Research report

Gender differences in the treatment of patients with bipolar disorder:A study of 7354 patients

Alina Karanti a,n,1, Christian Bobeck b, Maja Osterman b, Mathias Kardell a, Dag Tidemalm c,Bo Runeson c, Paul Lichtenstein d, Mikael Landén a,d

a Institute of Neuroscience and Physiology, Department of Psychiatry and Neurochemistry, The Sahlgrenska Academy, University of Gothenburg,Gothenburg, Swedenb Statistics and Data Analysis, Department of Computer and Information Science, University of Linköping, Linköping, Swedenc Department of Clinical Neuroscience, Center for Psychiatric Research, Karolinska Institutet, Stockholm, Swedend Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden

a r t i c l e i n f o

Article history:Received 16 June 2014Received in revised form10 November 2014Accepted 29 November 2014Available online 8 December 2014

Keywords:Bipolar disorderGenderDrug therapyElectroconvulsive therapyPsychotherapy

a b s t r a c t

Background: Gender differences in treatment that are not supported by empirical evidence have beenreported in several areas of medicine. Here, the aim was to evaluate potential gender differences in thetreatment for bipolar disorder.Methods: Data was collected from the Swedish National Quality Assurance Register for bipolar disorder(BipoläR). Baseline registrations from the period 2004–2011 of 7354 patients were analyzed. Multiplelogistic regression analysis was used to study the impact of gender on interventions.Results: Women were more often treated with antidepressants, lamotrigine, electroconvulsive therapy,benzodiazepines, and psychotherapy. Men were more often treated with lithium. There were no genderdifferences in treatment with mood stabilizers as a group, neuroleptics, or valproate. Subgroup analysesrevealed that ECT was more common in women only in the bipolar I subgroup. Contrariwise, lamotriginewas more common in women only in the bipolar II subgroup.Limitations: As BipoläR contains data on outpatient treatment of persons with bipolar disorder inSweden, it is unclear if these findings translate to inpatient care and to outpatient treatment in othercountries.Conclusions: Men and women with bipolar disorder receive different treatments in routine clinicalsettings in Sweden. Gender differences in level of functioning, bipolar subtype, or severity of bipolardisorder could not explain the higher prevalence of pharmacological treatment, electroconvulsivetherapy, and psychotherapy in women. Our results suggest that clinicians' treatment decisions are tosome extent unduly influenced by patients' gender.

& 2014 Elsevier B.V. All rights reserved.

1. Introduction

Bipolar disorder is a serious psychiatric disease with an estimatedlifetime prevalence of 0.6% for bipolar disorder type I (BP I), 0.4% forbipolar disorder type II (BP II), and 2.4% for bipolar spectrumdisorders (BPS) according to a recent study conducted by the WorldHealth Organization (Merikangas et al., 2011). Although the lifetimeprevalence of bipolar disorder appears to be roughly equal in bothgenders (Angst, 1998; Bebbington and Ramana, 1995; Gold, 1998;Grant et al., 2005), a number of reports suggest gender differences inthe symptomatic picture among bipolar patients. Many, but not all,studies report that women with bipolar disorder are more likely to

suffer from subsyndromal depressed mood and dysphoria (Altshuleret al., 2010; Diflorio and Jones, 2010; Morgan et al., 2005; Rasgonet al., 2005a), even though the number of depressives episodes andthe time spent in syndromal depression do not differ between menand women (Baldassano et al., 2005; Diflorio and Jones, 2010; Grantet al., 2005; Hendrick et al., 2000; Kawa et al., 2005; Kessing, 2004;Suominen et al., 2009). A majority of studies shows that women aremore likely than men to be diagnosed with the BP II subtype, and toexperience hypomanic episodes (Angst, 1998; Baldassano et al.,2005; Cassano et al., 1992; Diflorio and Jones, 2010; Merikangaset al., 2011; Schneck et al., 2008). Finally, a number of studies pointout that women are more likely than men to suffer from mixedepisodes (Benazzi, 2003; Diflorio and Jones, 2010; Grant et al., 2005;Kessing, 2004, 2008; Suppes et al., 2005).

There is a range of treatment options for bipolar disorder,including mood stabilizers such as lithium, neuroleptics (both firstand second generation neuroleptics), electroconvulsive treatment

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/jad

Journal of Affective Disorders

http://dx.doi.org/10.1016/j.jad.2014.11.0580165-0327/& 2014 Elsevier B.V. All rights reserved.

n Corresponding author. Fax: þ46 31 829835.E-mail address: alina.karanti@vgregion.se (A. Karanti).1 Postal address: Sahlgrenska University Hospital, Blå Stråket 15, Floor 3,

SE-413 45 Gothenburg, Sweden.

Journal of Affective Disorders 174 (2015) 303–309

(ECT), psychotherapy, and psychoeducation. By and large, there isno clear-cut evidence suggesting that treatment response to thesemodalities differs between men and women, and hence no justifi-cation for adjusting the bipolar disorder treatment according togender. Data are inconclusive with respect to the response to moodstabilizers; some studies find possible gender differences (Hendricket al., 2000; Kawa et al., 2005; Leibenluft, 1996; Viguera et al., 2000)while others do not (Arnold et al., 2000; Dilsaver et al., 1993;Freeman et al., 1992; Post et al., 1987; Secunda et al., 1985; Swannet al., 1997; Viguera et al., 2001). There is no evidence to suggestthat bipolar men and women respond differently to ECT orpsychotherapy. High teratogenic risk in some mood stabilizers, suchas valproate and carbamazepine (Yonkers et al., 2004), as well asrisk for menstrual abnormalities and polycystic ovary syndrome(PCOS) (O’Donovan et al., 2002; Rasgon et al., 2005b), should,however, be taken into account when treating women. Reproduc-tive considerations for women of a fertile age could therefore leadto gender differences in prescribing patterns of these specificmedications. Kriegshauser et al. (2010) found that women havemore fear than men of gaining weight as an adverse effect ofmedication, which could affect the choice of medication in women.

The importance of gender to treatment choice for bipolar disorderhas not previously been systematically studied. Hence, it is notknown whether men and women are offered different treatmentsin clinical practice. In the STEP-BD study (Systematic TreatmentProgram for Bipolar Disorder), there was no gender difference inthe use of antidepressants in bipolar patients (Baldassano et al.,2005), but we lack data on potential gender differences with respectto the use of lithium, mood stabilizers, ECT, and psychotherapy inbipolar patients in routine clinical practice. We know from studies ofother medical disorders, such as coronary artery disease, that therehave been unjustified differences in treatment between men andwomen, and attention to this has led to adjustments in clinicalpractice (Heer et al., 2002; Tsuyuki et al., 1994).

