Elsevier

Maturitas

Volume 52, Issue 2, 16 October 2005, Pages 119-126
Maturitas

Hot flashes are associated with psychological symptoms of anxiety and depression in peri- and post- but not premenopausal women

https://doi.org/10.1016/j.maturitas.2005.01.005Get rights and content

Abstract

Objective:

To explore the relationship between anxiety, depression, vasomotor symptoms, and menopausal status among middle-aged women.

Design:

A population-based study involving a rural Taiwanese population. Participants received a structured questionnaire, which included the hospital anxiety and depression scale (HADS), gynecological history and a checklist of menopausal symptoms in the most recent 2 weeks.

Results:

A total of 1273 women with no history of surgical menopause and hormonal therapy history participated. The mean anxiety, depression, and total HADS scores were 4.3 ± 3.3, 3.3 ± 2.8 and 7.6 ± 5.3, respectively, and did not differ according to menopausal status. A total of 10.5% participants reported hot flashes within the previous 2 weeks. After controlling for educational status and insomnia, anxiety (6.0 ± 3.8 versus 4.1 ± 3.1) and depression scores (4.0 ± 3.3 versus 3.2 ± 2.7) were significantly higher (p < 0.001) compared with those without hot flashes. These differences were attributed to peri- and postmenopausal subjects.

Conclusions:

Hot flashes in peri- and postmenopausal women were associated with anxious and depressive symptoms in East Asian population with low prevalence of vasomotor symptoms.

Introduction

Whether depression is increased in peri- and postmenopausal women has been an issue of debate for decades. Earlier clinic-based studies found that depression is more frequent in postmenopausal women [1]. In contrast, large-scale community-based research shows that the prevalence of depression is not higher in peri- or postmenopausal women [2], [3]. The divergence of results may result from different sampling methods. Since women with either menopausal syndrome or depression are more likely to seek help in clinical settings, subjects drawn from these clinics may be different from the general population in community settings.

Recent studies have begun to find risk factors of depression during menopausal transition. For example, Bosworth et al. [4] suggested that it is the menopausal symptoms, rather than menopausal status, that are related to depression. The authors reported that a telephone interview of women wishing to discuss hormone therapy revealed that mood swing, troubled sleeping, and poor concentration can account for the association between menopausal status and depression. Furthermore, Joffe et al. [5] showed a more complicated relationship among vasomotor symptoms, menopausal status, and depression, by applying the Center for Epidemiological Studies Depression Scale (CES-D). In that study, perimenopausal women visiting general practice physicians were shown to experience increased frequencies of depression only if vasomotor symptoms were present. Pre-, postmenopausal, and perimenopausal women without vasomotor symptoms did not show an increased frequency of depression. It appears that the simultaneous presence of both vasomotor symptoms and menopausal changes is associated with depression in middle-aged women. However, these two studies were conducted among specific groups of women in clinical settings, and not in the general population. Since both hot flashes and depression may bring women to clinical settings, their association may be caused by sampling bias. More community-based studies are needed in order to elucidate the relationship between menopausal status, menopausal syndrome, and depression.

In addition to sampling issues, measurement bias may also influence the investigation of depression and menopausal symptoms. When two methods of measurement overlap in some area, they appear to be correlated even if they are not. Depression and certain physical conditions share some common symptoms, such as fatigue. Therefore, in the assessment of depression, these shared symptoms may overlap, resulting in confounding and measurement bias [6]. For example, night sweating may disturb sleep and thus increase depression scores that include insomnia as a symptom. A measurement of depression that accounts for these shared symptoms would potentially reduce confounding between depression and menopausal syndrome. Besides, none of the previous studies explored the relationship between anxiety and menopausal changes.

The purpose of the current study was to explore the relationship between depression, vasomotor symptoms, and menopausal status in a community population. We adopted the hospital anxiety and depression scale (HADS) [7] as a measurement tool. The HADS is a depression and anxiety scale that excludes physical symptoms of depression and anxiety; therefore, it prevents physical symptoms from confounding the measure of depression. Because insomnia may be a possible mediator between vasomotor symptoms and emotional symptoms, we also tried to control insomnia in the analysis. The present study was conducted among middle-aged women in a community-based setting. We hypothesized that there would be higher scores of depression and anxiety among subjects with hot flashes.

Section snippets

Subjects

The study is a part of the Kinmen Woman's Health Investigation (KIWI) [8] that assessed the health condition of middle-aged women in the community of Kinmen, Taiwan. Kinmen is an island of 176 km2 located 248 km (154 miles) west of Taiwan and 41 km (25 miles) east of Mainland China. It consists of four townships, with a total population of 51,060 in 1998. Using the KIWI cohort, we surveyed all female citizens between ages 40 and 54 in 1998 who resided in two of the four townships in Kinmen Island.

Characteristics of the study population

Of the 2256 targeted women, 1497 (66%) participated in the study. Most of the non-participants actually live in Taiwan and return to Kinmen occasionally, so we could not find them even after several house calls. The participants and non-participants did not differ in age (45.5 ± 4.0 versus 45.8 ± 4.2 years; t = −1.3, d.f. = 2.54, p = 0.19).

After excluding subjects with incomplete data (n = 56), mental retardation (n = 5), past history of hysterectomy (n = 50) and/or bilateral oophorectomy (n = 42), and HT users (

Discussion

In the present study, conducted with a community population of middle-aged women, we found that both anxiety and depression are associated with hot flashes but not menopausal status. The association remains significant even after insomnia has been controlled. The results are compatible with our hypothesis that the presence of hot flashes is a risk factor for psychological symptoms of anxiety and depression. The lack of association between hot flashes and anxiety and/or depression in the

Acknowledgements

This study was supported by grants from the National Health Research Institutes (NHRI-GT-EX89P923C, NHRI-GT-EX90-8923PC and NHRI-EX91-8923PC) and Taipei VGH.

References (17)

There are more references available in the full text version of this article.

Cited by (98)

  • Symptoms assessed in studies on perimenopausal depression: A narrative review

    2020, Sexual and Reproductive Healthcare
    Citation Excerpt :

    Finally, publications using the same sample as well as measuring the same relevant symptoms were removed. Altogether, 37 articles were considered eligible for inclusion in this review [17,28,37,48,53–85]. These consisted of eight longitudinal and 29 cross-sectional studies, encompassing a total of 52′489 participants aged 35–70 years.

  • Assessment of perimenopausal depression: A review

    2019, Journal of Affective Disorders
    Citation Excerpt :

    The studies consistently reported a high prevalence of depression in the menopausal transition. Although some of the studies found no differences between the three stages of menopausal transition regarding depressive symptoms (Almeida et al., 2016; Anniverno et al., 2017; Bosworth et al., 2001; Zang et al., 2016), most reported a particularly high prevalence of depressive symptoms in the peri‑ or postmenopause (Bromberger et al., 2007; Campbell et al., 2017; Gonçalves et al., 2013; Jafari et al., 2014; Joffe et al., 2002; Juang et al., 2005; Lin et al., 2013; Maartens et al., 2002; Pimenta et al., 2016; Tangen and Mykletun, 2008; Terauchi et al., 2013; Yen et al., 2009; Zainal, 2008). Seven studies reported a prevalence peak during the perimenopause (Campbell et al., 2017; Gonçalves et al., 2013; Joffe et al., 2002; Lin et al., 2013; Tangen and Mykletun, 2008; Yen et al., 2009; Zainal, 2008).

View all citing articles on Scopus
View full text