Poor Baseline Tracking Performance Changed the Analyzed Cannabis Sample

Three occasional smokers’ critical-tracking data were excluded because their baseline performance limited the ability to measure further deterioration. The resulting task comparison did not include every enrolled occasional participant.

The task required an interpretable starting point

The Critical Tracking Task measured the limit at which participants could maintain control of a changing signal. Poor baseline performance can leave little usable room for detecting a postdose decline. The investigators excluded three occasional users from that task analysis, leaving eight occasional participants compared with fourteen frequent users.

The denominator depended on the outcome

Those exclusions did not mean that the participants disappeared from every other study measure. A statement that all task results came from the same twenty-five people would therefore be inaccurate. Excluding difficult baselines may improve interpretation of a task’s change score while narrowing the population represented by that analysis.

Separate the overall test from a specific contrast

The overall critical-tracking session, group and interaction tests were nonsignificant, although a specific 1.5-hour contrast favored frequent users relative to baseline. Both findings should be retained. The task-specific exclusion cannot establish how the omitted participants would have performed under better testing conditions or supply an individual determination of driving impairment.

Related analysis: blood THC and task effects across use groups.

Research cited

Desrosiers and colleagues (2015). Smoked Cannabis’ Psychomotor and Neurocognitive Effects in Occasional and Frequent Smokers. DOI: 10.1093/jat/bkv012.