R

Tp

Ra

Ub

a

ARA

KPDVMH

C

1

bt[

Hb

0h

Maturitas 77 (2014) 111– 117

Contents lists available at ScienceDirect

Maturitas

journa l h om epa ge: www.elsev ier .com/ locate /matur i tas

eview

he association between vasomotor symptoms and depression duringerimenopause: A systematic review

oisin Worsleya,b,∗, Robin Bell a, Jayashri Kulkarnib, Susan R. Davisa

Women’s Health Research Program, School of Public Health and Preventive Medicine; Department of Epidemiology and Preventive Medicine, Monashniversity, 99 Commercial Road, Melbourne, VIC 3004, AustraliaMonash Alfred Psychiatry Research Centre, Monash University, Level 4, 607 St Kilda Road, Melbourne, VIC 3004, Australia

r t i c l e i n f o

rticle history:eceived 14 November 2013ccepted 24 November 2013

a b s t r a c t

There is a high incidence of depression in women presenting to menopause clinics. The aim of this reviewwas to determine if there is an association between depressive symptoms or major depressive disorder(MDD) and vasomotor symptoms (VMS). A systematic review of the literature was conducted accordingto PRISMA guidelines. 33 relevant publications were found, 12 from three large studies. Overall, we found

eywords:erimenopauseepressionasomotorajor depressive disorder

that there is a bidirectional association between VMS and depressive symptoms. This has been establishedin well-conducted, large observational studies. There does not appear to be a relationship between VMSand MDD. However, studies examining VMS and MDD were prone to bias making it difficult to draw anyconclusions.

© 2013 Elsevier Ireland Ltd. All rights reserved.

ot flushes

ontents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1112. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1123. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

3.1. Studies that have reported on VMS and depressive symptoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1123.1.1. Cross-sectional studies that have reported on VMS and depressive symptoms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1123.1.2. Directionality of association between VMS and depressive symptoms; findings from cohort studies . . . . . . . . . . . . . . . . . . . . . . . . . . . 113

3.2. Studies that have reported on VMS and major depressive disorder. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1134. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1135. Conclusion. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115Competing interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115Provenance and peer review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

. Introduction

Vasomotor and psychological symptoms cause a significant

symptoms (VMS) increases in the menopause transition and aftermenopause [6]. There is universal acceptance that the biologicalchanges associated with the menopause are the cause of hot

urden for many women during and following the menopauseransition [1,2], impacting on their quality of life [3], work ability4] and relationships [5]. The incidence and severity of vasomotor

∗ Corresponding author at: Women’s Health Research Program, School of Publicealth and Preventive Medicine Monash University, 99 Commercial Road, Mel-ourne, VIC 3004, Australia. Tel.: +61 399030827; fax: +61 399030828.

E-mail address: roisin.worsley@monash.edu (R. Worsley).

378-5122/$ – see front matter © 2013 Elsevier Ireland Ltd. All rights reserved.ttp://dx.doi.org/10.1016/j.maturitas.2013.11.007

flushes and night sweats. Debate exists about the extent to whichother symptoms in particular mood disturbance, relates to themenopause transition. The perimenopause has been described as a‘window of vulnerability’ for depression [7], though not all authorsagree with this notion [8]. In clinical practice, many womenpresenting to menopause clinics feel depressed and attribute their

symptoms to menopausal changes. There is a high incidence ofboth self reported depressive symptoms and clinically diagnoseddepression in these women [9–12].

1 turita

aom

2

[wiwFsopae

sCftccgcfiddl

i2

nc(ddap

tniscSnticoofmsuI

sra

12 R. Worsley et al. / Ma

The objective of this review was to examine evidence for anssociation between VMS and self-reported depressive symptomsr major depressive disorder (MDD) in otherwise healthy peri-enopausal women.

. Methods

The review was conducted according to PRISMA guidelines13]. Longitudinal and cross-sectional studies published in Englishere considered without any restriction on publication date. Stud-

es were considered eligible if they involved: (1) perimenopausalomen (as defined by pattern of menstrual bleeding and/or serum

SH measure), (2) women without major illness (other than depres-ion), (3) at least one measure of VMS, and (4) a validated measuref depression. Studies were excluded if the sample included onlyremenopausal or postmenopausal women, or only women with

medical illness such as breast cancer. Clinical trials were alsoxcluded.

Cochrane reviews and published peer reviews in the area wereearched to identify relevant search terms. Ovid Medline, EMBASE,INAHL and PsychInfo were searched using combinations of theollowing terms: hot flashes, hot flushes, night sweats, vasomo-or symptoms, depression, mood disorders, depressive disorders,limacteric, menopause, perimenopause, menopause transition,ohort studies, epidemiological studies, cross sectional studies, lon-itudinal studies, retrospective studies, prospective studies. Afterombining search terms the following limits were used: human,emale, English language. The search strategy for Medline is shownn Appendix I. The last search was run on 2 July 2013. Papers fromifferent databases were collated in EndNote, allowing automaticetection and then manual deletion of duplicate articles. The title

ist was then manually reviewed again for duplicates.Supplementary material related to this article can be found,

n the online version, at http://dx.doi.org/10.1016/j.maturitas.013.11.007.

