BACKGROUND: The US guidelines recommend avoiding marijuana during breastfeeding given concerns about infant's neurodevelopment. In this setting, some physicians and hospitals recommend against or prohibit breastfeeding when marijuana use is detected during pregnancy. However, breastfeeding is beneficial for infants and women, and stigmatization of substance use in pregnancy has been historically linked to punitive approaches with a disproportionate impact on minority populations. We advance an empirically informed ethical analysis of this issue.

METHODS: First, we performed a retrospective cross-sectional qualitative study of prenatal and postpartum records from a random sample of 150 women delivered in an academic hospital system in 2017 to provide evidence and context regarding breastfeeding management in relation to marijuana use. We then perform a scoping literature review on infant risks from breastmilk marijuana exposure and risks associated with not breastfeeding for infants and women. Finally, we analyze this issue vis-a-vis ethical principles of beneficence, autonomy, and justice.

RESULTS: (1) Medical records reveal punitive language pertaining to the medicinal use of marijuana in pregnancy and misinterpretation of national guidelines, e.g., "patient caught breastfeeding and instructed to stop." (2) Though there are plausible neurodevelopmental harms from breastmilk exposure to THC, evidence of infant effects from breastmilk exposure to marijuana is limited and largely confounded by concomitant pregnancy exposure and undisclosed exposures. By contrast, health benefits of breastfeeding for women and infants are well-established, as are harms of forgoing breastfeeding. (3) Discouraging breastfeeding for women with marijuana use in pregnancy contradicts beneficence, as it neglects women's health considerations and incorrectly assumes that risks exceed benefits for infants. Restrictive hospital practices (e.g., withholding lactation support) compromise maternal autonomy and exploit power asymmetry between birthing persons and institutions, particularly when compulsory toxicology screening prompts child welfare investigations. Finally, recommending against breastfeeding during prenatal care and imposing restrictions during postpartum hospitalization may exacerbate racial disparities in breastfeeding and related health outcomes.

CONCLUSIONS: Policy interpretations which discourage rather than encourage breastfeeding among women who use of marijuana may cause net harm, compromise autonomy, and disproportionately threaten health and wellbeing of underserved women and infants.