The aim of this study was to survey whether men and womenwith bipolar disorder in psychiatric outpatient care receive differ-ent treatments. The null hypothesis was that we would find nogender difference in the treatment of bipolar disorder.

2. Subjects and methods

2.1. Sources of data and study population

Data were derived from the Swedish national quality assuranceregister for bipolar disorders (BipoläR). BipoläR contains individua-lized data concerning case-mix, medical interventions, and treatmentoutcomes. It was established in 2004 with the aim of improving thequality of care for bipolar patients in Sweden. The register capturesbasic clinical epidemiological data along with longitudinal data onthe natural history and clinical course of the disease. Data is collectedby the treating psychiatrist and entered into a web-based application.Participation in this register is voluntary for the clinician as well asthe patients, even though there have been incentives from healthcare providers to increase the clinician's rate of participation.Registering units include both private and public psychiatric out-patient health care units in Sweden.

The baseline data include a multi-axial psychiatric DSM-diagnosis[bipolar I, II, schizoaffective disorder or NOS, other psychiatricdisorders; axis II-disorders; somatic diseases; axis IV data; and aglobal assessment of functioning (GAF)]. Data on the number of life-time depressive, hypomanic, manic and mixed episodes, suicideattempts, educational level, occupation, sick benefits, history ofcompulsory institutional care, current psychotropic drugs, electro-convulsive treatment (ECT), psychological treatment, Clinical GlobalImpression Severity (CGI-S), weight, and height are also collected.

The treating physician receives reminders by email when the annualfollow-up is due. At the annual follow-up, the following data arecollected: psychiatric diagnoses as above, the number of depressive,hypomanic, manic, and mixed episodes during the past 12 months,suicide attempts, occupation, sick benefits, compulsory institutionalcare during the past 12 months, current psychotropic drug treatment,ECT, psychological treatment, psychoeducation, Clinical GlobalImpression Improvement (CGI-I), weight, and height. By the timeof data extraction for this study in 2011, 7354 baseline registrationswere analyzed to identify differences in treatment choice betweenbipolar men and women in Sweden.

2.2. Measures

Gender was the main exposure of interest in this study. Theoutcome measures were ECT, antidepressants, lithium, valproate,lamotrigine, mood stabilizers as a group (i.e., lithium, valproate,lamotrigine, carbamazepine, and oxcarbazepine), neuroleptics(both first and second generation), benzodiazepines, and psy-chotherapy. All these outcomes were binary measures (yes/no)derived from baseline registrations. Data on ECT was missing for168 individuals. Data on the use of benzodiazepines was onlyavailable for a subset of the patients (n¼4827).

As potential confounders in the model, we used age, bipolarsubtype, GAF-symptom score, comorbid anxiety disorder, comorbidsubstance use disorder, previous suicide attempts, and number ofdepressive, manic, and mixed episodes. “Comorbid anxiety disor-der” is coded as a group in BipoläR and includes panic disorder,agoraphobia, generalized anxiety disorder, social phobia, obsessive–compulsive disorder, and specific phobias, but not post-traumatic oracute stress disorders. The number of affective episodes was codedas “none”, “1–3 episodes”, and “4 or more episodes” (see Table 1).

2.3. Statistics

The statistical analyses were based on logistic regressions, wherefemale gender was chosen as the reference category. Thus, an odds ratioover one corresponds to an increased probability for a man to have acertain medicine or therapy. The significance level to reject the nullhypothesis was set to Po0.05. Separate analyses were conducted forbipolar types I and II. A subanalysis was performed for patients 45 yearsof age or younger, in order to investigate possible gender differences inthe use of valproate in light of adverse effects of valproate for women inreproductive age. We tested for multicollinearity using the varianceinflation factor (VIF). There were no signs of multicollinearity.

2.4. Ethics

The study was approved by the Regional Ethics Committee inGothenburg and conducted in accordance with the latest HelsinkiProtocol. All analyses were conducted on a de-identified dataset.

3. Results

Demographic and clinical characteristics of the analyzed group aresummarized in Table 1. There was an overrepresentation of women(61%) in our study populationwith amean (SD) age of 47.9 (23.3) yearsfor women and 50.1 (15.5) years for men. With respect to diagnosticsubgroups, 46% of the patients were diagnosed with BP I, 37% with BPII, 14% with NOS, and 3% with SAD (percentages not shown in table).Womenwere more likely than men to be diagnosed with BP II and BPNOS, while men were more likely to be diagnosed with BP I. Womenhad more depressive and mixed episodes, and men had more manicepisodes. Comorbid anxiety disorder as well as previous suicideattempts was more frequent in women while comorbid substance

A. Karanti et al. / Journal of Affective Disorders 174 (2015) 303–309304

abuse was more frequent in men. GAF functioning level was lower inwomen. Educational level did not differ significantly between genders.

The associations between gender and the types of interventionsin patients with any bipolar disorder are shown in Table 2a. Womenwere more likely than men to be treated with ECT, antidepressants,benzodiazepines, lamotrigine, and psychotherapy. The resultsremained after adjusting for age, bipolar subtype, GAF-symptomscore, comorbid anxiety disorder, comorbid substance use disorder,previous suicide attempts, and number of depressive, manic, andmixed episodes. Men with bipolar disorder were more likely thanwomen to be treated with lithium after adjusting for the sameconfounders. There were no significant gender differences withrespect to the treatment with neuroleptics, valproate, or use ofmood stabilizers as a group. The prescription of valproate did notshow any gender difference even when we restrict analyses to thereproductive age of women.

Separate analyses of bipolar type I and II groups revealed thatwomenwere treated with ECT more than men only in the bipolar Igroup. Conversely, lamotrigine was more common in women onlyin the bipolar type II group. Additionally, neuroleptics were morefrequently used in women than men in the bipolar I group, but notin the bipolar II group. Finally, mood stabilizers as a group weremore common in men than women in the bipolar I group, but notin the bipolar II group (see Tables 2b and 2c).

Pregnancy is a gender specific life event that affects medication.We therefore conducted a sensitivity analysis where we excludedthe 27 registrations that occurred during pregnancy. This did not

change the results, which was expected given the low number ofevents (data not shown).

