Stimulant-induced Psychosis: A Comparative Systematic Review and Meta-analysis of Psychotic Outcomes from Therapeutic and Nontherapeutic Use of Stimulants.
Deepika Jangra, Ritika Tejwani, Yashita Ahluwalia et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis compared psychotic outcomes associated with prescribed and nontherapeutic stimulant use. It included 77 studies involving 687,912 people: 9 studies of therapeutic use and 68 of nontherapeutic use. Pooled psychosis prevalence was 0.2% (95% CI: 0.0-0.3) among therapeutic users and 32.8% (95% CI: 25.5-40.2) among nontherapeutic users; pooled incidence in therapeutic use was 0.6% (95% CI: 0.3-0.9). The authors describe a low but clinically relevant risk with prescribed stimulants and a substantially greater burden with nontherapeutic use.
Key findings
- Pooled psychosis prevalence was 0.2% (95% CI: 0.0-0.3) in therapeutic users versus 32.8% (95% CI: 25.5-40.2) in nontherapeutic users.
- The pooled incidence of stimulant-induced psychosis in therapeutic use was 0.6% (95% CI: 0.3-0.9).
- Nontherapeutic psychosis was associated with higher dosages, acute onset, persecutory delusions, hallucinations, and greater recurrence risk.
- Therapeutic cases often resolved after dose reduction or discontinuation; evidence on treatment effectiveness, neurobiology, and genetic factors was limited and heterogeneous.
Why it matters
The markedly different prevalence estimates across use contexts highlight the importance of distinguishing prescribed use from nonmedical exposure when interpreting stimulant-associated psychosis. The review also identifies psychosis as a clinically relevant, though uncommon, outcome in therapeutic use and describes a substantial burden in nontherapeutic settings. These findings support attention to monitoring and risk profiling while recognizing that evidence on several clinical and biological questions remains limited.
What to keep in mind
The review found limited and heterogeneous evidence on treatment effectiveness, neurobiology, and genetic factors. The number of included studies differed substantially by context: 9 therapeutic and 68 nontherapeutic.