Prevalence and characteristics of prenatal cannabis use in Michigan, USA: A statewide population-based pregnancy cohort.

Abstract

To estimate the prevalence of prenatal cannabis use in Michigan, USA, using self-reports and urine toxicology and to examine factors associated with use. Observational study using data from the Michigan Archive for Research on Child Health (MARCH) cohort, part of the National Institutes of Health's Environmental influences on Child Health Outcomes (ECHO) program. It is a prospective statewide pregnancy cohort recruited in 2017-2023 using births as a sampling frame. A three-stage stratified cluster sample design that randomly selected hospitals, prenatal clinics and mothers across the lower peninsula of Michigan, where 97% of Michigan hospital births take place. Pregnant individuals recruited at their first prenatal visit primarily from 20 clinics serving 11 hospitals. Of 1105 eligible pregnant participants, 1092 (98.8%) had information on cannabis use either from self-reports (n = 999), urine toxicology (n = 1028) or both (n = 945). Study participants had a median age of 29.3 years (interquartile range = 7.6). Cannabis use was self-reported by interviewer-administered phone surveys twice during pregnancy and further ascertained by measuring tetrahydrocannabinol (THC) metabolites in up to two urine samples collected during pregnancy. The prevalence of cannabis use was estimated using sampling weights and clustering by sampled hospital to account for the complex survey design. The weighted prevalence of prenatal cannabis use was 16.8% [95% confidence interval (CI) = 8.0%-25.7%] based on either self-report or urinalysis. Prevalence based on self-report alone was 12.3% (95% CI = 6.1%-18.6%) and from urinalysis alone was 13.3% (95% CI = 5.1%-21.6%). Adjusted for creatinine concentrations, levels of 11-nor-9-carboxy-tetrahydrocannabinol (THC-COOH) in positive samples ranged from 5.43 to 4694.89 ng/mg with a median of 122.00 ng/mg. Multivariate analysis revealed that participants who were single [adjusted prevalence rate ratio (aPRR) = 2.08, 95% CI = 1.15-3.78], had lower educational attainment (aPRR = 3.76, 95% CI = 1.23-11.53 for some college education and aPRR = 3.15, 95% CI = 0.97-10.25 for high school education as compared with undergraduate education or above), reported symptoms of depression (aPRR = 1.72, 95% CI = 1.19-2.48) and had a history of adverse childhood events (ACEs; aPRR = 2.04, 95% CI = 1.21-3.45 for ACEs ≥ 3 and aPRR = 1.64, 95% CI = 1.07-2.52 for 1-2 ACEs as compared to 0 ACEs) were more likely to use cannabis prenatally. Cannabis use is estimated to have occurred in one of six pregnancies in a population-based sample in Michigan, USA.