4. Discussion

We analyzed individual data frommore than 7000 bipolar patientsto elucidate potential gender differences with respect to clinicalinterventions for bipolar disorder. The data were collected from andreflect routine clinical practice in Sweden. The study demonstrateddistinct gender differences. Women with bipolar disorder were morelikely to receive treatment with antidepressants, benzodiazepines, ECT,lamotrigine, and psychotherapy. Particularly in bipolar type I, womenwere more likely than men to be treated with ECT, antidepressants,benzodiazepines, psychotherapy, and neuroleptics. In the bipolar typeII group, women received antidepressants, lamotrigine, benzodiaze-pines and psychotherapy more often than men. Men with bipolardisorder were more likely treated with lithium than women, and alsowith mood stabilizers in the bipolar type I group.

Women's higher prevalence than men of psychotherapy, ECT, andseveral drug therapies could not be explained by gender differences inage, GAF score, bipolar subtype, comorbid anxiety disorder, comorbidsubstance use disorder, previous suicide attempts, or number ofdepressive, manic and mixed episodes.

4.1. Gender and antidepressants/anxiolytics

The finding that women with bipolar disorder are more likely toreceive antidepressants than men is in line with previous reportsshowing that women in general are more often prescribed antide-pressants (Simoni-Wastila, 1998), and receive more combinationtreatment than men (The National Board of Health and Welfare,2000). At odds with our finding, however, the STEP-BD project thatfocused on bipolar disorder found that the rates of antidepressant usewere equal among bipolar men and women (Baldassano et al., 2005).

A possible explanation for the higher frequency of antidepres-sant treatment among women with bipolar disorder is that womensuffered from more depressive episodes than men, but the resultswithstood adjustment for the number of depressive episodes.Women might, however, also spend more time with subsyndromaldepressive symptoms than men (Altshuler et al., 2010; Diflorio andJones, 2010; Morgan et al., 2005; Rasgon et al., 2005a), which wecould not adjust for and which hence might explain the difference.In the same vein, there are studies suggesting that women withbipolar disorder are more likely to suffer from anxiety disorder thanmen (Altshuler et al., 2010; Barnes and Mitchell, 2005; McIntyreet al., 2006; Sasson et al., 2003; Saunders et al., 2012; Strakowskiet al., 1992), which is a possible explanation for the higher use ofbenzodiazepines in women. However, women had higher use ofbenzodiazepines also after adjusting for comorbid anxiety disorderin our study. Lastly, clinicians would be more prone to prescribeantidepressants to women if there were established gender differ-ences in antidepressant efficacy. In unipolar depression, a possibleeffect of gender on response rates to antidepressants has beensuggested (Hildebrandt et al., 2003; Malhi et al., 2003). There arethus suggestive data that women of a fertile age with unipolardepression would respond better to SSRI and MAO inhibitors whilemen would respond better to tricyclics (Baca et al., 2004; Hamiltonet al., 1995; Kornstein et al., 2000; Quitkin et al., 2002; Raskin, 1974).There is also data suggesting that this gender difference disappearswith age so that there is no difference between men and post-menopausal women (Grigoriadis et al., 2003; Kornstein et al., 2000).Such a pattern would imply an effect of sex hormones on theresponse to antidepressants. However, these data have not beenconfirmed by other studies (Hildebrandt et al., 2003; Quitkin et al.,2002; Wohlfarth et al., 2004), and there are no studies on bipolar

Table 1Demographics, clinical characteristics, and bivariate statistics for 7354 bipolarpatients recorded in the quality register BipoläR 2004–2011.

Men(n¼2855)

Women(n¼4499)

P-value

Age, years, mean (SD) 50.1 (15.5) 47.9 (23.3) o0.01Diagnosis, n (%)BP I 1435 (50.3%) 1925 (42.8%) o0.01BP II 954 (33.4%) 1768 (39.3%) o0.01BP NOS 375 (13.1%) 679 (15.1%) 0.02SAD 91 (3.2%) 127 (2.8%) 0.43Education, n (%) 0.46Less than high school 682 (23.9%) 1114 (24.8%)Graduated high school 924 (32.4%) 1495 (33.2%)Graduated college or higher 1071 (37.5%) 1635 (36.3%)Unknown 178 (6.2%) 255 (5.7%)GAF symptom, mean (SD) 65.3 (16.7) 63.2 (16.8) o0.01GAF function, mean (SD) 65.4 (17.6) 63.6 (17.4) o0.01Number of depressive episodes,n (%)

None 151 (5.3%) 167 (3.7%) o0.011–3 depressive episodes 771 (27.0%) 922 (20.5%) o0.014 or more depressive episodes 1933 (67.7%) 3410 (75.8%) o0.01Number of manic episodes, n(%)

None 1188 (41.6%) 2177 (48.4%) o0.011–3 manic episodes 956 (33.5%) 1359 (30.2%) o0.014 or more manic episodes 711 (24.9%) 963 (21.4%) o0.01Number of mixed episodes, n(%)

None 1890 (66.2%) 2669 (59.3%) o0.011–3 mixed episodes 483 (16.9%) 867 (19.3%) o0.014 or more mixed episodes 483 (16.9%) 962 (21.4%) o0.01Psychiatric comorbidityComorbid anxiety disorder, n (%) 231 (8.1%) 553 (12.3%) o0.01Comorbid substance abuse, n (%) 223 (7.8%) 174 (3.9%) o0.01Missing data on comorbidity 86 (3.0%) 166 (3.7%)Previous suicide attempts, n (%) 839 (29.4%) 1795 (39.9%) o0.01Missing data on suicide attempts 134 (4.7%) 171 (3.8%)

BP I¼bipolar disorder type I, BP II¼ bipolar disorder type II, BP NOS¼bipolardisorder not otherwise specified, SAD¼schizoaffective disorder of bipolar type,GAF¼Global assessment of functioning.

A. Karanti et al. / Journal of Affective Disorders 174 (2015) 303–309 305

depression. Hence, there is at present no evidence supporting thatwomen with bipolar depression would respond better to antide-pressants than men. Quite the reverse, there are studies suggestinga higher prevalence of mixed state and rapid cycling in women(Altshuler et al., 2010) and that these could be related to the use ofantidepressants. This would motivate a lower rate of antidepressantuse in women.

4.2. Gender and lithium

Our study shows that bipolar men are more likely to beprescribed lithium even when results are adjusted for bipolarsubtype and the number of manic episodes. The literature is scarceon the importance of gender to lithium prescriptions in routineclinical settings. One previous study by Bramness et al. (2009) forthe period 1981–2006 showed that more women than men usedlithium in Scandinavia, with the ratio of females to males beingeven higher in Sweden than in Norway and Denmark, but thosefigures can be explained by a higher prevalence of bipolar disorderper se in women than men (61:39 in this study).

