Effects of risperidone/paliperidone versus placebo on cognitive functioning over the first 6 months of treatment for psychotic disorder: secondary analysis of a triple-blind randomised clinical trial

Abstract

The drivers of cognitive change following first-episode psychosis remain poorly understood. Evidence regarding the role of antipsychotic medication is primarily based on naturalistic studies or clinical trials without a placebo arm, making it difficult to disentangle illness from medication effects. A secondary analysis of a randomised, triple-blind, placebo-controlled trial, where antipsychotic-naive patients with first-episode psychotic disorder were allocated to receive risperidone/paliperidone or matched placebo plus intensive psychosocial therapy for 6 months was conducted. A healthy control group was also recruited. A cognitive battery was administered at baseline and 6 months. Intention-to-treat analysis involved 76 patients (antipsychotic medication group: 37; 18.6_Mage [2.9] years; 21 women; placebo group: 39; 18.3_Mage [2.7]; 22 women); and 42 healthy controls (19.2_Mage [3.0] years; 28 women). Cognitive performance predominantly remained stable (working memory, verbal fluency) or improved (attention, processing speed, cognitive control), with no group-by-time interaction evident. However, a significant group-by-time interaction was observed for immediate recall (p = 0.023), verbal learning (p = 0.024) and delayed recall (p = 0.005). The medication group declined whereas the placebo group improved on each measure (immediate recall: p = 0.024; η_p² = 0.062; verbal learning: p = 0.015; η_p² = 0.072 both medium effects; delayed recall: p = 0.001; η_p² = 0.123 large effect). The rate of change for the placebo and healthy control groups was similar. Per protocol analysis (placebo n = 16, medication n = 11) produced similar findings. Risperidone/paliperidone may worsen verbal learning and memory in the early months of psychosis treatment. Replication of this finding and examination of various antipsychotic agents are needed in confirmatory trials. Antipsychotic effects should be considered in longitudinal studies of cognition in psychosis.Trial registration: Australian New Zealand Clinical Trials Registry ( http://www.anzctr.org.au/ ; ACTRN12607000608460).

Fig. 1

A diagram of participant flow into the study.

Fig. 2. Plots of estimated trends for cognitive measures with no significant group × time interaction from baseline to 6 months and associated 95% confidence intervals for the means at endpoints.

Each panel represents a different cognitive test score, with the test score shown on the y-axis and the timepoint shown on the x-axis.

Fig. 3. Plots of estimated trends for significant group × time interaction for verbal paired associate learning and memory from baseline to 6 months and associated 95% confidence intervals for the means at the endpoints.

Each panel represents a different cognitive test score (a immediate recall; b verbal learning; and c delayed recall), with the test score shown on the y-axis and the timepoint shown on the x-axis.