Saliva and Blood THC Can Correlate Without Supporting a Reliable Conversion

A correlation between oral-fluid and blood THC does not establish that one specimen can reliably predict the other person’s concentration or effects. Controlled vaporization research found broad oral-fluid variability despite robust overall associations.

The study included alcohol co-use

Nineteen cannabis smokers participated in a within-person study of vaporized cannabis, with and without oral alcohol. Researchers collected oral fluid, blood and plasma and repeatedly measured subjective effects.

The analysis used measured THC and breath-alcohol concentrations after nominal cannabis-dose groups did not produce sufficiently distinct concentrations. Nominal product potency and actual exposure were therefore not interchangeable even inside the experiment.

Association is not agreement

Oral-fluid cannabinoid concentrations correlated with blood and plasma, but specimen ratios varied considerably. A correlation can describe a shared overall trend while leaving wide uncertainty around the prediction for an individual sample.

The authors found blood THC more consistent than oral fluid for modeling the reported intoxicating effects under the tested conditions. This was not validation of a universal blood threshold or a saliva-to-blood conversion factor.

Alcohol changed the time course

Alcohol prolonged some cannabis subjective effects. Concentration–effect relationships also showed hysteresis, meaning that time within the exposure course affected interpretation.

These were ratings of experience, not proof that every driving skill changed by the same amount. The study did not directly establish the cause of a particular crash or a universal waiting interval.

Important population limits

Participants using cannabis more than three days weekly were excluded, so the findings do not automatically describe very frequent users. Product, route, alcohol sequence and sampling time also matter.

When reviewing paired results, identify the collection device, analytes, matrices and timing. Treat saliva as its own specimen with its own uncertainties. An impressive correlation coefficient should not be used to conceal the variability that matters when interpreting one person’s result.

For context, see our discussion of THC concentrations and the separate question of impairment.

Primary research

Rebecca L. Hartman and colleagues. Controlled Vaporized Cannabis, With and Without Alcohol: Subjective Effects and Oral Fluid-Blood Cannabinoid Relationships. Drug testing and analysis (2015). DOI: 10.1002/dta.1839.