Gender does not appear to influence the response to lithium(Arnold et al., 2000; Freeman et al., 1992; Rubinow and Moore,2004; Secunda et al., 1985; Swann et al., 1997; Viguera et al., 2000)even though one study suggested that women with bipolardisorder had a slightly superior treatment response to lithium,even in bipolar type II (Viguera et al., 2001).

4.3. Gender and other mood stabilizers

The use of lamotrigine was significantly more common inwomen with bipolar disorder but particularly pronounced in thebipolar type II group. Our results are adjusted for number ofdepressive episodes and therefore cannot be explained by thehigher rate of depressive episodes in the women. We have foundno study suggesting that the effect of lamotrigine depends ongender. It therefore remains to be explained why women are moreoften treated with lamotrigine in Sweden.

The likelihood of being treated with valproate does not differsignificantly between sexes in our study, which is surprising. Onewould expect that fertile women in particular would be treated with

Table 2aGender and odds ratios for interventions among 7354 patients with any bipolar disorder recorded in BipoläR 2004–2011.

N OR (95%CI) P aOR (95%CI) P

ECT 7186 0.80 (0.71–0.90) o0.001 0.80 (0.69–0.89) o0.001Antidepressants 7354 0.72 (0.65–0.79) o0.001 0.81 (0.73–0.91) o0.001Lithium 7354 1.39 (1.26–1.53) o0.001 1.25 (1.12–1.39) o0.001Valproate 7354 1.21 (1.04–1.40) 0.015 1.15 (0.98–1.35) 0.086Lamotrigine 7354 0.69 (0.61–0.77) o0.001 0.85 (0.75–0.97) 0.017Mood stabilizers 7354 1.15 (0.99–1.32) 0.060 1.14 (0.98–1.32) 0.098Neuroleptics 7354 0.97 (0.88–1.07) 0.561 0,93 (0.84–1.04) 0.220Benzodiazepines 4827 0.70 (0.62–0.79) o0.001 0.72 (0.63–0.83) o0.001Psychotherapy 7354 0.58 (0.53–0.64) o0.001 0.67 (0.60–0.74) o0.001

OR¼odds ratio. aOR: OR adjusted for age, bipolar subtype, GAF-symptom score, comorbid anxiety disorder, comorbid substance use disorder, previous suicide attempts,number of depressive, manic, and mixed episodes. An OR lower than 1 indicates lower likelihood for men to receive the treatment.

Table 2bGender and odds ratios for interventions among 3039 persons diagnosed with bipolar disorder type I in BipoläR 2004–2011.

N OR (95%CI) P aOR (95%CI) P

ECT 2971 0.66 (0.56–0.78) o0.001 0.72 (0.60–0.86) o0.001Antidepressants 3039 0.73 (0.63–0.85) o0.001 0.82 (0.69–0.96) 0.015Lithium 3039 1.25 (1.07–1.46) 0.005 1.22 (1.03–1.44) 0.019Valproate 3039 1.15 (0.94–1.40) 0.191 1.13 (0.91–1.40) 0.267Lamotrigine 3039 0.95 (0.77–1.17) 0.629 0.88 (0.71–1.10) 0.274Mood stabilizers 3039 1.31 (1.05–1.31) 0.017 1.33 (1.04–1.70) 0.021Neuroleptics 3039 0.82 (0.71–0.95) 0.007 0.79 (0.68–0.92) 0.003Benzodiazepines 1966 0.73 (0.61–0.87) o0.001 0.81 (0.67–0.99) 0.040Psychotherapy 3039 0.64 (0.56–0.74) o0.001 0.71 (0.62–0.84) o0.001

OR¼odds ratio. aOR: OR adjusted for age, bipolar subtype, GAF-symptom score, comorbid anxiety disorder, comorbid substance use disorder, previous suicide attempts,number of depressive, manic, and mixed episodes. An OR lower than 1 indicates lower likelihood for men to receive the treatment.

Table 2cGender and odds ratios for interventions among 2462 persons diagnosed with bipolar disorder type II in BipoläR 2004–2011.

N OR (95%CI) P aOR (95%CI) P

ECT 2462 1.03 (0.76–1.28) 0.764 0.98 (0.78–1.24) 0.891Antidepressants 2492 0.76 (0.65–0.88) o0.001 0.80 (0.67–0.95) 0.009Lithium 2492 1.39 (1.18–1.63) o0.001 1.22 (1.02–1.45) 0.027Valproate 2492 1.18 (0.89–1.56) 0.244 1.24 (0.92–1.66) 0.158Lamotrigine 2492 0.60 (0.50–0.71) o0.001 0.69 (0.57–0.84) o0.001Mood stabilizers 2492 1.02 (0.82–1.28) 0.861 1.05 (0.83–1.35) 0.673Neuroleptics 2492 1.00 (0.71–0.95) 0.982 1.12 (0.92–1.38) 0.262Benzodiazepines 1659 0.67 (0.54–0.81) o0.001 0.67 (0.54–0.83) o0.001Psychotherapy 2492 0.55(0.46–0.65) o0.001 0.62 (0.52–0.75) o0.001

OR¼odds ratio. aOR: OR adjusted for age, bipolar subtype, GAF-symptom score, comorbid anxiety disorder, comorbid substance use disorder, previous suicide attempts,number of depressive, manic, and mixed episodes. An OR lower than 1 indicates lower likelihood for men to receive the treatment.

A. Karanti et al. / Journal of Affective Disorders 174 (2015) 303–309306

valproate to a lesser extent considering the teratogenic effect ofvalproate as well as the risk for menstrual abnormalities and PCOS(Joffe et al., 2006a, 2006b; McIntyre et al., 2003; O’Donovan et al.,2002; Rasgon et al., 2005b; Yonkers et al., 2004). We ran a subanalysisfor the age span that corresponds to women's reproductive age –

where there should be a gender difference with less prescription towomen – but found no gender difference in the use of valproate. Thisfinding highlights the importance of informing clinicians on the effectsof valproate in fertile women.

The use of mood stabilizers as a group was higher in men thanwomen in the bipolar type I group, but not in the bipolar type IIgroup or in the total group of bipolar patients. The reason for thisgender difference is unclear.