Data retrieved from eligible studies included: study design;umber of participants; whether sampling was random or usedonvenience methods; the setting participants were recruited fromcommunity or clinical); type of VMS measure used; which vali-ated depression measure was used and the cut off score used toefine depressive symptoms; the relationship found, in particular,n odds ratio. Where possible, the data extracted was specific toerimenopausal women within each study.

Depression is very broad concept, from transient mood changeshrough to persistent debilitating melancholic depression witheurovegetative symptoms. Hence the term “depressive disorders”

s used to encompass a number of conditions with the commonymptom of lowered mood. A commonly used, highly validatedlassification system of depressive disorders is the Diagnostic andtatistical Manual (DSM) version 5 [14]. The criteria for the diag-osis of MDD requires five or more of the following symptomso be present for at least 2 weeks: depressed mood, diminishednterest, significant weight changes, sleep pattern changes, psy-homotor agitation or retardation, loss of energy, feeling worthlessr inappropriate guilt, decreased concentration, suicidal ideationr attempts. Exclusion criteria include recent losses or other causalactors such as substance use or a medical illness, the presence of

anic episodes, or the diagnosis of psychotic disorders. In researchettings MDD is diagnosed using a validated clinical interview,ndertaken by a skilled clinical rater, such as the Structured Clinical

nterview for the DSM (SCID) [15].

There are many depressive symptoms that cause considerable

uffering for the individual but do not meet the diagnostic crite-ia for MDD. Subsyndromal symptoms can be just as debilitatings MDD. Examples of such symptoms include fluctuating anger,

s 77 (2014) 111– 117

irritability, anxiety, suicidality, cognitive changes, paranoia, lossof libido, transient changes in sleep and appetite and low energy.Screening for depressive symptoms in the general population withself-rating scales such as the Center for Epidemiological StudiesDepression Scale (CES-D) [16] can yield very different results com-pared to diagnosing MDD with the SCID [17]. Therefore, the resultsof included studies have been divided into those measuring depres-sive symptoms, and those that examine formally diagnosed MDD.

An assessment of the risk of bias was conducted on all includedstudies using the 11-item tool validated by Hoy [18]. This toolincludes items that assess the external and internal validity ofprevalence studies, with an overall rating of low (L), moderate (M)or high (H) risk of bias. Studies rated as M or H are less likely toaccurately reflect the actual prevalence of any particular outcomein the general or target population [18]. The full bias assessmentfor each study is included in Appendix II.

Supplementary material related to this article can be found,in the online version, at http://dx.doi.org/10.1016/j.maturitas.2013.11.007.

3. Results

The search of Medline, PsychInfo, CINAHL and Embase retrieved326 articles. After title screening, 107 papers remained and allabstracts were then reviewed. After reviewing abstracts 66 papersremained for which the full text was read. Of these 31 met inclu-sion criteria. A further 3 papers were found through searching thebibliographies of included studies. Three large longitudinal stud-ies accounted for a large number of the included publications; 5from The Study of Women’s Health Across the Nation (SWAN); 4from the Seattle Midlife Women’s Health Study (SMWH); and 3from the Penn Ovarian Aging Study (POA). Different publicationsfrom the same studies have been included in the following tables,as different aspects of the association between VMS and depressionhave been assessed. It should be noted that multiple publicationsfrom one study do not represent independent analyses as the samewomen are being reanalyzed. Therefore, it has been clearly identi-fied where a publication has been derived from SWAN, SMWH orPOA.

3.1. Studies that have reported on VMS and depressive symptoms

Cross-sectional studies can provide evidence for an associationbut not directionality so cross-sectional and longitudinal studiesare reported separately below.

3.1.1. Cross-sectional studies that have reported on VMS anddepressive symptoms

Seventeen cross-sectional studies that assessed depressivesymptoms and VMS were found (Table 1). Eleven studies involvedcommunity dwelling women and seven involved women in clinicalsettings. The studies ranged in size from 70 to 1280 participants,with a total of 9615 women across all studies. In four studies dataspecific to perimenopausal women was extracted, whereas otherstudies presented analyses based on pooled data of women frommultiple menopausal stages.

