BRIEF REPORT Early clinical predictors of long-term morbidity in major depressive disorder Giulia Serra 1,2,3,4 | Athanasios Koukopoulos 3,4 | Lavinia De Chiara 5 | Alexia E. Koukopoulos 2,4 | Gabriele Sani 4,5 | Leonardo Tondo 1,2,3,6 | Paolo Girardi 4,5 | Daniela Reginaldi 3,4 | Ross J. Baldessarini 2,3 1 Child Psychiatry Unit, Dept. of Neuroscience, I.R.C.C.S. Children Hospital Bambino Gesù, Rome, Italy 2 Department of Psychiatry, Harvard Medical School, Boston, Massachusetts 3 International Consortium for Mood & Psychotic Disorders Research, Mailman Research Center, McLean Hospital, Belmont, Massachusetts 4 Lucio Bini Mood Disorder Center, Rome, Italy 5 NESMOS Department, Sant'Andrea Hospital, Sapienza University of Rome, Rome, Italy 6 Lucio Bini Mood Disorder Center, Cagliari, Sardinia, Italy Correspondence Dr Giulia Serra, Child Psychiatry Unit, Department of Neuroscience, I.R.C.C.S. Children Hospital Bambino Gesù, Piazza S. Onofrio, 4, 00165 Rome, Italy. Email: giuliaserra@gmail.com Aims: To identify early clinical factors predictive of later morbidity in major depressive disor- der (MDD). Methods: We analysed factors associated with long-term depressive morbidity (%-time ill) between a first-lifetime major depressive episode and last follow-up of 116 adults diagnosed with DSM-IV major depressive disorder. Bivariate comparisons were followed by multivariable linear regression modelling. Results: Three factors were independently associated with an average of 25%-time-depressed over 17 years at risk: (a) agitated-mixed, or psychotic features in initial major depressive epi- sodes, (b) anxiety syndromes prior to a first-lifetime major depressive episode, and (c) anxiety symptoms in childhood. Conclusion: Early anxiety symptoms and syndromes and agitated-mixed or psychotic initial depressive episodes predicted more long-term depressive morbidity in MDD. KEYWORDS anxiety, depression, mixed, outcome, prediction 1 | INTRODUCTION Clinical features arising before or at the start of major depressive disor- der (MDD) in adults that predict the severity of future morbidity are not well defined. Mood disorders with juvenile vs adult onset are reported to be generally more severe and more recurrent (van Noorden et al., 2011; Zisook et al., 2004). In particular, juvenile onset of MDD has been associated with more later recurrences/year, longer-lasting episodes, and greater symptom-severity, with more agitation or hypo- manic features, more suicidal risk, more other co-occurring psychiatric disorders, as well as lower educational or vocational achievement and marriage rates (Zisook et al., 2004). Also, early anxiety disorders may increase risk for later MDD (Fichter, Quadflieg, Fischer, & Kohlboeck, 2010; Pine, Cohen, Gurley, Brook, & Ma, 1998) as well as predicting prevalent long-term depressive polarity in bipolar disorder (Baldessarini, Tondo, & Visioli, 2014). Less clear is whether these or other features may predict long-term morbidity in MDD in adult years. Given limited information available about potential predictive value of early clinical features, before and at first episodes for long- term depressive morbidity in MDD, we analysed such relationships in MDD patients who were systematically evaluated, treated clinically, and closely followed over nearly 5 years. 2 | METHODS Based on previously detailed methods (Serra et al., 2015, 2017), we extracted historical and prospective data during repeated clinical assessment and long-term treatment from detailed, semi-structured clinical records of adult outpatients diagnosed with DSM-IV-TR MDD and followed for several years by the same mood disorder expert, the late Athanasios Koukopoulos, M.D., at the Lucio Bini Mood Disorder Center of Rome. We included charts of MDD subjects followed at the study centre for at least 1 year to allow evaluation of long-term mor- bidity, and providing adequate information. Received: 5 December 2017 Revised: 4 June 2018 Accepted: 4 November 2018 DOI: 10.1111/eip.12768 Early Intervention in Psychiatry. 2018;1–4. wileyonlinelibrary.com/journal/eip © 2018 John Wiley & Sons Australia, Ltd 1