4.4. Gender and neuroleptics

We did not see any gender difference in the prescription ofneuroleptics in the whole group of patients with bipolar disorder,but in the bipolar type I group neuroleptics were more common inwomen. Russo et al. (2002) found that females with bipolardisorder were more than 27% more likely than males to useneuroleptics. The use of neuroleptics and specifically second-generation neuroleptics has increased in the treatment of bipolardisorder in the past decade. It has been postulated that the responseto neuroleptics differs in women compared to men, and thisdifference at least partially can be related to gender differences inpharmacokinetics, i.e., absorption, distribution, bioavailability, andmetabolism (Anderson, 2008; Smith, 2010). Previous studies haveshown that women tend to require less neuroleptics than men, butthe results are not conclusive regarding possible gender differencesin treatment response (Pinals et al., 1996; Robinson et al., 1999;Salokangas, 2004; Szymanski et al., 1995; Usall et al., 2007).

4.5. Gender and ECT

An additional finding in this study is that women with bipolardisorder type I were more likely than men to be treated with ECT.This is in line with a recent study in Norway that found a male:female ratio of 1:2 (Schweder et al., 2011).

Even though women with bipolar disorder have a greater like-lihood of having subsyndromal depressive symptoms (Altshuleret al., 2010), those are not a clinical indication for ECT and areunlikely to explain the gender difference. One explanation could bethat women with depression may undergo fewer pharmacologicaltrials before being referred to ECT (Bloch et al., 2005), but thiscontradicts our finding that women are prescribed more antide-pressants than men. The efficacy of ECT is generally considered tobe unrelated to gender (Godfroid, 1999), but there is a relativepaucity of data regarding gender differences in ECT and moststudies have focused on unipolar depression. ECT has been foundto be more effective for female than male schizophrenic patients(Bloch et al., 2005). It has been suggested that women may haveless cognitive impairment after ECT than men (Sackeim et al., 1986),but also that women can be at higher risk of depressive relapse afterECT (Thase et al., 2000). The reason why ECT was more prevalent inwomen than men in the bipolar type I but not the type II group isunclear.

4.6. Gender and psychotherapy

In our study, female gender increased the likelihood for bipolarpatients to receive psychotherapeutic intervention. The results wereadjusted for age, bipolar subtype, GAF-symptom score, comorbidanxiety disorder, comorbid substance use disorder, previous suicideattempts, and number of depressive, manic or mixed episodes. Thereare few studies examining the role of gender in psychotherapy.

A literature review on unipolar depression did not show any clearevidence that gender influences the response to psychotherapyincluding cognitive behavioral therapy and interpersonal therapy(Parker et al., 2011). Specifically for bipolar disorder, there are no dataas to possible gender differences in response to psychotherapy. Giventhe lack of data suggesting that the outcome of psychotherapy differsbetween bipolar men and women, it should be investigated whywomen with bipolar disorder are referred more often to psychother-apy than men. One possible explanation could be that women have apreference to psychotherapy compared to men, but this has yet tobe shown.

4.7. Strengths and limitations of this study

BipoläR includes a large number of persons with bipolar disorder(over 7000 unique baseline registrations) and our study based onBipoläR constitutes the largest nationally representative study to-dateto examine information on gender differences in individuals withbipolar disorder. The psychiatrists who register patients are oftenspecialized in the treatment of mood disorders in general andtreatment of bipolar disorders in particular, which provide conditionsfor good validity and high data quality (Sellgren et al., 2011). BipoläRcontains much more collateral information and more detailed pheno-typic information than other Swedish national registers. This allowedus to adjust for potential confounders. It should, however, be notedthat at time of the study, BipoläR covered approximately 30% of allbipolar patients receiving outpatient care in Sweden. This is becausethe rate of participation in quality registers differs across countycouncils. Although there thus might be geographical bias, this isunlikely to have influenced the results which were controlled forage, bipolar subtype, GAF-symptom score, comorbid anxiety andsubstance abuse, previous suicide attempts, and number of moodepisodes. Regarding the generalizability of the findings, BipoläRcontains only psychiatric outpatient treatment data, and it is unknownwhether these findings translate to inpatient care or to patientstreated in primary care settings. It should also be noted that theremight be other variables that might differ between men and womenthat we could not control for since the register do not record theinformation. Such potential confounders include rapid cycling(Degenhardt et al., 2012), psychosis (Sato et al., 2002), migraineheadaches (Saunders et al., 2014), and the number of years of illness(Baldassano, 2006). On a final note, patients with bipolar disorderoften receive various combinations of available treatment options. Thelist of possible combinations is long and this study did not survey ifthese combinations differ by gender.

5. Conclusion

Our study demonstrates significant gender differences in theroutine clinical treatment of bipolar disorders in Sweden. Womenare more likely to be treated with antidepressants, ECT, lamotrigine,benzodiazepines, and psychotherapy, while men are more likely to betreated with lithium. The higher prevalence of medical interventionsin women could not be explained by differences in global assessmentof functioning, bipolar subtype, previous suicide attempts, number ofdepressive, mixed andmanic episodes, comorbid anxiety or substanceabuse. Given that there is scarce empirical evidence to suggest thattreatment for bipolar disorder should be adjusted according togender, our results suggest that clinicians' treatment decisions areto some extent unduly influenced by patients' gender. However,gender differences could also arise if preferences differed amongwomen and men, and could also arise if there was a gender bias inthe choice of doctors. We suggest that outcome research report datastratified on gender. Such data could justify or discourage differenttreatment regimens based on gender in the future. We finally suggest

A. Karanti et al. / Journal of Affective Disorders 174 (2015) 303–309 307

that since gender evidently plays a role in a naturalistic setting, futureguidelines need to address the evidence base for taking gender intoaccount when prescribing treatment for bipolar disorder.

Role of funding sourceThis research was supported by grants from the Regional Research and Develop-

ment Unit Västra Götaland FoU (VGFOUREG-158591), the Swedisfrh Medical ResearchCouncil (K2011-61X-14647-09-3, K2010-61X-21569-01-1, and K2010-61P-21568-01-4),the Swedish foundation for Strategic Research, the Swedish Brain foundation, and theSwedish Federal Government under the LUA/ALF agreement (ALFGBG-142041).

The funding source had no involvement in study design, data collection, dataanalysis, data interpretation, writing of the report, or in the decision to submit thepaper for publication.

Conflict of interestAlina Karanti, M.D.AK declares that, over the past three years, she has received compensation for

lectures from Eli Lilly Sweden. No other equity ownership, profit-sharing agree-ments, royalties, or patent.

- Consultant None- Grant/Research None- Speakers Bureau None- Major stockholder None- Other financial or Material support None

As of 11/06/14Bo Runeson, M.D., Ph.D.BR declares that he has received compensation for lectures from AstraZeneca,

Eli Lilly Sweden and Lundbeck pharmaceuticals. No other equity ownership, profit-sharing agreements, royalties, or patent.

