Heavy prenatal alcohol exposure and obstetric and birth outcomes: a Danish nationwide cohort study from 1996 to 2018

Marcella Broccia*, Bo Mølholm Hansen, Julie Marie Winckler, Thomas Larsen, Katrine Strandberg-Larsen, Christian Torp-Pedersen, Ulrik Schiøler Kesmodel

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

6 Citations (Scopus)
69 Downloads (Pure)


BACKGROUND: Heavy alcohol use during pregnancy can harm the fetus, but the relation to most obstetric outcomes remains unclear. We therefore aimed to describe maternal characteristics and estimate the association between heavy prenatal alcohol exposure and 22 adverse obstetric and birth outcomes. METHODS: We carried out a Danish nationwide register-based historical cohort study, including all singleton births from Jan 1, 1996, to Dec 31, 2018. Births of women who had emigrated to Denmark were excluded from the study due to missing data and women who migrated within 1 year before or during pregnancy were also excluded due to loss to follow-up. Data were extracted from the Danish Medical Birth Register, the Danish National Patient Registry, the Danish National Prescription Registry, the Danish Civil Registration System, and the Population Education Register. Logistic regression models were used to estimate crude and adjusted odds ratios (ORs) of obstetric and birth outcomes. Heavy alcohol use was defined by hospital contacts for alcohol-attributable diagnoses given to the mother, her infant, or both, or maternal redeemed prescriptions for drugs to treat alcohol dependence within 1 year before or during pregnancy. FINDINGS: Of 1 191 295 included births, 4823 (0·40%) were defined as heavily alcohol-exposed and 1 186 472 were categorised as a reference group with no identified heavy prenatal alcohol exposure. Heavy-alcohol-exposed births more often had mothers with psychiatric diagnoses (49·8% vs 9·6%), substance use (22·0% vs 0·4%), tobacco use (64·3% vs 15·8%), and low educational level (64·1% vs 17·6%) than did the reference group. For heavy-alcohol-exposed births, significantly increased adjusted ORs were found for small for gestational age (OR 2·20 [95% CI 1·97-2·45]), preterm birth (OR 1·32 [1·19-1·46]), haemorrhage in late pregnancy (OR 1·25 [1·05-1·49]), and preterm prelabour rupture of membranes (OR 1·18 [1·00-1·39]). Decreased adjusted ORs were found for postpartum haemorrhage (500-999 mL; OR 0·80 [95% CI 0·69-0·93]), gestational diabetes (OR 0·81 [0·67-0·99]), planned caesarean section (OR 0·82 [0·72-0·94]), pre-eclampsia and eclampsia (OR 0·83 [0·71-0·96]), and abnormalities of forces of labour (OR 0·92 [0·86-0·99]). INTERPRETATION: Heavy prenatal alcohol exposure is associated with adverse obstetric and birth outcomes and high proportions of maternal low educational level, psychiatric disease, and lifestyle risk behaviours. These findings highlight a need for holistic public health programmes and policy attention on improving pre-conceptional care and antenatal care. FUNDING: The Obel Family Foundation, The Health Foundation, TrygFonden, Aase and Ejnar Danielsens Foundation, The North Denmark Region Health Science and Research Foundation, Holms Memorial Foundation, Dagmar Marshalls Foundation, the A.P. Møller Foundation, King Christian X Foundation, Torben and Alice Frimodts Foundation, the Axel and Eva Kastrup-Nielsens Foundation, the A.V. Lykfeldts Foundation.

Original languageEnglish
JournalThe Lancet. Public health
Issue number1
Pages (from-to)e28-e35
Publication statusPublished - Jan 2023

Bibliographical note

Copyright © 2023 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY-NC-ND 4.0 license. Published by Elsevier Ltd.. All rights reserved.


  • Cesarean Section
  • Cohort Studies
  • Denmark/epidemiology
  • Female
  • Humans
  • Infant
  • Infant, Newborn
  • Pregnancy
  • Premature Birth
  • Prenatal Exposure Delayed Effects/epidemiology


Dive into the research topics of 'Heavy prenatal alcohol exposure and obstetric and birth outcomes: a Danish nationwide cohort study from 1996 to 2018'. Together they form a unique fingerprint.

Cite this