Six cross-sectional studies found no relationship between VMSand depressive symptoms [19–24]. Two studies performed a factoranalysis finding only a weak relationship [25,26]. A statistically sig-nificant relationship was found in 9 of 17 studies [27–34] with oddsratios ranging between 1.27 (1.08–1.51) [35] and 8.1 (2.5–26.4)

[28]. Estimates of odds ratios were lower (1.27–1.76) in studiesthat controlled for multiple factors known to be associated withVMS or depression such as demographics, menopausal status andinsomnia [29,30,33,35].

R. Worsley et al. / Maturitas 77 (2014) 111– 117 113

Table 1Cross-sectional studies that assessed the association between VMS and depressive symptoms.

Reference Sample Sample size Age Validateddepression scaleused

Association Finding Risk of bias

Hunter (1986) Convenience, clinical 682 45–65 WHQ Yes Correlationbetween factors0.17 (p < 0.05)

M

Woods (1997) SMWH Random, community 337 35–55 CES-D, SCL90 No LBosworth (2001) Random, community 581 45–54 aCES-D No OR 0.72 (0.39–1.34) MJoffe (2002) Convenience, primary care 141 40–60 CES-D Yes OR 4.27

(1.33–13.66)M

Schmidt (2002) Convenience, clinical 92 40–55 CES-D No MFreeman (2003) POA Random, community 324 38–52 CES-D Yes Pearson correlation

coefficient 0.24p < 0.001

L

Blumel (2004) Convenience 300 40–59 GCS Yes OR 8.1 (2.5–26.4) MJuang (2005) Random, community 1273 40–54 HADS Yes OR 1.76 (1.03–3.04) LGallicchio (2007) Random, community 634 45–54 CES-D No HLi (2008) Convenience, community 1280 45–59 Zung Yes OR 1.3 (1.11–1.58) H

Thurston (2008) SWAN Random, community 1042 49–60 Composite scoreincluding CES-D

Yes OR 1.27 (1.08–1.51)for VMS distress

L

Brown (2009) SWAN 639 45–54 CES-D No OR 1.48 (1.0–2.19)for any VMS

M

Reed (2009) Convenience, clinical 770 45–70 PHQ-8 Yes OR 1.67 (1.04–2.68) MFlores-Ramos (2010) Convenience, clinical 141 45–55 CES-D Yes Women with VMS

had a CE-D score5.23 points higherthan those withoutp = .018

H

Yen (2009) Convenience, community 89 40–60 CES-D No OR 1.43 (0.76–2.68) MOpperman (2012) Random, community 324 36–62 SRQ-20 Yes OR 1.61 (1.09–2.34) LHunter (2013) Convenience, community 896 45–55 WHQ Yes OR 6.64 (3.29–13.4) M

L, low; M, moderate; H, high; CES-D, Center for Epidemiological Studies Depression scale; aCES-D, abbreviated CES-D; GCS Greene Climacteric Scale; HADS, Hospital Anxietyand Depression Scale; POA, Penn Ovarian Aging Study; PHQ, Patient Health Questionnaire; SCL-90, Symptom Checklist; SNWH, Seattle Midlife Women’s Health Study; SRQ,S WHQ,

Vtwsa

3s3Vsfipwd[b

3ssTssbtr

elf Reporting Questionnaire; SWAN, Study of Women’s Health Across the Nation;

Overall, these studies support a positive association betweenMS and depressive symptoms. Studies that reported an associa-

ion tended to be larger than those that did not, otherwise thereere no specific methodological differences. The majority of these

tudies had a moderate to high risk of bias because of poor internalnd external validity.

.1.2. Directionality of association between VMS and depressiveymptoms; findings from cohort studies.1.2.1. Likelihood of women with depressive symptoms developingMS. Two studies examined the risk of women with high depres-ive symptom scores subsequently developing VMS (Table 2) withollow-up length of one and ten years [36,37]. Both reported a pos-tive association, though Woods found this only in women withersistently high depression scores [36]. In Freeman’s study of 170omen, followed over ten years, the odds ratio of women withepressive symptoms developing hot flushes was 3.06 (1.43–6.58)37]. These were both well-conducted studies each with low risk ofias.

.1.2.2. Likelihood of women with VMS developing depressiveymptoms. Ten studies assessed the risk of women with VMS sub-equently developing depressive symptoms (Table 3) [30,37–45].he length of follow-up ranged from 3 to 15 years, with sampleizes of 36–3292. VMS were found to increase the risk of depres-

ive symptoms in eight of the studies [37,39–45] with odds ratiosetween 1.62 (1.43–1.84) [44] and 8.88 (2.57–30.68) [23]. Most ofhese studies were methodologically sound; they mostly had a lowisk of bias and adjusted for a considerable number of covariates.

Women’s Health Questionnaire; Zung Self Rating Depression Scale.

