State and trait hopelessness in a prospective five-year study of patients with depressive disorders

Abstract

BackgroundHopelessness is a common experience of patients with depressive disorders (DD) and an important predictor of suicidal behaviour. However, stability and factors explaining state and trait variation of hopelessness in patients with DD over time are poorly known. MethodsPatients with DD (n = 406) from the Vantaa Depression Study and the Vantaa Primary Care Depression Study filled in the Beck Hopelessness Scale (BHS), Beck Depression Inventory (BDI), Beck Anxiety Inventory (BAI), Perceived Social Support Scale–Revised (PSSS-R), and Eysenck Personality Inventory–Q (EPI-Q) at baseline, at six and eighteen months, and at five years. We conducted a multilevel linear regression analyses predicting BHS with these covariates. ResultsDuring the five-year follow-up half of the variance in BHS was attributable to between-patient variance (50.6%, CI = 41.2–61.5%), and the rest arose from within-patient variance and measurement errors. BDI and BAI explained 5.6% of within-patient and 28.4% of between-patient variance of BHS. High Neuroticism and low Extraversion explained 2.6% of the between-patient variance of BHS. PSSS-R explained 5% of between-patient variance and 1.7% of within-patient variance of BHS. LimitationsNo treatment effects were controlled. ConclusionsHopelessness varies markedly over time both within and between patients with depression; it is both state- and trait-related. Concurrent depressive and anxiety symptoms and low social support explain both state and trait variance, whereas high Neuroticism and low Extraversion explain only trait variance of hopelessness. These variations influence the utility of hopelessness as an indicator of suicide risk.