Characteristics of interpretation bias and relationship with suicidality in a psychiatric hospital sample

Background: Interpretation bias, the tendency to interpret ambiguous situations negatively (or to lack a positive bias), is a cognitive vulnerability associated with psychopathology. However, there is a lack of research characterizing this bias in psychiatric samples, including whether it is also a risk factor for suicidality. This study characterized interpretation bias in a psychiatric sample at risk for suicide and examined the relationship between interpretation bias and suicidality cross-sectionally and prospectively. Methods: Patients (N=65) attending a partial hospital program completed the Word-Sentence Association Paradigm (WSAP), which results in four variables reflecting different types of interpretation bias: endorsement rates and reaction time bias scores for negative and benign interpretations. We conducted logistic regression models to predict high suicidality (ideation, plans, attempts, etc. assessed via a structured interview at admission) and suicidal ideation (assessed via self-report at admission and discharge). Results: Logistic regression models predicting suicide outcomes upon admission and discharge indicated that benign interpretation endorsement was the most robust predictor of suicidality concurrently and prospectively, controlling for baseline suicidal ideation. Limitations: Lack of gold standard self-report suicide assessment. Unable to assess additional constructs such as hopelessness or perfectionism, which may better elucidate how lacking a benign bias influences suicidality. Modest sample size. Conclusions: A lower endorsement of positive interpretations was the strongest predictor of prospective suicidal ideation, even after controlling for baseline suicidal ideation. Future research should examine how targeting interpretation bias influences suicidality.