A Quiet Funnel Plot Could Not Rule Out Missing Cannabis Crash Studies
The older-driver review did not see obvious publication bias on funnel-plot inspection. With only a few studies per outcome, that observation could not establish that no relevant findings were missing.
The two evidence pools were small
The review included four collision-involvement studies and three culpability studies. Funnel plots were inspected for evidence of publication bias. A statement that inspection did not indicate bias describes the observed plot, rather than proving complete retrieval of every conducted study.
Sparse plots provide limited discrimination
When very few studies contribute, the shape is difficult to interpret. Differences in study design or sampling can also affect the pattern. This methodological limitation does not demonstrate that missing studies existed or predict which direction they would change the pooled estimate.
Search effort and certainty are distinct
The review used a systematic search, which is a strength. It still could not establish that every relevant analysis had been published, indexed and located. A defensible summary can report no obvious asymmetry while avoiding the stronger claim that publication bias was excluded. This matters when a nonsignificant pooled association is presented as definitive safety evidence: imprecision, observational design and possible reporting limitations remain even when a funnel plot looks unremarkable. The conclusion should match the small evidence base rather than treating a screening check as a guarantee.
Related analysis: the study’s main findings and limitations.
Research cited
Chinna-Meyyappan and colleagues. Risk of Motor Vehicle Collisions and Culpability among Older Drivers Using Cannabis: A Meta-Analysis. Brain Sciences (2023). DOI: 10.3390/brainsci13030421.
