Relapse predictors in frist-episode schizophrenia: a 3-year longitudinal study

Abstract

The relapsing nature of schizophrenia poses great challenge in terms of treatment and recovery. Identification of relapse predictors would be of utmost importance and relevance in the management of schizophrenia. In this naturalistic study, potential relapse predictors in terms of demographics, clinical symptomatology and cognition were investigated. 153 consecutive patients with first-episode psychosis were recruited, of which 138 fulfilled DSM-IV criteria for schizophrenia, schizophreniform disorder or schizoaffective disorder upon review at 3 years. Of the 138 eligible subjects, 93 completed the study at 3 years. They were 42 men and 51 women with a mean age of 31.2 (S.D.= 9.6), mean duration of untreated psychosis of 474 days (S.D.= 768), and mean educational attainment of 10.54 years (S.D.=2.9). 55 patients (605) experienced no relapse at 3 years, 25 (27%) had one relapse and 12 (13%) had 2 or more relapses. Those who relapsed have significantly poorer medication adherence (year 1: t=-3.4, p=0.002; year 2: t=-3.6, p=0.001; year 3: t=-3.4, p=0.002) and lower age at presentation (t=3.27, p=0.002). They also tended to have more life events prior to onset (t=-1.687, p=0.097) and more years of education (t=-1.99, p=0.049). No significant effect was found with regard to pre-morbid personality and gender. Our findings suggest that medication adherence and younger age at presentation are significant predictors of the number of relapses in first-episode schizophrenia patients in the first three years. Efforts should target at enhancing medication adherence and engagement of patients, especially those of younger age.