OBJECTIVE: To determine the association between ethnic group and likelihood of admission to intensive care in pregnancy and the postnatal period. DESIGN: Cohort study. SETTING: Maternity and intensive care units in England and Wales. POPULATION OR SAMPLE: A total of 631 851 women who had a record of a registerable birth between 1 April 2015 and 31 March 2016 in a database used for national audit. METHODS: Logistic regression analyses of linked maternity and intensive care records, with multiple imputation to account for missing data. MAIN OUTCOME MEASURES: Admission to intensive care in pregnancy or postnatal period to 6 weeks after birth. RESULTS: In all, 2.24 per 1000 maternities were associated with intensive care admission. Black women were more than twice as likely as women from other ethnic groups to be admitted (odds ratio [OR] 2.21, 95% CI 1.82-2.68). This association was only partially explained by demographic, lifestyle, pregnancy and birth factors (adjusted OR 1.69, 95% CI 1.37-2.09). A higher proportion of intensive care admissions in Black women were for obstetric haemorrhage than in women from other ethnic groups. CONCLUSIONS: Black women have an increased risk of intensive care admission that cannot be explained by demographic, health, lifestyle, pregnancy and birth factors. Clinical and policy intervention should focus on the early identification and management of severe illness, particularly obstetric haemorrhage, in Black women, in order to reduce inequalities in intensive care admission. TWEETABLE ABSTRACT: Black women are almost twice as likely as White women to be admitted to intensive care during pregnancy and the postpartum period; this risk remains after accounting for demographic, health, lifestyle, pregnancy and birth factors.