For questions about whether a drug is safe for nursing mothers, few doctors are considered as authoritative as Thomas W. Hale. A clinical pharmacologist at Texas Tech who specializes in drug exposure during pregnancy and breastfeeding, Hale is also the author of Medications and Mothers' Milk, the definitive text on drug safety during breastfeeding, and the founder of InfantRisk, a website dedicated to the subject.
Like the World Health Organization, which recommends mothers breastfeed exclusively for the first six months after their infants are born, Hale is a staunch advocate of breastfeeding, and his work has a fundamental mission: to improve breastfeeding rates by demystifying the interaction between various commonly prescribed drugs and breast milk.
"[S]tudies suggest that the use of medications is one of the major reasons women discontinue breastfeeding prematurely," the website for Medications and Mothers' Milk says. "Discontinuing breastfeeding is often the wrong decision and most mothers could easily continue to breastfeed their infants and take the medication without risk to the infant."
The book, as well as the InfantRisk website, are chock full of handy reference information for busy clinicians, like how much oxycodone a mom can safely take while nursing, or whether Adderall is safe to breastfeed on (it is). When it comes to cannabis, however, Hale is decidedly not down. Cannabis is listed as an L5, or "contraindicated," substance in his book, carrying the highest risk rating and the following warning:
Studies in breastfeeding mothers have demonstrated that there is significant and documented risk to the infant based on human experience, or it is a medication that has a high risk of causing significant damage to an infant. The risk of using the drug in breastfeeding women clearly outweighs any possible benefit from breastfeeding. The drug is contraindicated in women who are breastfeeding an infant.
In other words, the leading authority on what is and isn't safe for pregnant and nursing mothers thinks you would be better off not breastfeeding at all than smoking weed while you do it. To him, it's non-negotiable, and what he writes becomes de facto policy.
Joelle Puccio, a perinatal nurse at a childbirth center outside of Seattle, learned this the hard way. Last January, she joined a committee working to draft a new drug policy for the hospital. The hospital had no previous policy on substance exposure during breastfeeding, so Puccio, also a staunch advocate of breastfeeding as well as an active member of drug policy reform group the People's Harm Reduction Alliance, volunteered to write one.
"Initially, I had included some myth-busting," she said. Along with clearing up misconceptions about having hepatitis C—it's safe to breastfeed if you have the condition as long as your nipples aren't cracked or bleeding—Puccio's policy was going to include a recommendation that cannabis use was not a contraindication for breastfeeding. Her own research on the subject yielded little evidence of risk—certainly not enough to discourage occasional marijuana users from breastfeeding if they want to; it has massive benefits for infants and moms alike. However, when it came time for the committee's final review, Puccio was dismayed to discover that, despite her extensive research, she was the lone voice in support of this particular recommendation.
"I argued for a while, but it was agreed that the best thing to do was follow the established expert," Puccio said. "All the managers agreed that we had to go by Hale and not by my literature review, which is the exact right thing for them to do most of the time. It's just that in this instance, Hale is wrong."
Frustrated that her colleagues—"good, smart nurses"—refused to look past Hale's book, Puccio took it on herself to fact-check his work. She began by picking through his article on perinatal cannabis use for InfantRisk.com, which warns nursing mothers of scary health consequences in infants exposed to pot-laden breast milk. It ends "strongly advis[ing]" mothers to avoid "any form of THC" while pregnant or breastfeeding.
Puccio found what she characterizes as a clear overstatement of risk on Hale's part. While the article cites a number of studies regarding cannabis use during pregnancy, it cites only two on cannabis use during lactation: one conducted with 27 women, and another with 16 women. While the InfantRisk article notes that "[d]ata continues to suggest that cannabis may produce long-term... reduced cognition and changes in mood and reward," neither of these studies found many concrete indicators of harm. Indeed, the first study Hale cites was conducted with 27 women who used cannabis every day while breastfeeding, and it found "no differences in growth, mental and motor development."
Even the studies with actual data—the ones on pot use during pregnancy—were often out of date, out of context, or unsupportive of Hale's conclusions, according to Puccio. Angry at what she felt was unnecessary bias against cannabis-using moms, Puccio wrote a polite yet scathing email to Hale, disputing his article's claims one by one.
She started with three studies on in utero cannabis exposure, cited by Hale to back up his claim that "THC crosses the placenta readily, and there is increasing evidence that it may increase rates of growth retardation and adverse neurodevelopment following prenatal exposure." Puccio found none of this research convincing:
The first article (Day et al) used to support this claim finds no effect of marijuana exposure on any of the studied outcomes. The second (Fried et al) found transient effects in the immediate newborn period and at 4 years. They also found subtle possible long-term effects, but were unable to find a causal link, noting that 'the only conclusive statement would be that, if there are long term consequences of prenatal exposure to marihuana, such effects are very subtle.' The 3rd article (Hurd et al) did find that 'there was a 0.08-cm and 14.53-g significant reduction of foot length and body weight.' This is statistical significance, not clinical significance. These articles do not support your claim.
The three studies that Puccio mentioned form the basis of a literature review on cannabis and lactation that Hale cites as a suggestion of "severe long-term neurobehavioral consequences." Building on those shaky foundations, and using carefully qualified statements, he goes on to introduce a number of unnecessarily ominous conclusions.
"Recent studies have suggested a reduction in long and short-term memory retrieval and retention in children exposed to prenatal cannabis," he writes. "These children were also weak in planning, integration, and judgment skills." No citation.
"Both human cohort studies and studies in animals clearly suggest that early exposure to cannabis is not benign and that cannabis exposure in the perinatal period may produce long-term changes in behavior and mental health." No citation.