3.2. Studies that have reported on VMS and major depressivedisorder

Seven studies examined the association between VMS and MDD(Table 4) [11,43,46–50]. These studies were of a smaller scale withsample sizes from 29 to 266 participants. No statistically signifi-cant relationship was found in five of the seven studies. Schmidtreported a positive association, though no odds ratio was calcu-lated, and the sample size was very small with 29 participants [46].Overall, these studies do not support an association between VMSand MDD. However, most of these studies had a moderate to highrisk of bias because of small sample size, convenience sampling andhigh risk of missing information bias making it difficult to draw anyconclusions.

4. Discussion

The major finding of this review is that there is a positiveassociation between vasomotor and depressive symptoms duringperimenopause. This association is bidirectional, with women withdepressive symptoms more likely to develop VMS, and women withVMS more likely to develop depressive symptoms.

We found this association to be evident in both communitydwelling women and women presenting to clinical services. Wealso found the association to be independent of other knownrisk factors for depression including age, body mass index, mari-

tal status, education, health and financial status, perceived stress,insomnia, and sensitivity to symptoms [29,30,33,35,39,42,44].

We did not find an association between VMS and MDD dur-ing perimenopause. This dichotomy between depressive symptoms

114 R. Worsley et al. / Maturitas 77 (2014) 111– 117

Table 2Cohort studies that assessed the likelihood of women with depressive symptoms developing VMS.

Reference Sample Years offollow up

Samplesize

Age atbaseline

Validateddepression scaleused

Association Finding Risk ofbias

Woods (1996)SMWH

Random, community 1 347 35–55 CES-D Yes VMS distinguishedwomen withcontinuing vsresolvingdepression only

L

Freeman (2009)POA

Random, community, noVMS or depressive

10 170 35–47 CES-D Yes OR 3.06 (1.43–6.58) L

C g Stud

ad

ssirca

rttcl

s

TC

LW

symptoms at baseline

ES-D, Center of Epidemiological Studies Depression Scale; POA, Penn Ovarian Agin

nd MDD was also found in a review of the relationship betweenepression and the menopause transition [8].

Our finding of an association between VMS and depressiveymptoms but not MDD suggests that VMS are associated with sub-yndromal depressive symptoms. The severity of these symptomss not clear but it is possible that the women in these studies expe-ience moderate or severe depressive symptoms that do not persistonsistently for two consecutive weeks, but are intermittent or waxnd wane.

VMS are most closely linked with declining estradiol levels [6],aising the possibility that declining estradiol levels also contributeo depressive symptoms. If declining estradiol levels do contributeo depressive symptoms during perimenopause, an intermittentourse of symptoms would fit with the erratic nature of estradiol

evels known to occur during the perimenopause [51].

A shared biological pathway connecting VMS and depressiveymptoms is further supported by observations from clinical

able 3ohort studies that assessed the likelihood of women with VMS developing depressive sy

Reference Sample Years offollow up

Sample size Age atbaseline

Hunter(1990)

Convenience, clinical 3 36 45–55

Dennerstein(1999)

Random, community 6 354 45–55

Woods(2002)SMWH

Random, communityAt least 6 years ofmood data

6 201 Mean 41

Woods(2008)SMWH

At least one year ofmood data

15 302

Avis (2001) Random, community 3 309 43–53

Cohen(2006)

Random, community 4.9–7.6 326 36–45

POA Random, community

Freeman(2006) POA

No depressivesymptoms at baseline

8 23135–47

Freeman(2009) POA

No VMS or depressivesymptoms at baseline

10 170

Bromberger(2010)SWAN

Random, community 8 3292 42–52

Burleson(2010)

Convenience,community

5 55 42–52

, low; M, moderate; H, high; CES-D, Center for Epidemiological Studies Depression scale; Pomen’s Health Study; SWAN, Study of Women’s Health Across the Nation; WHQ, Wom

y; SMWH, Seattle Midlife Women’s Health Study.

practice. The most effective therapy for VMS is estrogen therapy[52], which may also be helpful for depressive symptoms [53].Conversely, low dose antidepressant therapies have some efficacyin reducing VMS [54,55].

Psychosocial factors are also important contributors to the asso-ciation. We found that psychological factors such as insomnia,perceived stress and anxiety were associated with both VMS anddepressive symptoms. Depressive symptom scales, such as theCES-D, may produce high scores in those with anxiety disorders[16,56,57–59]. The association between depressive symptom scoreand VMS may in part reflect the increase in anxiety symptoms seenat the menopause transition [60].

Many of the studies reviewed had methodological limitations,such as high risk of bias from poorly representative study popu-

lations, lack of random selection, missing information and usingunvalidated measures of VMS [11,22,30,32]. Small sample sizes[38,46] and unadjusted analyses limit the applicability of other

mptoms.

