Disability multilevel modelling in first episodes of psychosis at 3-year follow-up Blanca Mellor-Marsá a,b,c , Francisco Félix Caballero a,b,c , María Cabello a,b,c , José Luis Ayuso-Mateos a,b,c , Esther Setién-Suero b,d , Javier Vázquez-Bourgon b,d , Benedicto Crespo-Facorro b,d , Rosa Ayesa-Arriola b,d, ⁎ a Department of Psychiatry, Universidad Autónoma de Madrid, Spain b Instituto de Salud Carlos III, Centro de Investigación Biomédica en Red de Salud Mental, CIBERSAM, Spain c Instituto de Investigación de la Princesa (IIS-IP), Hospital Universitario de la Princesa, Madrid, Spain d Department of Psychiatry, Marqués de Valdecilla University Hospital, IDIVAL, School of Medicine, University of Cantabria, Santander, Spain abstract article info Article history: Received 10 September 2016 Received in revised form 12 December 2016 Accepted 14 December 2016 Available online xxxx The description of longitudinal bio-psycho-social profiles in FEP samples may be useful for the prediction of dis- ability trajectories. This study aimed to describe the differences between disability status of FEP patients at base- line and their change over time, analysing how variables associated to the psychological status, and the environment of the patient can affect his or her disability trajectory, once the influence of health condition and socio-demographic variables has been controlled for. Using data from a 3-year follow-up study on early psychosis (PAFIP), a multilevel structure in which the longitudinal measurements (within level) were nested within the in- dividuals (between level), was modeled. The contribution of the different time-varying and time-invariant var- iables to the patients' disability outcomes was tested through eight nested models. Consecutive models, that successively added health related, socio-demographic, psychological and environmental variables to the uncon- ditional model were estimated, by means of deviance and fit statistics. The present work revealed the importance of psychological and environmental factors in the explanation of dis- ability changes in the context of FEP. We may conclude that longitudinal assessments of time-varying predictors – living situation (b = -0.10, p b 0.05), economic support (b = 0.11, p b 0.01) and insight (b = -0.08, p b 0.05) – explain a relevant amount of disability variation over time, independently from symptoms' severity, duration of untreated psychosis, age, gender and years of education. Additionally, the level of premorbid adjustment (b = 0.05, p b 0.001) was asso- ciated to differences in disability outcomes among FEP patients. © 2016 Elsevier B.V. All rights reserved. Keywords: First-episode psychosis Disability Multilevel analysis Longitudinal study 1. Introduction Psychosocial difficulties frequently constitute a disabling disadvan- tage that adds up to the intrinsic burden of schizophrenia spectrum dis- order symptoms (Switaj et al., 2012). The dimension of a patient's perceived disability should be considered when analysing functional restoration (Harvey and Bellack, 2009). Wiersma et al. (2000) found that the remission of symptoms during the first two years in a sample of patients with first episode of schizophrenia from six European co- horts did not play a very important role in explaining the variation of disability scores. Notwithstanding, few studies have considered disabil- ity as a relevant outcome variable using longitudinal designs in the con- text of FEP (Cornblatt et al., 2012; Hodgekins et al., 2015). Changes over time in relevant factors such as psychosocial abilities (DeSisto et al., 1995), stressful life events, social and family support net- works, patient reactions to illness, and self-attitudes (Wing, 1988) and antipsychotic medication (Robinson et al., 1999), may have an impact on the disability evolution of first episode of psychosis (FEP) patients. In fact, past research addressing risk factors for disability has found that variables measured at baseline do not contribute significantly to variance in the functioning outcome (Gonzalez-Blanch et al., 2015). As Ayesa-Arriola et al. (2013) have hypothesized, the low predictive capacity of baseline cognitive and health-related factors on follow-up functioning outcomes might be explained by chronicity and other disor- der traits. Baseline measures of gender (Jablensky et al., 1992), cognitive functions (Allott et al., 2011), negative symptoms (Milev et al., 2005), premorbid adjustment (Jeppesen et al., 2008) or years of education (Ayesa-Arriola et al., 2013) have shown an association with FEP pa- tients' disability status, yet with a high percentage of disability variance remaining unexplained in the long term. Schizophrenia Research xxx (2016) xxx–xxx ⁎ Corresponding author at: Department of Psychiatry, Marqués de Valdecilla University Hospital, IDIVAL, School of Medicine, University of Cantabria, Santander, Spain. E-mail address: rayesa@humv.es (R. Ayesa-Arriola). SCHRES-07090; No of Pages 6 http://dx.doi.org/10.1016/j.schres.2016.12.014 0920-9964/© 2016 Elsevier B.V. All rights reserved. Contents lists available at ScienceDirect Schizophrenia Research journal homepage: www.elsevier.com/locate/schres Please cite this article as: Mellor-Marsá, B., et al., Disability multilevel modelling in first episodes of psychosis at 3-year follow-up, Schizophr. Res. (2016), http://dx.doi.org/10.1016/j.schres.2016.12.014