cohorts sampled from tertiary centers, emphasizeschronicity and frequent recurrences. We investigatedthe long-term outcome of a regionally representativepsychiatric MDD cohort comprising mainly outpatients.Method: The Vantaa Depression Study included 163patients with DSM-IV MDD (71.8% of those eligible)diagnosed using structured and semistructured interviewsand followed up at 6months, 18months, and 5 yearswith alife-chart. The effects of comorbid disorders and otherpredictors on outcomewere comprehensively investigated.Results: Over the 5-year follow-up, 98.8% of patientsachieved a symptom state below major depressive epi-sode (MDE) criteria and 88.4% reached full remission,with themedian time to full remission being 11.0months.Nearly one-third (29.3%) had no recurrences, whereas30.0% experienced one, 12.9% two, and 27.9% three ormore recurrences. Double depression, cluster C person-ality disorder, and longer MDE duration prior to entrywere the most significant predictors of longer time inachieving full remission. Severity of MDD and comor-bidity, especially social phobia, predicted probability of,shorter time to, and number of recurrences.Conclusion: Previous literature on mostly inpatientMDD may have, by generalizing from patients with themost severe psychopathology, overemphasized chronicityof MDD. The long-term outcome of MDD in psychiatriccare is variable, with about one-tenth of patients havingpoor, one-third intermediate, and a half favorable out-comes. In addition to known predictors, cluster C person-ality disorders and social phobia warrant further attentionas predictors of MDD outcome among outpatients.

Keywords: Major depressive disorder, Outcome,Recurrence, Comorbidity

doi:10.1016/j.jad.2007.12.015

[O1.7] Asenapine in the treatment of acute mania inbipolar I disorder: Outcomes from two randomizedand placebo-controlled trials

R. McIntyre*,a, R. Hirschfeldb, L. Alphsc,M. Cohend, T. Macekc, J. Panagidese

aUniversity of Toronto, CanadabUniversity of Texas Medical Branch, USAcPfizer Global at the time of employment, USAdOrganon Biosciences, USAeOrganon International Inc, USA

Introduction: Asenapine is a novel psychopharmacologicagent in development for the treatment of schizophrenia

and bipolar disorder. We assessed the use of asenapine inpatients with bipolar I disorder in a matched pair of pla-cebo-controlled trials (Ares 7501004 and Ares 7501005).Method: Adults experiencing manic or mixed episodesof bipolar I disorder, with Young Mania Rating Scale(YMRS) total score≥20, were randomly assigned (2:2:1)to 3weeks of flexible-dose asenapine (10mg adjustable to5 mg BID), olanzapine (15 mg adjustable to 5, 10, or20 mg QD), or placebo following single-blind run-inperiods of up to 7 days. A mixed model for repeatedmeasures analysis was performed for the primary efficacyoutcome, the least squares (LS) mean change in YMRStotal score from baseline to day 21 for asenapine versusplacebo (there were no statistical comparisons ofasenapine vs olanzapine).Results: For both studies combined, 976 patients took≥1dose of study medication (asenapine, n=185 and 194 inAres 7501004 and 7501005, respectively; olanzapine,n=205 and 190; placebo, n=98 and 104). Mean dosageswere similar in the 2 studies (asenapine, 18.4 and 18.2 mg/d; olanzapine, 15.9 and 15.8 mg/d). LS mean changes inYMRS total scorewere significantly (Pb0.05) greater withasenapine (−14.2 and −13.1) and olanzapine (−16.1 and−13.9) than with placebo (−10.8 and −7.4). The superi-ority of asenapine over placebo was observed as early astreatment day 2. Asenapine was generally well tolerated.Incidence rates of clinically significant weight gain (≥7%)in the 2 studies were 7% and 6% for asenapine, 19% and13% for olanzapine, and 1.2% and 0% for placebo.Discussion: In patients with acute bipolar I disorder,asenapine was effective and well tolerated, with arelatively low risk of incurring weight gain.

