Marijuana use in young mothers and adverse pregnancy outcomes: a retrospective cohort study.

To evaluate the association between marijuana use and a composite adverse pregnancy outcome using biological sampling.Retrospective cohort study.Single tertiary center.Young women (13-22 years old) with singleton, non-anomalous pregnancies delivered from September 2011 to May 2017.Exposure was defined as marijuana detected on universal urine toxicology testing or by self-report. Multivariable logistic regression modelling was used to estimate the effect of any marijuana use on the primary composite outcome. The effect of marijuana exposure was also estimated for self-reported use, toxicology-detected use, and multiple use detected by toxicology.The primary composite outcome included spontaneous preterm birth, hypertensive disorders of pregnancy, stillbirth, or small for gestational age.Of 1206 pregnant young women, 17.5% (n = 211) used marijuana. Among the women who used marijuana, 8.5% (n = 18) were identified by self-report alone, 63% (n = 133) by urine toxicology alone, and 28.4% (n = 60) by both. Urine toxicology testing results were available for 1092 (90.5%) births. The composite outcome occurred more frequently in pregnancies exposed to marijuana (46 versus 34%, P < 0.001). This remained significant after adjusting for race/ethnicity and tobacco in the multivariable model (adjusted OR 1.50, 95% CI 1.09-2.05). When marijuana exposure was defined by self-report only, the association with adverse pregnancy outcome became non-significant (adjusted OR 1.01, 95% CI 0.62-1.64).In a population of young women with nearly universal biological sampling, marijuana exposure was associated with adverse pregnancy outcomes. The heterogeneity of findings in existing studies evaluating the impact of marijuana on mothers and neonates may result from the incomplete ascertainment of exposure.Marijuana use, as detected by universal urine testing, was associated with a composite adverse pregnancy outcome among young mothers.