Prescription psychostimulants, atomoxetine and the risk of psychosis in adults with history of psychosis: a population-based cohort study
Concerns exist that psychostimulants may increase the risk of psychosis, particularly in individuals with a history of psychosis. This study assessed whether psychosis risk increases after initiating psychostimulant or atomoxetine treatment in individuals with prior psychotic events. In this cohort...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article (Journal) |
| Language: | English |
| Published: |
31 March 2026
|
| In: |
Translational Psychiatry
Year: 2026, Volume: 16, Pages: 1-9 |
| ISSN: | 2158-3188 |
| DOI: | 10.1038/s41398-026-03998-4 |
| Online Access: | Verlag, kostenfrei, Volltext: https://doi.org/10.1038/s41398-026-03998-4 Verlag, kostenfrei, Volltext: https://www.nature.com/articles/s41398-026-03998-4 
|
| Author Notes: | Patrick Bach, Johan Franck, Jonas Hällgren, Härje Widing, Mika Gissler and Jeanette Westman |
| Summary: | Concerns exist that psychostimulants may increase the risk of psychosis, particularly in individuals with a history of psychosis. This study assessed whether psychosis risk increases after initiating psychostimulant or atomoxetine treatment in individuals with prior psychotic events. In this cohort study, we used Swedish population-based register data that contain data on all Swedish residents. We identified adults aged 18-64, who had a history of psychosis and initiated psychostimulant or atomoxetine treatment between 2008 and 2021. A within-individual design compared the rate of psychotic events in two 6-month periods before and after treatment initiation. Rate ratios (RR) with 95% confidence intervals (CI) were estimated. Sensitivity analyses were stratified by sex, age groups, and medication classes and also investigated substance-related psychosis as alternative outcome. We identified 3,770 individuals with a history of psychosis that received psychostimulant or atomoxetine prescriptions during the study period. Compared to the period before treatment initiation, the RR for psychotic events in the six months following treatment initiation was 0.95 (95% CI 0.84-1.08). Sensitivity analyses, considering only substance-induced psychotic events, or eight-week intervals and subgroup analyses by sex, age groups and separately for individuals receiving either (dex-)methyphenidate or (lis-)dexamphetamine or atomoxetine or (ar-)modafinil showed consistent results. Contrary to concerns, psychostimulant or atomoxetine treatment was not associated with an increased risk of psychotic events in adults with a history of psychosis. These findings may inform clinicians considering psychostimulant prescriptions in this population. |
|---|---|
| Item Description: | Gesehen am 23.04.2026 |
| Physical Description: | Online Resource |
| ISSN: | 2158-3188 |
| DOI: | 10.1038/s41398-026-03998-4 |