturitas 1995; 22: 71–8.
31. Girdler SS, O’Briant C, Steege J, Grewen K, Light KC. A comparison of the effect of estrogen with or without progesterone on mood and physical symptoms in postmenopausal women. J Women Health Gender Based Med 1999; 8: 637–646.
32. Dennerstein L. Depression in the menopause. Obstet Gynecol Clin North Am 1987; 4: 33–45.
33. Gath D, Iles S. Depression and menopause. BMJ 1990; 300: 1387–8.
34. Schmidt PJ, Nieman L, Danaceau MA, Tobin MB, Roca CA, Murphy JH, Rubinow DR. Estrogen replacement in perimenopause related depression: a preliminary report. Am J Obstet Gynecol 2000; 183: 414–20.
35. Slavin L, Lee C. Mood and symptom reporting among middle-aged women: the relationship between menopausal status, hormone replacement therapy and exercise participation. Health Psychol 1997; 16: 203–8.
36. Blumenthal JA, Babyak MA, Moore KA, Craighead WE, Herman S, Khatri P, Waugh R, Napolitano MA, Forman LM, Appelbaum M, Doraiswamy PM, Krishnan KR. Effects of exercise training on older patients with major depression. Arch Intern Med 1999; 25;159 (19): 2349–56.
37. Hammar M, Berg C, Lindgren R. Does physical exercise influence the frequency of postmenopausal hot flushes? Acta Obstet Gynecol Scand 1990; 69: 409–12.
38. Harlow BL, Cohen LS, Otto MW, Spiegelman D, Cramer DW. Prevalence and predictors of depressive symptoms in older premenopausal women: the Harvard Study of Moods and Cycles. Arch Gen Psychiatry 1999; 56 (5): 418–24.
39. Son BK, Markovitz JH, Winders S, Smith D. Smoking, nicotine dependence, and depressive symptoms in the CARDIA Study: effects of educational status. Am J Epidemiol 1997; 145: 110–6.
40. Glassman AH, Helzer JE, Covey LS, Cottler LB, Stetner F, Tipp JE, Johnson J. Smoking, smoking cessation, and major depression. JAMA 1990; 264: 1246–9.
41. Glassman AH. Cigarette smoking: implications for psychiatric illness. Am J Psychiatry 1993; 15: 546–53.
42. Hall SM, Munoz RF, Reus VI, Sees KL. Nicotine, negative affect, and depression. J Consult Clin Psychol 1993; 61: 761–7.
43. Perez-Stable EJ, Marin G, Marin BV, Katz MH. Depressive symptoms and cigarette smoking among Latinos in San Francisco. Am J Public Health 1990; 80: 1500–2.
44. Kaplan GA, Roberts RE, Camacho TC, Coyne JC. Psychosocial predictors of depression: prospective evidence from the Human Population Laboratory Studies. Am J Epidemiol 1987; 125: 206–20.
45. Black SA, Markides KS, Miller TQ. Correlates of depressive symptomatology among older community-dwelling Mexican Americans: the Hispanic EPESE. J Gerontol Soc Sci 1998; 53B: S198–S208.
46. Eaton WW, Kessler LG. Rates of symptoms of depression in a national sample. Am J Epidemiol 1981; 114: 528–38.
menopause; depression; climacteric symptoms
Copyright © 2001 by American Psychosomatic Society
Highlight selected keywords in the article text.