Deviation from expected cognitive ability across psychotic disorders

Abstract

Patients with schizophrenia show a deficit in cognitive ability compared to estimated premorbid and familial intellectual abilities. However, the degree to which this pattern holds across psychotic disorders and is familial is unclear. The present study examined deviation from expected cognitive level in schizophrenia, schizoaffective disorder, and psychotic bipolar disorder probands and their first-degree relatives. Using a norm-based regression approach, parental education and WRAT-IV Reading scores (both significant predictors of cognitive level in the healthy control group) were used to predict global neuropsychological function as measured by the composite score from the Brief Assessment of Cognition in Schizophrenia (BACS) test in probands and relatives. When compared to healthy control group, psychotic probands showed a significant gap between observed and predicted BACS composite scores and a greater likelihood of robust cognitive decline. This effect was not seen in unaffected relatives. While BACS and WRAT-IV Reading scores were themselves highly familial, the decline in cognitive function from expectation had lower estimates of familiality. Thus, illness-related factors such as epigenetic, treatment, or pathophysiological factors may be important causes of illness related decline in cognitive abilities across psychotic disorders. This is consistent with the markedly greater level of cognitive impairment seen in affected individuals compared to their unaffected family members.

Keywords: Brief assessment of cognition in schizophrenia; Cognitive decline; First-degree relatives; Premorbid cognition; Psychotic disorders.

Figure 1

All proband groups show a lower deviation scores compared to both the relative groups and controls. Mean deviation from expectation scores differed significantly among proband groups from bipolar to schizoaffective to schizophrenia. This pattern suggests disruption of premorbid cognitive trajectories across psychotic disorders. ^†Deviation Score = Difference between Observed and Predicted BACS Composite z-score.

Figure 2

Scatterplot of observed BACS composite score and predicted BACS based on WRAT and parent education. The solid line represents the lower end of the 95% confidence interval derived from the healthy control regression equation.

Figure 2

Scatterplot of observed BACS composite score and predicted BACS based on WRAT and parent education. The solid line represents the lower end of the 95% confidence interval derived from the healthy control regression equation.

Figure 2

Scatterplot of observed BACS composite score and predicted BACS based on WRAT and parent education. The solid line represents the lower end of the 95% confidence interval derived from the healthy control regression equation.

Figure 3

Rates of cognitive impairment based on cut-scores for level of performance (left figure) and deviation from cognitive expectation (right figure). ^†Deviation Score = Difference between Observed and Predicted BACS Composite z-score.

Figure 4

Cases were classified based on significant deviation from expectation (below the 95% CI cutoff) in probands. The magnitude of deviation between probands and relatives increased dramatically, regardless of diagnosis, as a function the proband deviation status. Thus, the cognitive decline seen in probands below the cut score does not stem from a familial pattern of cognitive vulnerability. ^†Deviation Score = Difference between Observed and Predicted BACS Composite z-score.