Schizophrenia research based on traditional assessment measures for negative symptoms appears to be, to some extent, unreliable. The limitations of the Positive and Negative Syndrome Scale (PANSS) and the Scale for the Assessment of Negative Symptoms (SANS) have been extensively acknowledged and should be taken into account. The aim of this study is to show how the PANSS and the SANS conflate negative symptoms and cognition and to offer alternatives for the limitations found.

A sample of 117 participants with schizophrenia from two independent studies was retrospectively investigated. Linear regression models were computed to explore the effect of negative symptoms and illness duration as predictors of cognitive performance.

For the PANSS, the item “abstract thinking” accounted for the association between negative symptoms and cognition. For the SANS, the “attention” subscale predicted the performance in verbal memory, but illness duration emerged as a stronger predictor than negative symptoms for outcomes of processing speed, verbal and working memory.

Utilizing alternative models to the traditional PANSS and SANS formats, and accounting for illness duration, provide more precise evidence on the relationship between negative symptoms and cognition. Since these measures are still extensively utilized, we recommend adopting more rigorous approaches to avoid misleading results.

Patients with major depression show reduced hippocampal volume compared to healthy controls. However, the contribution of patients’ cumulative illness severity to hippocampal volume has rarely been investigated. It was the aim of our study to find a composite score of cumulative illness severity that is associated with hippocampal volume in depression.

We estimated hippocampal gray matter volume using 3-tesla brain magnetic resonance imaging in 213 inpatients with acute major depression according to DSM-IV criteria (employing the SCID interview) and 213 healthy controls. Patients’ cumulative illness severity was ascertained by six clinical variables via structured clinical interviews. A principal component analysis was conducted to identify components reflecting cumulative illness severity. Regression analyses and a voxel-based morphometry approach were used to investigate the influence of patients’ individual component scores on hippocampal volume.

Principal component analysis yielded two main components of cumulative illness severity: Hospitalization and Duration of Illness. While the component Hospitalization incorporated information from the intensity of inpatient treatment, the component Duration of Illness was based on the duration and frequency of illness episodes. We could demonstrate a significant inverse association of patients’ Hospitalization component scores with bilateral hippocampal gray matter volume. This relationship was not found for Duration of Illness component scores.

Variables associated with patients’ history of psychiatric hospitalization seem to be accurate predictors of hippocampal volume in major depression and reliable estimators of patients’ cumulative illness severity. Future studies should pay attention to these measures when investigating hippocampal volume changes in major depression.