- Consultant None- Grant/Research None- Speakers Bureau None- Major stockholder None- Other financial or Material support None

As of 11/06/14Mikael Landén, M.D., Ph.D.ML declares that, over the past three years, he has received compensation for

lectures from AstraZeneca, Biophausia, Bristol Myers-Squibb, Eli Lilly Sweden,Servier Sweden, and previously served at advisory board for Lundbeck pharma-ceuticals. No other equity ownership, profit-sharing agreements, royalties, orpatent.

- Consultant None- Grant/Research None- Speakers Bureau None- Major stockholder None- Other financial or Material support None

As of 11/06/14

AcknowledgmentsThe quality register BipoläR is acknowledged for providing data for this study.

This research was supported by Grants from the Regional Research and Develop-ment Unit Västra Götaland FoU (VGFOUREG-158591), the Swedish MedicalResearch Council (K2011-61X-14647-09-3, K2010-61X-21569-01-1, and K2010-61P-21568-01-4), the Swedish Foundation for Strategic Research (KF10-0039), theSwedish Brain Foundation (FO2014-0125), and the Swedish Federal Governmentunder the LUA/ALF agreement (ALFGBG-142041).

References

Altshuler, L.L., Kupka, R.W., Hellemann, G., Frye, M.A., Sugar, C.A., McElroy, S.L.,Nolen, W.A., Grunze, H., Leverich, G.S., Keck, P.E., Zermeno, M., Post, R.M.,Suppes, T., 2010. Gender and depressive symptoms in 711 patients with bipolardisorder evaluated prospectively in the Stanley Foundation bipolar treatmentoutcome network. Am. J. Psychiatry 167, 708–715.

Anderson, G.D., 2008. Gender differences in pharmacological response. Int. Rev.Neurobiol. 83, 1–10.

Angst, J., 1998. The emerging epidemiology of hypomania and bipolar II disorder.J. Affect. Disord. 50, 143–151.

Arnold, L.M., McElroy, S.L., Keck Jr., P.E., 2000. The role of gender in mixed mania.Compr. Psychiatry 41, 83–87.

Baca, E., Garcia-Garcia, M., Porras-Chavarino, A., 2004. Gender differences intreatment response to sertraline versus imipramine in patients with nonme-lancholic depressive disorders. Prog. Neuropsychopharmacol. Biol. Psychiatry28, 57–65.

Baldassano, C.F., 2006. Illness course, comorbidity, gender, and suicidality inpatients with bipolar disorder. J. Clin. Psychiatry 67 (Suppl 11), S8–S11.

Baldassano, C.F., Marangell, L.B., Gyulai, L., Ghaemi, S.N., Joffe, H., Kim, D.R.,Sagduyu, K., Truman, C.J., Wisniewski, S.R., Sachs, G.S., Cohen, L.S., 2005.Gender differences in bipolar disorder: retrospective data from the first 500STEP-BD participants. Bipolar Disord. 7, 465–470.

Barnes, C., Mitchell, P., 2005. Considerations in the management of bipolar disorderin women. Aust. N. Z. J. Psychiatry 39, 662–673.

Bebbington, P., Ramana, R., 1995. The epidemiology of bipolar affective disorder.Soc. Psychiatry Psychiatr. Epidemiol. 30, 279–292.

Benazzi, F., 2003. The role of gender in depressive mixed state. Psychopathology 36,213–217.

Bloch, Y., Ratzoni, G., Sobol, D., Mendlovic, S., Gal, G., Levkovitz, Y., 2005. Genderdifferences in electroconvulsive therapy: a retrospective chart review. J. Affect.Disord. 84, 99–102.

Bramness, J.G., Weitoft, G.R., Hallas, J., 2009. Use of lithium in the adult populationsof Denmark, Norway and Sweden. J. Affect. Disord. 118, 224–228.

Cassano, G.B., Akiskal, H.S., Savino, M., Musetti, L., Perugi, G., 1992. Proposedsubtypes of bipolar II and related disorders: with hypomanic episodes (orcyclothymia) and with hyperthymic temperament. J. Affect. Disord. 26,127–140.

Degenhardt, E.K., Gatz, J.L., Jacob, J., Tohen, M., 2012. Predictors of relapse orrecurrence in bipolar I disorder. J. Affect. Disord. 136, 733–739.

Diflorio, A., Jones, I., 2010. Is sex important? Gender differences in bipolar disorder.Int. Rev. Psychiatry 22, 437–452.

Dilsaver, S.C., Swann, A.C., Shoaib, A.M., Bowers, T.C., Halle, M.T., 1993. Depressivemania associated with nonresponse to antimanic agents. Am. J. Psychiatry 150,1548–1551.

Freeman, T.W., Clothier, J.L., Pazzaglia, P., Lesem, M.D., Swann, A.C., 1992. A double-blind comparison of valproate and lithium in the treatment of acute mania.Am. J. Psychiatry 149, 108–111.

Godfroid, I.O., 1999. Sex differences relating to psychiatric treatment. Can. J.Psychiatry 44, 362–367.

Gold, J.H., 1998. Gender differences in psychiatric illness and treatments: a criticalreview. J. Nerv. Ment. Dis. 186, 769–775.

Grant, B.F., Stinson, F.S., Hasin, D.S., Dawson, D.A., Chou, S.P., Ruan, W.J., Huang, B.,2005. Prevalence, correlates, and comorbidity of bipolar I disorder and axis Iand II disorders: results from the National Epidemiologic Survey on Alcohol andRelated Conditions. J. Clin. Psychiatry 66, 1205–1215.

Grigoriadis, S., Kennedy, S.H., Bagby, R.M., 2003. A comparison of antidepressantresponse in younger and older women. J. Clin. Psychopharmacol. 23, 405–407.

Hamilton, J.A., Grant, M., Jensvold, M.F., 1995. Sex and the Treatment of Depression:When Does It Matter? 1st ed American Psychiatric Press, Washington, DC.

Heer, T., Schiele, R., Schneider, S., Gitt, A.K., Wienbergen, H., Gottwik, M., Gieseler,U., Voigtlander, T., Hauptmann, K.E., Wagner, S., Senges, J., 2002. Genderdifferences in acute myocardial infarction in the era of reperfusion (the MITRAregistry). Am. J. Cardiol. 89, 511–517.

Hendrick, V., Altshuler, L.L., Gitlin, M.J., Delrahim, S., Hammen, C., 2000. Gender andbipolar illness. J. Clin. Psychiatry 61, 393–396 (quiz 397).

