Cognitive changes following antidepressant or antipsychotic treatment in adolescents at clinical risk for psychosis

Abstract

Background: Improving neurocognitive abilities is a treatment priority in schizophrenia, however, pharmacological efforts to enhance deficits after illness onset have resulted in quite modest results that are of questionable clinical meaningfulness. Individuals at clinical risk for psychosis demonstrate neurocognitive impairments intermediate to the level of deficits observed in schizophrenia and normative performance, suggesting that a similar magnitude of improvement might result in more clinically meaningful change. In this study, we examined neurocognitive changes after six months of treatment in adolescents with clinical signs of risk for psychosis.

Methods: Adolescents who were referred to the Recognition and Prevention program, which is focused on treatment and research for individuals at a clinical high risk for psychosis, were followed in a naturalistic treatment design. At study entry and approximately six months after starting treatment, we examined neuropsychological functioning and clinical symptoms for patients who remained off medications (OFF; N=27), started selective serotonin reuptake inhibitor antidepressant medication (AD; N=15), or started a second-generation antipsychotic medication (AP; N=11) within three months of study entry. We also included a locally recruited healthy comparison group (HC; N=17).

Results: The clinical groups were not significantly different on baseline demographic, neurocognitive, or clinical symptom measures. Linear mixed models were used to examine cognitive changes, with time between assessments, depressive symptom severity, and attenuated positive symptom severity as random effects. Group by time effects were observed in sustained attention and verbal learning, with the AD group showing a more favorable response than the AP group. The AD group's improvements were not significantly different from the HC or OFF group.

Conclusion: Early intervention for those at clinical high risk for psychosis may result in neurocognitive improvements. These improvements were observed for those prescribed antidepressant, but not antipsychotic medications even though the groups did not differ in clinical symptom severity or treatment response.

Fig. 1

Six-month change in Continuous Performance Test — Identical Pairs, 4-digit condition by group (estimated marginal means, standard error bars). HC = healthy comparison group (N=17), Off/Off = clinical high risk for psychosis subjects off of medication at both time points (N=27), Off/AD = clinical high risk for psychosis subjects off medication at baseline and on an antidepressant at follow-up (N=15), Off/AP = clinical high risk for psychosis subjects off medication at baseline and on a second generation antipsychotic at follow-up (N=11).

Fig. 2

Six-month change in CVLT total learning by group (estimated marginal means, standard error bars). HC = healthy comparison group (N=17), Off/Off = clinical high risk for psychosis subjects off of medication at both time points (N=27), Off/AD = clinical high risk for psychosis subjects off medication at baseline and on an antidepressant at follow-up (N=15), Off/AP = clinical high risk for psychosis subjects off medication at baseline and on a second generation antipsychotic at follow-up (N=11).