Validity of self-report measures of cannabis use compared to biological samples among women of reproductive age: a scoping review

Abstract

BackgroundMost existing evidence about the prevalence of prenatal cannabis use relies on self-reported measures, which is limited by social desirability bias and recall bias. To date, several studies have examined the validity of self-reported measures of prenatal cannabis use, but this evidence has yet to be synthesized. To address this gap, we performed a scoping review to systematically identify and synthesize existing evidence on the validity of self-reported measures of cannabis use among pregnant women.MethodsWe searched PubMed, PyschINFO, CINAHL, Cochrane/CENTRAL, and Google Scholar for peer-reviewed studies published in English between January 2010 and June 2021. We included studies that compared self-reported measures of cannabis use to a biochemical measure of cannabis (e.g., urine, hair, meconium) in pregnant women. We excluded studies reporting solely on prenatal cannabis use prevalence as well as those that examined self-reported drug use in which cannabis use was not a distinct category.ResultsWe found 12 unique studies (11 primary studies and one systematic review) that examined the validity of self-reported prenatal cannabis use, compared to a biochemical sample. Most studies were conducted in the US and conducted in either a hospital or clinical setting. We found that self-report was more valid in populations with a current or prior history of drug use. Self-report was also more valid when assessed via interviews by research team members than health care provider screenings or self-administered surveys. The most commonly used biochemical measure used was urine drug testing, which was found to have the highest level of concordance with self-report.ConclusionsThis scoping review systematically mapped existing evidence on the validity of self-reported prenatal cannabis use. Although much remains unknown in this area, an important next step is a systematic review that would provide robust evidence on clinical utilization of self-reported use in conjunction with biochemical samples. Further research is needed to examine validity by type of measure and mode of administration. Additionally, future studies could assess factors associated with disclosure of use across different critical maternal health periods beyond pregnancy.