Hildebrandt, M.G., Steyerberg, E.W., Stage, K.B., Passchier, J., Kragh-Soerensen, P.,2003. Are gender differences important for the clinical effects of antidepres-sants? Am. J. Psychiatry 160, 1643–1650.

Joffe, H., Cohen, L.S., Suppes, T., Hwang, C.H., Molay, F., Adams, J.M., Sachs, G.S., Hall,J.E., 2006a. Longitudinal follow-up of reproductive and metabolic features ofvalproate-associated polycystic ovarian syndrome features: a preliminaryreport. Biol. Psychiatry 60, 1378–1381.

Joffe, H., Cohen, L.S., Suppes, T., McLaughlin, W.L., Lavori, P., Adams, J.M.,Hwang, C.H., Hall, J.E., Sachs, G.S., 2006b. Valproate is associated with new-onset oligoamenorrhea with hyperandrogenism in women with bipolar dis-order. Biol. Psychiatry 59, 1078–1086.

Kawa, I., Carter, J.D., Joyce, P.R., Doughty, C.J., Frampton, C.M., Wells, J.E., Walsh, A.E.,Olds, R.J., 2005. Gender differences in bipolar disorder: age of onset, course,comorbidity, and symptom presentation. Bipolar Disord. 7, 119–125.

Kessing, L.V., 2004. Gender differences in the phenomenology of bipolar disorder.Bipolar Disord. 6, 421–425.

Kessing, L.V., 2008. The prevalence of mixed episodes during the course of illness inbipolar disorder. Acta Psychiatr. Scand. 117, 216–224.

A. Karanti et al. / Journal of Affective Disorders 174 (2015) 303–309308

Kornstein, S.G., Schatzberg, A.F., Thase, M.E., Yonkers, K.A., McCullough, J.P., Keitner,G.I., Gelenberg, A.J., Davis, S.M., Harrison, W.M., Keller, M.B., 2000. Genderdifferences in treatment response to sertraline versus imipramine in chronicdepression. Am. J. Psychiatry 157, 1445–1452.

Kriegshauser, K., Sajatovic, M., Jenkins, J.H., Cassidy, K.A., Muzina, D., Fattal, O.,Smith, D., Singer, B., 2010. Gender differences in subjective experience andtreatment of bipolar disorder. J. Nerv. Ment. Dis. 198, 370–372.

Leibenluft, E., 1996. Women with bipolar illness: clinical and research issues. Am. J.Psychiatry 153, 163–173.

Malhi, G.S., Mitchell, P.B., Salim, S., 2003. Bipolar depression: management options.CNS Drugs 17, 9–25.

McIntyre, R.S., Mancini, D.A., McCann, S., Srinivasan, J., Kennedy, S.H., 2003.Valproate, bipolar disorder and polycystic ovarian syndrome. Bipolar Disord.5, 28–35.

McIntyre, R.S., Soczynska, J.K., Bottas, A., Bordbar, K., Konarski, J.Z., Kennedy, S.H.,2006. Anxiety disorders and bipolar disorder: a review. Bipolar Disord. 8,665–676.

Merikangas, K.R., Jin, R., He, J.P., Kessler, R.C., Lee, S., Sampson, N.A., Viana, M.C.,Andrade, L.H., Hu, C., Karam, E.G., Ladea, M., Medina-Mora, M.E., Ono, Y.,Posada-Villa, J., Sagar, R., Wells, J.E., Zarkov, Z., 2011. Prevalence and correlatesof bipolar spectrum disorder in the world mental health survey initiative. Arch.Gen. Psychiatry 68, 241–251.

Morgan, V.A., Mitchell, P.B., Jablensky, A.V., 2005. The epidemiology of bipolardisorder: sociodemographic, disability and service utilization data from theAustralian National Study of Low Prevalence (Psychotic) Disorders. BipolarDisord. 7, 326–337.

O’Donovan, C., Kusumakar, V., Graves, G.R., Bird, D.C., 2002. Menstrual abnormal-ities and polycystic ovary syndrome in women taking valproate for bipolarmood disorder. J. Clin. Psychiatry 63, 322–330.

Parker, G., Blanch, B., Crawford, J., 2011. Does gender influence response to differingpsychotherapies by those with unipolar depression? J. Affect. Disord. 130,17–20.

Pinals, D.A., Malhotra, A.K., Missar, C.D., Pickar, D., Breier, A., 1996. Lack of genderdifferences in neuroleptic response in patients with schizophrenia. Schizophr.Res. 22, 215–222.

Post, R.M., Uhde, T.W., Roy-Byrne, P.P., Joffe, R.T., 1987. Correlates of antimanicresponse to carbamazepine. Psychiatry Res. 21, 71–83.

Quitkin, F.M., Stewart, J.W., McGrath, P.J., Taylor, B.P., Tisminetzky, M.S., Petkova, E.,Chen, Y., Ma, G., Klein, D.F., 2002. Are there differences between women's andmen's antidepressant responses? Am. J. Psychiatry 159, 1848–1854.

Rasgon, N., Bauer, M., Grof, P., Gyulai, L., Elman, S., Glenn, T., Whybrow, P.C., 2005a.Sex-specific self-reported mood changes by patients with bipolar disorder.J. Psychiatr. Res. 39, 77–83.

Rasgon, N.L., Altshuler, L.L., Fairbanks, L., Elman, S., Bitran, J., Labarca, R., Saad, M.,Kupka, R., Nolen, W.A., Frye, M.A., Suppes, T., McElroy, S.L., Keck Jr., P.E.,Leverich, G., Grunze, H., Walden, J., Post, R., Mintz, J., 2005b. Reproductivefunction and risk for PCOS in women treated for bipolar disorder. BipolarDisord. 7, 246–259.

Raskin, A., 1974. Age-sex differences in response to antidepressant drugs. J. Nerv.Ment. Dis. 159, 120–130.

Robinson, D.G., Woerner, M.G., Alvir, J.M., Geisler, S., Koreen, A., Sheitman, B.,Chakos, M., Mayerhoff, D., Bilder, R., Goldman, R., Lieberman, J.A., 1999.Predictors of treatment response from a first episode of schizophrenia orschizoaffective disorder. Am. J. Psychiatry 156, 544–549.

Rubinow, D.R., Moore, M., 2004. Sex-dependent modulation of treatment response.Dialogues Clin. Neurosci. 6, 39–51.

