ObjectivePrenatal alcohol exposure (PAE) places children at risk of fetal alcohol spectrum disorder (FASD) but ascertainment of PAE is problematic. Early intervention for children at risk of FASD may help mitigate long-term difficulties. Phosphatidylethanol (PEth), a metabolite of alcohol, is incorporated into red cell membranes and can be measured in dried blood spot (DBS) cards. In the UK, DBS samples are collected on day 5 for routine newborn screening. We sought to examine if PEth measured from DBS correlates with postnatal maternal self-report of alcohol consumption in pregnancy.DesignObservational population-based study. Comparison of infant PEth concentration and self-report of maternal alcohol use during pregnancy.SettingLarge maternity unit in Glasgow, Scotland.ParticipantsAll singleton mother–infant dyads delivered during each fourth consecutive 24-hour period.InterventionsMother: direct, confidential, immediate postnatal interview by a single researcher examining alcohol use during pregnancy. Infant: one extra DBS collected coincident with routine newborn screening if bleeding continued.Results92.5% of eligible mothers agreed to participate. 510 DBS were obtained of which 502 were successfully analysed. 216 (43%) samples contained PEth at a concentration of ≥8 ng/mL and 148 (29.5%) at ≥20 ng/mL. The sensitivity of PEth ≥8 ng/mL and ≥20 ng/mL in identifying women who self-reported modest alcohol use after 36 weeks’ gestation was 50% and 36.4%, respectively.ConclusionPEth measured from DBS obtained on day 5 of life does not reliably identify modest PAE after 36 weeks’ gestation from maternal self-report.