"Chronic cannabis use may cause depression, anxiety, and bipolar disorder in adolescents and adults." Not only does it not have a citation, but it also has nothing to do with the topic at hand.
Hale also suggests a link between prenatal cannabis exposure and childhood leukemia, citing a 1989 study from the journal Cancer. However, he conveniently fails to mention that the same study was duplicated about 15 years later by the same author with a much larger cohort. On the second go, the author actually found a slightly negative association between cannabis and leukemia.
Upon receiving Puccio's email, Hale forwarded it to one of his employees at the InfantRisk center, Dr. James Abbey. Initially, Abbey was receptive to Puccio's complaints, noting that it was one of the first things he'd flagged for review upon taking the job at InfantRisk.
"In my opinion, [the article] overstates the risks and misrepresents the level of evidence used to support its conclusions," Abbey wrote back to Puccio. "I haven't gotten around to addressing it yet." Abbey pulled the article offline for revisions.
However, a few days later, Puccio found the original article reposted with no changes. She got back in touch with Abbey, but he'd changed his tune, justifying cannabis's L5 listing as necessary in light of, vexingly, the fact that we don't have enough studies on the subject. After that exchange, the line went dead. When asked to comment for this article, Abbey did not respond. Hale declined.
Abbey's description of cannabis fits Hale's L3 category—"moderately safe"—to a tee. L3 substances, according to Medication and Mothers' Milk, are substances for which "there are no controlled studies in breastfeeding women, however the risk of untoward effects to a breastfed infant is possible; or, controlled studies show only minimal non-threatening adverse effects. Drugs should be given only if the potential benefit justifies the potential risk to the infant."
In her subsequent discussion with Abbey, Puccio urged him to drop cannabis to an L3. Abbey, downplaying the harm of keeping cannabis at an L5 level, suggested that any good clinician would conduct a thorough review of all available information, not just Hale's book, and arrive at their own decision. Puccio dismissed this as fantasy—and a willful denial of how Hale's book is supposed to function in the first place.
"Your average nurse practitioner or neonatologist is going to go to that book, open it up, look at the L5, close the book, and go tell their patient that they can't breastfeed [if they ever smoke weed]," Puccio said. "That's what happens in real life."
Switching cannabis to an L3 classification, Puccio says, would at least inform providers that the situation wasn't cut and dry and perhaps prompt them to conduct that deep dive Abbey suggested, instead of simply grabbing the bottle and calling it a day.
While this might seem like an obscure technical quibble, it's not. Women in America are constantly shamed and stigmatized for using cannabis during pregnancy or breastfeeding, no matter how responsibly or infrequently they use. (They are also constantly shamed and stigmatized for pretty much any choices they make when it comes to breastfeeding in general.) Though we know very little about the actual effects of cannabis use during lactation, there is a common perception that it is extremely dangerous and horribly irresponsible to ingest THC while breastfeeding. There are very real consequences for pregnant and breastfeeding women whose urine shows up positive for pot, even in states where the drug is legal. Many child protection agencies and state health organizations still consider exposing infants to THC, no matter how trace the amount, to be abuse or endangerment; according to the Family Law and Cannabis Alliance, some states even apply "delivery of drugs" to minors laws to cases of women using marijuana while breastfeeding.
Bernadette Hoppe, a lawyer in New York who defends parents and represents children in abuse cases, confirmed this. "I certainly have cases where the only allegation against the parent is that they were using marijuana," she said. She said that, while Hale's book itself could not be used as evidence in an abuse case, it could certainly inform expert witnesses.
But even when there are no legal proceedings involved, the stigma rears its ugly head. One InfantRisk.com forum user, who identified himself as a county health official from an area with heavy medical marijuana use, noted that "Dr. Hale's book lists marijuana in the most dangerous category, and consequently our State Health Division, relying on that information, will not provide any supports to lactating women who are marijuana users." This support could take the form of government assistance (WIC) to purchase breast pumps, for example. "As I read the literature, the known effects of marijuana in this situation are largely undetermined, so I am confused by its categorization," the user continues. "As the Health Officer for our county I want to be sure that we support breastfeeding activity when appropriate, and of course discourage it when it would be dangerous. How can grouping marijuana along with chemotherapeutic drugs and heavy metals be justified by the existing literature?"
In his response to this user, Hale decried that policy, saying that he does not agree "that you should deny women WIC services, pumps, and other support simply because they occasionally use marijuana, or have used it in the past." Nevertheless, he seemed unaware of or unwilling to acknowledge the role he played in creating it. (He did note he has "struggled for years" with marijuana's L5 classification.)
Breast milk has all sorts of benefits, including, according to one study, increasing one's IQ later in life; some would argue that if drinking pot-laced breast milk makes babies dumber, well, so does not drinking it. As Abbey himself says in a forum post from 2014, "The [benefits of breastfeeding] are tangible and well-established, while the [risks of cannabis use during breastfeeding] are vague and scary."
"The literature is not well-established enough in this area to be of much help in making useful generalizations," Abbey continues. "It is our considered opinion that the benefits of breastfeeding probably outweigh the detriment caused by occasional or 'casual' exposure to cannabis. That detriment increases with more frequent use and, at some point, will exceed the benefits. We do not have enough information to say where that division lies." Yet marijuana's L5 categorization remains, promoting stigma as well as disproportionately affecting low-income mothers.
Hoppe, who also happens to have a master's of public health, was a little more direct.
"What I do know from my public health background," she said, "is that there are very few reasons to absolutely not breastfeed. Being HIV positive is one of them. There are a handful of drugs, including prescription drugs, that you should not breastfeed if you're taking. But aside form that, the benefits of breastfeeding are so strong and so well documented that I think we need to be extremely careful when we tell people to not breastfeed."