Validateddepressionscale used

Association Finding Risk of bias

WHQ No M

Affectometre-2

Yes A higher VMS scoreassociated with a highermean score for negativemood p < 0.001

L

CES-D Yes Apparent greater severityof VMS in women withresolving depressioncompared with otherpatterns p = 0.004

L

No

CES-D Yes OR 1.83 (1.02–3.28) LCES-D Yes OR 2.5 (1.1–5.8) L

OR 2.16 (1.07–4.33) LCES-D

Yes

OR 8.88 (2.57–30.68) L

CES-D Yes OR 1.62 (1.43–1.84) M

SCL-90 andmood diary

Yes Higher VMS scoreassociated with next daynegative mod in womenwith depression p < 0.01

M

OA, Penn Ovarian Aging Study; SCL-90, Symptom Checklist; SMWH, Seattle Midlifeen’s Health Questionnaire.

R. Worsley et al. / Maturitas 77 (2014) 111– 117 115

Table 4Studies that examined the association between VMS and MDD.

Reference Sample Design Years offollow up

Samplesize

Age atbaseline

Validateddiagnostic tool

Association Finding Risk of bias

Schmidt (2004) Convenience,community, no MDDat baseline

Longitudinal 5.2 29 40–50 SCID-IV Yes VMS morecommon in womenwho developeddepression p < 0.01(Fisher’s exact test)

M

Freeman (2006) POA Random, community,no past history ofMDD at baseline

Longitudinal 8 231 35–47 PRIME-MD No OR 1.34 (0.42–4.25) L

Ozturk (2006) Convenience, clinical Cross-sectional,case-controlled

141 48 SCID-I No M

Silva (2008) Convenience,menopause clinic

Cross-sectional 70 45 HDRS Yes Women with VMSmore likely to beclassified asdepressed p = 0.03

H

Steinberg (2008) Convenience, clinical,all with depression

Cross-sectional 116 40–55 SCID-IV No Presence of VMSnot associated withmajor vs minordepression

M

Bromberger (2009)SWAN

Random, communityNo past history ofMDD at baseline

7 266 46 SCID-IV No M

Bromberger (2011)SWAN

With and without apast history of MDD atbaseline

10 221 42–52 No OR 1.76 (1.0–1.09)

L depreso ome

sa

[rc[

asgai

diimVlump

5

caMt

C

mst

, low; M, moderate; H, high; HDRS, Hamilton Depression Rating Scale; MDD, major

f Mental Disorders; SCID, Structured Clinical Interview for DSM; SWAN, Study of W

tudies [24,28,32]. The small number of studies that assessed thessociation between VMS and MDD were particularly prone to bias.

The CES-D has been criticized for over-inflating scores in women61], which is concerning for a number of studies included in thiseview. However, the association was also found using a higherut off score of the CES-D [27] and multiple other instruments29–31,33,35,39,45].

The results of this review are limited by the lack of data avail-ble, which apply specifically to perimenopausal women, as mosttudies pooled the data from women in multiple menopausal statusroups. Furthermore, the definition of perimenopause varied from

description of bleeding patterns to more sophisticated definitionsncluding serum hormone levels.

Overall, we found a positive association between VMS andepressive symptoms. VMS and depressive symptoms are clin-

cally important. Depressive symptoms are associated with anncrease in the risk of ischemic heart disease [57], obesity [62],

etabolic syndrome [63,64] and reduced levels of functioning [65].MS, particularly severe VMS are associated with poorer quality of

ife [3], and an increased risk of coronary heart disease [66]. Bynderstanding the complex interplay between symptoms at peri-enopause, strategies can be developed to improve the health for

erimenopausal women.

. Conclusion

During perimenopause there is a positive, bidirectional asso-iation between VMS and depressive symptoms. There does notppear to be an association between VMS and clinically diagnosedDD but methodologically sound studies are needed to confirm

his finding.

ontributors

Dr. Worsley performed the literature search and wrote theanuscript. Professors Davis and Bell provided advice on the

earch strategy, inclusion criteria, tabulation of results and revisedhe manuscript. Professor Kulkarni provided advice regarding the

sive disorder; POA, Penn Ovarian Aging Study; PRIME-MD, Primary Care Evaluationn’s Health Across the Nation.

interpretation of results and revised the manuscript. All authorshave seen and approved the final draft.

Competing interest

The authors report no potential conflict of interest.

Funding

No funding was provided for the preparation of the manuscript.Dr Worsley is supported by an NHMRC postgraduate researchscholarship. Professor Davis is an Australian NHMRC PrincipalResearch Fellow (Grant No. 490939).

Provenance and peer review

Not commissioned, externally peer reviewed.

References

[1] Mishra GD, Kuh D. Health symptoms during midlife in relation tomenopausal transition: British prospective cohort study. British Medical Jour-nal 2012;344:e402.

