Method
Subjects
A total of 198 patients with a first episode psychosis (FEP) were included at baseline from the ongoing Thematically Organized Psychosis Research (TOP) study at the Oslo University Hospital and University of Oslo, Norway. Of these, a total of 127 patients were followed for one year. Romm and colleagues have previously reported the baseline characteristics of 119 of the 198 patients included in the present study.[4] The present study is a further exploration of depressive symptoms in a 12 months follow-up period.
These samples were included from March 2004 to February 2010. The inclusion criteria were (1) age 18 to 65 years and a first episode non-affective psychosis according to Diagnostic and Structural Manual of Mental Disorders, fourth edition (DSM-IV).[18] Exclusion criteria were a history of organic brain disorder, a significant comorbid medical condition, or an IQ below 70.
Inclusion was based on the time from the first onset of positive psychotic symptoms (the first week with a PANSS score of 4 or more on Positive Scale Items 1, 3, 5, 6 or General Scale item 9) to the start of first admission to the study.[19] Patients with previous treatment with antipsychotic medication in adequate dosage for more than 12 weeks were not considered as FEP patients.
The study is approved by the Regional Committee for Medical Research Ethics and the Norwegian Data Inspectorate. A written informed consent was obtained from all of the participants.
Clinical Assessment and Instruments
Diagnosis was set by the Structural Clinical Interview for DSM-IV axis I disorders (SCID I).[18] Current symptom severity was assessed on the Positive and Negative Syndrome Scale (PANSS)[20] and characterized by five subscales; positive, negative, excitative, depressive, and cognitive.[21] Depression was assessed using the Calgary Depression Scale for Schizophrenia (CDSS).[22] CDSS has been shown to be reliable and valid and it measures depression separate from negative or extrapyramidal symptoms.[23,24] In a systematic review of instruments measuring depressive symptoms, the CDSS outperformed other depression instruments in terms of reliability and validity in patients with schizophrenia.[25] The cut-off score for depression varies between three and nine in the different studies.[15,17,26–28] We chose to use a cut-off score of equal and/or above six, in line with some of the previous studies.[28–30] Suicidality was measured with the CDSS-suicide item. Premorbid functioning was assessed with the Premorbid Adjustment Scale (PAS).[31] The premorbid phase is defined as the time from birth until six months before onset of psychosis. PAS measures social and academic functioning in 4 time periods; childhood, early adolescence, late adolescence, and adult life. We decided to focus on social and academic functioning in childhood and early adolescence only. It may be difficult to discern between premorbid and psychotic symptoms in late adolescence when the first psychotic episode usually occur. General level of symptoms and functioning was assessed with the Global Assessment of Functioning scale (GAF), split version.[32,33] Problem drinking was screened by the Alcohol Use Disorders Identification Test (AUDIT),[34] and use of illegal drugs was assessed with Drug Use Disorders Identification Test (DUDIT).[35] DUP was measured following the criteria described by Larsen et al..[36] At 12-month follow-up the participants were re-assessed with the following instruments: GAF, PANSS, CDSS, AUDIT and DUDIT.
Reliability
Clinical interviews were carried out by trained psychologists and psychiatrists. The interviewers were all enrolled in the general training and reliability program in the TOP study. For DSM-IV diagnostics, mean overall kappa with training videos was 0.77 (95% confidence interval [CI]: 0.60–0.94). Interrater reliability, measured by the interclass correlation coefficient (1.1), was for the PANSS positive subscale 0.82 (95% CI: 0.66–0.94), for the PANSS negative subscale 0.76 (95% CI: 0.58–093), the PANSS general subscale 0.73 (95% CI: 0.54–0.90), the GAF symptom scale 0.86 (95% CI: 0.77–0.92), and the GAF functioning scale 0.85 (95% CI: 0.76–0.92).
Analysis
Statistical analysis was done by using SPSS for Windows (version 16.0). Descriptive statistics was performed on the whole sample to obtain the independent variables' frequencies, means and standard deviations. To identify different subgroups of patients with different trajectories of depressive symptoms, we used a cut-off score above or equal six on the CDSS. Based on the CDSS scores we wanted to identify patients who were depressed in the whole follow-up period, patients who were depressed at baseline, but recovered during the follow-up period, patients who became depressed during the follow-up period, and finally patients who never were scored as depressed during the follow-up period. One-Way ANOVA was performed in order to compare the means of the follow-up subgroups. Bivariate correlations were used to study the correlation of each independent variable with CDSS total scores at both baseline and 12-month follow-up.
DUP required transformation to its natural logarithm (Ln[DUP + 1]).
Hierarchical multiple linear regression was used to assess the ability of baseline variables to predict depression (CDSS total score) at 12-month follow-up. The predictors were selected based on theoretical considerations: gender differences, neurocognitive developmental issues (premorbid functioning in childhood), and basic psychopathological/psychotic symptoms that may have association with the development of depressive symptoms in FEP. To estimate how much of the variance each of the patient variables explained, we used a blockwise multiple hierarchic regression analysis. Gender was entered in the first block, because this variable must be considered as the most basic one. PAS childhood social and academic were entered in the second block as a measure of early social and academic functioning. The duration of psychosis was entered in the third block. Finally, we entered the five different PANSS subscales which we considered as the most important variables in predicting depressive symptoms at 12 months follow-up. By entering the PANSS subscale scores in the fourth block, we controlled for the amount of variance explained by the variables in the three first blocks.
BMC Psychiatry. 2013;13(106) © 2013 BioMed Central, Ltd.
© 1999-2006 BioMed Central Ltd