Russo, P., Smith, M.W., Dirani, R., Namjoshi, M., Tohen, M., 2002. Pharmacotherapypatterns in the treatment of bipolar disorder. Bipolar Disord. 4, 366–377.

Sackeim, H.A., Portnoy, S., Neeley, P., Steif, B.L., Decina, P., Malitz, S., 1986. Cognitiveconsequences of low-dosage electroconvulsive therapy. Ann. N. Y. Acad. Sci.462, 326–340.

Salokangas, R.K., 2004. Gender and the use of neuroleptics in schizophrenia.Schizophr. Res. 66, 41–49.

Sasson, Y., Chopra, M., Harrari, E., Amitai, K., Zohar, J., 2003. Bipolar comorbidity:from diagnostic dilemmas to therapeutic challenge. Int. J. Neuropsychophar-macol. 6, 139–144.

Sato, T., Bottlender, R., Kleindienst, N., Moller, H.J., 2002. Syndromes and phenom-enological subtypes underlying acute mania: a factor analytic study of 576manic patients. Am. J. Psychiatry 159, 968–974.

Saunders, E.F., Fitzgerald, K.D., Zhang, P., McInnis, M.G., 2012. Clinical features ofbipolar disorder comorbid with anxiety disorders differ between men andwomen. Depress. Anxiety 29, 739–746.

Saunders, E.F., Nazir, R., Kamali, M., Ryan, K.A., Evans, S., Langenecker, S., Gelenberg,A.J., McInnis, M.G., 2014. Gender differences, clinical correlates, and long-itudinal outcome of bipolar disorder with comorbid migraine. J. Clin. Psychiatry75, 512–519.

Schneck, C.D., Miklowitz, D.J., Miyahara, S., Araga, M., Wisniewski, S., Gyulai, L.,Allen, M.H., Thase, M.E., Sachs, G.S., 2008. The prospective course of rapid-cycling bipolar disorder: findings from the STEP-BD. Am. J. Psychiatry 165,370–377 (quiz 410).

Schweder, L.J., Lydersen, S., Wahlund, B., Bergsholm, P., Linaker, O.M., 2011.Electroconvulsive therapy in Norway: rates of use, clinical characteristics,diagnoses, and attitude. J. ECT 27, 292–295.

Secunda, S.K., Katz, M.M., Swann, A., Koslow, S.H., Maas, J.W., Chuang, S., Croughan,J., 1985. Mania. Diagnosis, state measurement and prediction of treatmentresponse. J. Affect. Disord. 8, 113–121.

Sellgren, C., Landen, M., Lichtenstein, P., Hultman, C.M., Langstrom, N., 2011.Validity of bipolar disorder hospital discharge diagnoses: file review andmultiple register linkage in Sweden. Acta Psychiatr. Scand. 124, 447–453.

Simoni-Wastila, L., 1998. Gender and psychotropic drug use. Med. Care 36, 88–94.Smith, S., 2010. Gender differences in antipsychotic prescribing. Int. Rev. Psychiatry

22, 472–484.Strakowski, S.M., Tohen, M., Stoll, A.L., Faedda, G.L., Goodwin, D.C., 1992. Comor-

bidity in mania at first hospitalization. Am. J. Psychiatry 149, 554–556.Suominen, K., Mantere, O., Valtonen, H., Arvilommi, P., Leppamaki, S., Isometsa, E.,

2009. Gender differences in bipolar disorder type I and II. Acta Psychiatr. Scand.120, 464–473.

Suppes, T., Mintz, J., McElroy, S.L., Altshuler, L.L., Kupka, R.W., Frye, M.A.,Keck Jr., P.E., Nolen, W.A., Leverich, G.S., Grunze, H., Rush, A.J., Post, R.M.,2005. Mixed hypomania in 908 patients with bipolar disorder evaluatedprospectively in the Stanley Foundation Bipolar Treatment Network: a sex-specific phenomenon. Arch. Gen. Psychiatry 62, 1089–1096.

Swann, A.C., Bowden, C.L., Morris, D., Calabrese, J.R., Petty, F., Small, J., Dilsaver, S.C.,Davis, J.M., 1997. Depression during mania. Treatment response to lithium ordivalproex. Arch. Gen. Psychiatry 54, 37–42.

Szymanski, S., Lieberman, J.A., Alvir, J.M., Mayerhoff, D., Loebel, A., Geisler, S.,Chakos, M., Koreen, A., Jody, D., Kane, J., et al., 1995. Gender differences in onsetof illness, treatment response, course, and biologic indexes in first-episodeschizophrenic patients. Am. J. Psychiatry 152, 698–703.

Thase, M.E., Frank, E., Kornstein, S.G., et al., 2000. Sex-related differences inresponse to treatments of depression. In: Frank, E. (Ed.), Gender and Its Effectson Psychopathology. American Psychiatric Press, Washington, DC, pp. 103–129.

The National Board of Health and Welfare, 2000. A quality perspective ofpharmacotherapy in psychiatric inpatient care (“Ett kvalitetsperspektiv påläkemedel i psykiatrisk sluten vård”), Linköping, Sweden.

Tsuyuki, R.T., Teo, K.K., Ikuta, R.M., Bay, K.S., Greenwood, P.V., Montague, T.J., 1994.Mortality risk and patterns of practice in 2,070 patients with acute myocardialinfarction, 1987–92. Relative importance of age, sex, and medical therapy. Chest105, 1687–1692.

Usall, J., Suarez, D., Haro, J.M., Group, S.S., 2007. Gender differences in response toantipsychotic treatment in outpatients with schizophrenia. Psychiatry Res. 153,225–231.

Viguera, A.C., Baldessarini, R.J., Tondo, L., 2001. Response to lithium maintenancetreatment in bipolar disorders: comparison of women and men. Bipolar Disord.3, 245–252.

Viguera, A.C., Tondo, L., Baldessarini, R.J., 2000. Sex differences in response tolithium treatment. Am. J. Psychiatry 157, 1509–1511.

Wohlfarth, T., Storosum, J.G., Elferink, A.J., van Zwieten, B.J., Fouwels, A., van denBrink, W., 2004. Response to tricyclic antidepressants: independent of gender?Am. J. Psychiatry 161, 370–372.

Yonkers, K.A., Wisner, K.L., Stowe, Z., Leibenluft, E., Cohen, L., Miller, L., Manber, R.,Viguera, A., Suppes, T., Altshuler, L., 2004. Management of bipolar disorderduring pregnancy and the postpartum period. Am. J. Psychiatry 161, 608–620.

A. Karanti et al. / Journal of Affective Disorders 174 (2015) 303–309 309