No serious adverse events occurred in the capsule trial’s cannabinoid group. With only 29 participants and a wide comparative interval, that observation did not demonstrate protection against serious harm.
The review separated event severity
The historical Cochrane synthesis reported no serious events in the cannabinoid arm and three among 31 placebo participants. The estimated risk ratio was 0.15, with a 95% confidence interval from 0.01 to 2.83. Serious placebo events included worsening colitis and a complicated pregnancy and were considered unrelated to study drug.
At the same time, nonserious adverse events were common in the cannabinoid arm. An absence of serious events did not mean an absence of adverse experiences.
Zero is a count, not a guaranteed future risk
A small sample can observe zero events even when an uncommon risk exists. The comparative interval did not exclude a materially higher risk, and the limited follow-up could not characterize every delayed outcome. The trial was not designed to certify a lifetime safety profile.
Preserve what was actually assessed
Report event counts, seriousness, attribution and exposure duration separately. Do not infer that cannabis prevented worsening disease because serious events happened only in the comparator arm. The review’s conclusion remained uncertainty about efficacy and safety; neither a zero count nor a favorable point estimate resolved that uncertainty.
Related reading: why endpoint definitions matter.
Research cited
Kafil and colleagues (2018). Cannabis for the treatment of ulcerative colitis. DOI: 10.1002/14651858.CD012954.pub2.
Leave a Reply