[2] Dibonaventura MD, Chandran A, Hsu MA, Bushmakin A. Burden of vasomotorsymptoms in France, Germany, Italy, Spain, and the United Kingdom. Interna-tional Journal of Women’s Health 2013;5:261–9.

[3] Avis NE, Colvin A, Bromberger JT, et al. Change in health-related qualityof life over the menopausal transition in a multiethnic cohort of middle-aged women: study of Women’s Health Across the Nation. Menopause2009;16(September–October (5)):860–9.

[4] Geukes M, van Aalst MP, Nauta MC, Oosterhof H. The impact of menopausalsymptoms on work ability. Menopause 2012;19(March (3)):278–82.

[5] Woods NF, Mitchell ES. Symptom interference with work and relationships dur-ing the menopausal transition and early postmenopause: observations from theSeattle Midlife Women’s Health Study. Menopause 2011;18(June (6)):654–61.

[6] Freeman EW, Sammel MD, Lin H, et al. Symptoms associated with menopausal

transition and reproductive hormones in midlife women. Obstetrics and Gyne-cology 2007;110(2 (Pt 1)):230–40.

[7] Soares CN, Zitek B. Reproductive hormone sensitivity and risk for depressionacross the female life cycle: a continuum of vulnerability? Journal of Psychiatryand Neuroscience 2008;33(July (4)):331–43.

1 turita

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

16 R. Worsley et al. / Ma

[8] Judd FK, Hickey M, Bryant C. Depression and midlife: are we overpathol-ogising the menopause? Journal of Affective Disorders 2012;136(February(3)):199–211.

[9] Hay AG, Bancroft J, Johnstone EC. Affective symptoms in women attendinga menopause clinic. British Journal of Psychiatry: Journal of Mental Science1994;164(April (4)):513–6.

10] Seritan AL, Iosif AM, Park JH, DeatherageHand D, Sweet RL, Gold EB.Self-reported anxiety, depressive, and vasomotor symptoms: a study of per-imenopausal women presenting to a specialized midlife assessment center.Menopause 2010;17(2):410–5.

11] Silva M, Brito LMO, Chein MBdC, Brito LGO, Navarro P. Depression in climactericwomen: analysis of a sample receiving care at a university hospital in Maran-hao, Brazil. Revista de Psiquiatria do Rio Grande do Sul 2008;30(May–August(2)):150–4.

12] Stewart DE, Boydell K, Derzko C, Marshall V. Psychologic distress during themenopausal years in women attending a menopause clinic. International Jour-nal of Psychiatry in Medicine 1992;22(3):213–20.

13] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reportingsystematic reviews and meta-analyses of studies that evaluate health careinterventions: explanation and elaboration. Journal of Clinical Epidemiology2009;62(October (10)):e1–34.

14] Association AP. Diagnostic and statistical manual of mental disorders. 5th ed.Arlington, VA: American Psychiatric Publishing; 2013.

15] Spitzer RL, Williams JB, Gibbon M, First MB. The structured clinical interviewfor DSM-III-R (SCID): I. History, rationale, and description. Archives of GeneralPsychiatry 1992;49(August (8)):624–9.

16] Breslau N, Depressive symptoms. major depression, and generalized anxiety:a comparison of self-reports on CES-D and results from diagnostic interviews.Psychiatry Research 1985;15(July (3)):219–29.

17] Pignone M, Gaynes B, Rushton J, et al. Screening for depression systematic evi-dence review. North Carolina: Research Triangle Institute, University of NorthCarolina; 2002.

18] Hoy D, Brooks P, Woolf A, et al. Assessing risk of bias in prevalence studies:modification of an existing tool and evidence of interrater agreement. Journalof Clinical Epidemiology 2012;65(September (9)):934–9.

19] Woods NF, Mitchell ES. Pathways to depressed mood for midlife women: obser-vations from the Seattle Midlife Women’s Health Study. Research in Nursingand Health 1997;20(2):119–29.

20] Bosworth HB, Bastian LA, Kuchibhatla MN, et al. Depressive symptoms,menopausal status, and climacteric symptoms in women at midlife. Psycho-somatic Medicine 2001;63(July–August (4)):603–8.

21] Schmidt PJ, Murphy JH, Haq N, Danaceau MA, St Clair L. Basal plasma hor-mone levels in depressed perimenopausal women. Psychoneuroendocrinology2002;27(November (8)):907–20.

22] Gallicchio L, Schilling C, Miller SR, Zacur H, Flaws JA. Correlates of depressivesymptoms among women undergoing the menopausal transition. Journal ofPsychosomatic Research 2007;63(September (3)):263–8.

23] Brown JP, Gallicchio L, Flaws JA, Tracy J. Relations among menopausal symp-toms, sleep disturbance and depressive symptoms in midlife. Maturitas2009;62(February (2)):184–9.

