rethinkTHC Search
Menu
Study breakdown

Cochrane review found emerging but inconclusive evidence for early psychosis interventions, including cannabis-specific therapy

Systematic ReviewModerate evidence
The takeaway

A Cochrane review of 18 RCTs (n=1808) found emerging evidence for specialized early intervention services in psychosis but inconclusive results for most specific treatments, including cannabis-focused psychosis therapy.

Read this if you want to understand what evidence supports early intervention for psychosis.

Cannabis-psychosis specific therapy did not show benefit (n=47)

What the researchers found

This Cochrane systematic review examined whether early intervention could improve outcomes in psychosis across 18 RCTs with 1,808 participants.

For psychosis prevention in prodromal patients: olanzapine showed little benefit, CBT was similarly inconclusive, risperidone plus CBT showed short-term benefit at 6 months but not 12 months, and omega-3 fatty acids showed a promising but unreplicated result.

For first-episode psychosis treatment: specialized early intervention teams showed some benefit for treatment retention and independent living at 5 years. Phase-specific cannabis and psychosis therapy did not show significant benefit (RR for cannabis use 1.30, CI 0.8-2.2). Vocational intervention improved employment.

The largest and highest quality study showed specialized teams reduced dropout and improved treatment compliance, with better independent living at 5 years but not at 1 year.

Why it matters

The finding that cannabis-specific psychosis therapy did not show measurable benefit was important for treatment planning, while the broader finding that early intervention services had emerging support informed mental health policy.

The numbers in context

18 RCTs, n=1808. Cannabis-psychosis therapy: RR 1.30 (0.8-2.2), not significant. Specialized teams: reduced dropout RR 0.59 (0.4-0.8), NNT 9. Independent living at 5 years: RR 0.42 (0.21-0.8), NNT 19.

How the study worked

Cochrane systematic review of RCTs for preventing psychosis in prodromal individuals or improving outcomes in first-episode psychosis. 18 RCTs identified (n=1808). Risk ratios with 95% CIs calculated. NNT/NNH computed where possible.

What this study cannot tell us

Studies were diverse, mostly small, and conducted by pioneering researchers with methodological limitations. Meta-analysis was inappropriate for most comparisons. Cannabis-psychosis therapy data came from a single small trial (n=47).

How to read the evidence

Cochrane systematic review representing the highest standard of evidence synthesis, though noting most included studies were small and methodologically limited.

When this study was published

Published in 2011. Early intervention services have expanded significantly since, with more robust evidence emerging.

The bigger picture

This authoritative Cochrane review established the evidence base for early intervention in psychosis as "emerging but inconclusive," influencing service development and research priorities globally.

Questions still open

  • Are early intervention gains maintained long-term? Can cannabis-specific therapy be improved and retested? What components of specialized early intervention are most effective?

Common questions

Does early intervention help in psychosis?
This review found emerging but not yet conclusive evidence that specialized early intervention teams improve outcomes, particularly for treatment retention and independent living. Individual specific treatments showed mixed results.
Is there a therapy specifically for cannabis and psychosis?
A cannabis-specific psychosis therapy was tested in one small trial (47 patients) included in this review but did not show significant benefit for reducing cannabis use.

Read the original research

Early intervention for psychosis.

The Cochrane database of systematic reviews, CD004718

Citation

Marshall, Max; Rathbone, John. (2011). Early intervention for psychosis.. The Cochrane database of systematic reviews, CD004718. https://doi.org/10.1002/14651858.CD004718.pub3

Explore the wider topic