Keywords: Bipolar I disorder, Mania, Asenapine

doi:10.1016/j.jad.2007.12.016

[O1.8] Mood stabiliser loading versus titration inacutemania:An audit of outcomes in clinical practice

A.J. Wheeler, P.G. Robinson*, A. Fraser

University of Auckland, New ZealandWaitemata District Health Board, New ZealandAuckland Mind, New Zealand

Introduction: There is evidence that serum level of moodstabiliser, and time taken to achieve that level are factorsin the degree and rapidity of response in acute mania. It islikely that higher serum levels are more effective thanlevels in the middle of the therapeutic ranges. The aim ofthis study was to investigate whether mood stabilisers

S56 Abstracts / Journal of Affective Disorders 107 (2008) S53–S122

administered as a loading compared to a titration regimenimproved clinical outcomes (total length of stay, time inintensive care and seclusion) in acutely manic inpatients.Method: 12-month retrospective review of admissions totwo adult psychiatric units in Auckland, New Zealand, wasconducted. Demographic, legal status, psychiatric admis-sions, clinical outcome (length of stay, intensive care andseclusion use) and medication data were collected for allpatients with a diagnosis of acute mania who started moodstabiliser treatment within three days of admission (n=103). Serum levels and adverse events were also recorded.Results: Data from 97 admissions were analysed (58.8%loading). Lithium was the treatment choice for two-thirdsof the admissions. Average length of stay was 30.6 dayswith no difference between loaded or titrated regimens.72.2% of patients were treated in intensive care (average8.4 days) and 34.0% spent time in seclusion; no differ-ence was found between treatment regimens. Serumlevels were taken inconsistently, particularly for val-proate. More adverse events occurred with loading (51%)compared to titrating (30%).Conclusion: No difference was found between loadingand titrating in this naturalistic study. Dosage does notpredict a greater likelihood of a positive response andthe literature supports a stronger link between rapidlyattained high serum levels and positive outcomes.Access to regular serum levels can lead to better clinicaldecision making and speed up response times. If clini-cians use loading doses they must regularly measureserum levels and monitor for adverse events.

Keywords: Acute mania, Mood stabiliser, Loading,Titration

doi:10.1016/j.jad.2007.12.017

Free rapid communicationsStream two

[O2.1] Cognitive vulnerability to depression inchildren and adolescents: The role of self-criticaland personal standards perfectionism

C.M. Parkinson*, C.M. McWhinnie, J.R.Z. Abela

McGill University, Canada

Perfectionism has traditionally been operationalized asa unidimensional construct comprised only of negativecognitive components (Ellis, 1962). More recent theories,however, operationalize perfectionism as a multidimen-sional construct containing positive and negative compo-

nents (Hamachek, 1978). Consistent with such anoperationalization, factor analytic studies indicate per-fectionism measures break into two distinct factors:(1) maladaptive ‘self-critical’ or ‘evaluative concerns’perfectionism (SCP) and (2) positive ‘personal standards’or ‘achievement striving’ perfectionism (PSP) (Frostet al., 1993; McWhinnie et al., 2007).

Most studies conducted to date have (1) examinedthe impact of perfectionism on depression within amain-effect as opposed to a vulnerability-stress frame-work and (2) utilized adult samples. Therefore, little isknown about the conditions under which perfectionismleads to depression — particularly in youth.

We examined whether SCP and PSP confer vulner-ability to depression in a sample of children andadolescents — particularly following negative events. Inorder to provide a powerful examination of hypotheses,we utilized a one-year multi-wave longitudinal designin which children (n=140; ages 6–14) completed(1) measures assessing SCP, PSP, and depressive symp-toms during an initial assessment, and (2) measuresassessing depressive symptoms and negative events onceevery 6 weeks for the subsequent 12 months. In order tomaximize the number of childrenwho exhibited clinicallysignificant increases in depressive symptoms during thecourse of the study, we utilized a sample of children ofparents with a history of major depressive episodes.

Hierarchical linear modelling analyses indicated thathigher levels of SCP, but not PSP, were significantlyassociatedwith greater within-subject increases in depres-sive symptoms following within-subject increases instress levels. The strength of this association was notmoderated by age suggesting results applied to both thechildren and adolescents in our sample. Implications offindings for the prevention and treatment of youthdepression will be discussed.

Keywords: Depression, Cognitive vulnerability, Youth,Perfectionism

doi:10.1016/j.jad.2007.12.018

[O2.2] The impact of multiple comorbid anxietydisorders on the course and severity of patients witha bipolar disorder

V.Q.Koppelaar,A. Spijker, E.Hoencamp*, J.Haffmans

Parnassia Bavo Group, The Netherlands

Introduction: To investigate the impact of multiple anx-iety disorders on the course severity of bipolar disorder,

S57Abstracts / Journal of Affective Disorders 107 (2008) S53–S122