24] Yen J-Y, Yang M-S, Wang M-H, Lai C-Y, Fang M-S. The associations betweenmenopausal syndrome and depression during pre-, peri-, and postmenopausalperiod among Taiwanese female aborigines. Psychiatry and Clinical Neuro-sciences 2009;63(October (5)):678–84.

25] Hunter M, Battersby R, Whitehead M. Relationships between psycho-logical symptoms, somatic complaints and menopausal status. Maturitas1986;8(October (3)):217–28.

26] Freeman EW, Sammel MD, Liu L, Martin P. Psychometric properties of amenopausal symptom list. Menopause 2003;10(3):258–65.

27] Joffe H, Hall JE, Soares CN, et al. Vasomotor symptoms are associated withdepression in perimenopausal women seeking primary care. Menopause2002;9(6):392–8.

28] Blumel JE, Castelo-Branco C, Cancelo MJ, et al. Relationship between psycho-logical complaints and vasomotor symptoms during climacteric. Maturitas2004;49(3):205–10.

29] Juang KD, Wang SJ, Lu SR, Lee SJ, Fuh JL. Hot flashes are associated with psy-chological symptoms of anxiety and depression in peri- and post- but notpremenopausal women. Maturitas 2005;52(2):119–26.

30] Li Y, Yu Q, Ma L, Sun Z, Yang X. Prevalence of depression and anxiety symptomsand their influence factors during menopausal transition and postmenopausein Beijing city. Maturitas 2008;61(3):238–42.

31] Reed SD, Ludman EJ, Newton KM, et al. Depressive symptoms and menopausalburden in the midlife. Maturitas 2009;62(3):306–10.

32] Flores-Ramos M, Heinze G, Silvestri-Tomassoni R. Association between depres-sive symptoms and reproductive variables in a group of perimenopausalwomen attending a menopause clinic in Mexico City. Archives of Women’sMental Health 2010;13(1):99–105.

33] Oppermann K, Fuchs SC, Donato G, Bastos CA, Spritzer PM. Physical, psycho-logical, and menopause-related symptoms and minor psychiatric disorders ina community-based sample of Brazilian premenopausal, perimenopausal, andpostmenopausal women. Menopause 2012;19(3):355–60.

34] Hunter MS, Gupta P, Chedraui P, et al. The International Menopause Study of Cli-mate, Altitude, Temperature (IMS-CAT) and vasomotor symptoms. Climacteric2013;16(1):8–16.

35] Thurston RC, Bromberger JT, Joffe H, et al. Beyond frequency: who is mostbothered by vasomotor symptoms? Menopause 2008;15(5):841–7.

[

s 77 (2014) 111– 117

36] Woods NF, Mitchell ES. Patterns of depressed mood in midlife women; obser-vations from the Seattle Midlife Women’s Health Study. Research in Nursingand Health 1996;19(2):111–23.

37] Freeman EW, Sammel MD, Lin H. Temporal associations of hot flashes anddepression in the transition to menopause. Menopause 2009;16(4):728–34.

38] Hunter MS. Psychological and somatic experience of the menopause: a prospec-tive study. Psychosomatic Medicine 1990;52(3):357–67 [corrected].

39] Dennerstein L, Lehert P, Burger H, Dudley E. Mood and the menopausal transi-tion. Journal of Nervous and Mental Disease 1999;187(November (11)):685–91.

40] Woods NF, Mariella A, Mitchell ES. Patterns of depressed mood across themenopausal transition: approaches to studying patterns in longitudinal data.Acta Obstetricia et Gynecologica Scandinavica 2002;81(7):623–32.

41] Avis NE, Crawford S, Stellato R, Longcope C. Longitudinal study of hormonelevels and depression among women transitioning through menopause. Cli-macteric 2001;4(September (3)):243–9.

42] Cohen LS, Soares CN, Vitonis AF, Otto MW, Harlow BL. Risk for new onset ofdepression during the menopausal transition: the Harvard study of moods andcycles. Archives of General Psychiatry 2006;63(4):385–90.

43] Freeman EW, Sammel MD, Lin H, Nelson DB. Associations of hormones andmenopausal status with depressed mood in women with no history of depres-sion. Archives of General Psychiatry 2006;63(4):375–82.

44] Bromberger JT, Schott LL, Kravitz HM, et al. Longitudinal change in reproduc-tive hormones and depressive symptoms across the menopausal transition:results from the Study of Women’s Health Across the Nation (SWAN). Archivesof General Psychiatry 2010;67(6):598–607.

45] Burleson MH, Todd M, Trevathan WR. Daily vasomotor symptoms, sleep prob-lems, and mood: using daily data to evaluate the domino hypothesis inmiddle-aged women. Menopause 2010;17(January–February (1)):87–95.

