One in Ten Women With Cannabis Use Disorder in Pregnancy Ended Up in a Psychiatric Emergency

Public health researchers have spent a decade documenting what happens to infants exposed to cannabis in the womb. Far less is known about what happens to the women themselves in the year after they give birth, and until now no study had followed a statewide population.

A cohort of more than 2 million California deliveries, published in Annals of Epidemiology, puts a number on it. Among women diagnosed with cannabis use disorder during pregnancy, 10.4 percent made a psychiatric emergency department visit in the postpartum year. Among women with neither cannabis nor alcohol use disorder, 2.7 percent did. Women diagnosed with alcohol use disorder, included as a benchmark, came in at 15.6 percent.

The adjusted relative risk for cannabis use disorder was 1.39 (95% CI 1.32–1.47) against the no-diagnosis group, and 1.80 (1.56–2.09) for alcohol use disorder. LaMecia De Alba and Sidra Goldman-Mellor at UC Merced, with Claire Margerison of Michigan State and Alison Gemmill of UCLA, built annual cohorts from statewide linked hospital and emergency department discharge records covering 2016 through 2021 — the years when legal recreational cannabis sales began in the state.

The trend line underneath the association is the part policymakers will notice. Prenatal cannabis use disorder prevalence rose from about 1.0 percent to 1.4 percent over the six-year window, an increase of roughly 40 percent. Alcohol use disorder diagnoses among pregnant women stayed flat.

What Two Million Birth Records Show

The outcome measured was not a self-reported symptom scale but an actual emergency department encounter for a psychiatric reason. As Goldman-Mellor put it in the university's announcement, these are not women saying they feel depressed — these are psychiatric symptoms serious enough that they had to go to the emergency room.

The study is part of a larger National Institutes of Health grant examining substance use, violence victimization and self-harm among pregnant and postpartum women. The exposure is a coded diagnosis, not any reported cannabis use, which means a woman who uses cannabis occasionally and is never diagnosed appears in the comparison group.

Why the Benchmark Group Changes the Story

Two details complicate the most obvious reading. The first is the benchmark itself. Alcohol use disorder during pregnancy carried a higher adjusted risk of postpartum psychiatric emergency care than cannabis use disorder did, which undercuts any framing of cannabis as a uniquely dangerous exposure during pregnancy. The legal substance that expectant mothers are counselled most aggressively to avoid produced the worse outcome in this data.

The second is arithmetic. Comparing the raw rates — 10.4 percent against 2.7 — yields roughly a fourfold difference. After adjustment, the relative risk is 1.39. Most of the gap between those two figures is explained by the other characteristics the model accounted for, which is a reminder that a diagnosis of cannabis use disorder in pregnancy travels with poverty, prior mental health contact and disrupted care.

Goldman-Mellor is direct about the causal limit. The study was not designed to show that cannabis use disorder causes psychiatric emergencies, she said, only that women who are addicted to cannabis during pregnancy are not a neutral group. Detection is another open question: a woman flagged for cannabis use during prenatal care is, by that fact, inside a system that screens, refers and records.

What the paper does establish is that this is no longer a rare presentation. For clinicians working with cannabis dependence, the finding argues for mental health screening that continues past delivery rather than stopping at the prenatal visit, and for dual diagnosis care that treats the psychiatric condition and the substance together instead of sequencing them.