46] Schmidt PJ, Haq N, Rubinow DR. A longitudinal evaluation of the relationshipbetween reproductive status and mood in perimenopausal women. AmericanJournal of Psychiatry 2004;161(12):2238–44.

47] Ozturk O, Eraslan D, Mete HE, Ozsener S. The risk factors and symptomatologyof perimenopausal depression. Maturitas 2006;55(2):180–6.

48] Steinberg EM, Rubinow DR, Bartko JJ, et al. A cross-sectional evaluationof perimenopausal depression. Journal of Clinical Psychiatry 2008;69(June(6)):973–80.

49] Bromberger JT, Kravitz HM, Matthews K, Youk A, Brown C, Feng W. Predictorsof first lifetime episodes of major depression in midlife women. PsychologicalMedicine 2009;39(1):55–64.

50] Bromberger JT, Kravitz HM, Chang YF, Cyranowski JM, Brown C, MatthewsKA. Major depression during and after the menopausal transition: studyof Women’s Health Across the Nation (SWAN). Psychological Medicine2011;41(9):1879–88.

51] Burger HG, Dudley EC, Robertson DM, Dennerstein L. Hormonal changes inthe menopause transition. Recent Progress in Hormone Research 2002;57:257–75.

52] de Villiers TJ, Gass ML, Haines CJ, et al. Global Consensus Statement onmenopausal hormone therapy. Maturitas 2013;74(April (4)):391–2.

53] Worsley R, Davis SR, Gavrilidis E, et al. Hormonal therapies for new onsetand relapsed depression during perimenopause. Maturitas 2012;73(October(2)):127–33.

54] De Sloover Koch Y, Ernst ME. Selective serotonin-reuptake inhibitors for thetreatment of hot flashes. Annals of Pharmacotherapy 2004;38(July–August(7/8)):1293–6.

55] Stearns V, Slack R, Greep N, et al. Paroxetine is an effective treatment for hotflashes: results from a prospective randomized clinical trial. Journal of Clin-ical Oncology: Official Journal of the American Society of Clinical Oncology2005;23(28):6919–30.

56] Dozeman E, van Schaik DJ, van Marwijk HW, Stek ML, van der Horst HE, Beek-man AT. The center for epidemiological studies depression scale (CES-D) is anadequate screening instrument for depressive and anxiety disorders in a veryold population living in residential homes. International Journal of GeriatricPsychiatry 2011;26(March (3)):239–46.

57] Fountoulakis KN, Bech P, Panagiotidis P, et al. Comparison of depressive indices:reliability, validity, relationship to anxiety and personality and the role ofage and life events. Journal of Affective Disorders 2007;97(January (1–3)):187–95.

58] Roberts RE, Rhoades HM, Vernon SW. Using the CES-D scale to screen fordepression and anxiety: effects of language and ethnic status. PsychiatryResearch 1990;31(January (1)):69–83.

59] Papassotiropoulos A, Heun R. Detection of subthreshold depression and sub-threshold anxiety in the elderly. International Journal of Geriatric Psychiatry1999;14(August (8)):643–50.

60] Bromberger JT, Kravitz HM, Chang Y, et al. Does risk for anxiety increase dur-ing the menopausal transition? Study of women’s health across the nation.Menopause 2013;20(May (5)):488–95.

61] Carleton RN, Thibodeau MA, Teale MJ, et al. The center for epidemiologic studiesdepression scale: a review with a theoretical and empirical examination of itemcontent and factor structure. PLoS ONE 2013;8(3):e58067.

62] Brown JM, Stewart JC, Stump TE, Callahan CM. Risk of coronary heart diseaseevents over 15 years among older adults with depressive symptoms. American

Journal of Geriatric Psychiatry: Official Journal of the American Association forGeriatric Psychiatry 2011;19(August (8)):721–9.

63] Lin KP, Liang TL, Liao IC, Tsay SL. Associations among depression, obesity, andmetabolic syndrome in young adult females. Biological Research for Nursing2013 (Epub ahead of print).

turitas

[

[

R. Worsley et al. / Ma

64] Marijnissen RM, Smits JE, Schoevers RA, et al. Association between metabolicsyndrome and depressive symptom profiles – sex-specific? Journal of AffectiveDisorders 2013;151(Dec (3)):1138–42.

65] Vahia IV, Meeks TW, Thompson WK, et al. Subthreshold depression and suc-cessful aging in older women. American Journal of Geriatric Psychiatry: Official

[

77 (2014) 111– 117 117

Journal of the American Association for Geriatric Psychiatry 2010;18(March(3)):212–20.

66] Gast GC, Pop VJ, Samsioe GN, et al. Vasomotor menopausal symptomsare associated with increased risk of coronary heart disease. Menopause2011;18(February (2)):146–51.