The expanded re-cap of the lecture by Drs. Teresa Baker and Thomas Hale on their recent publication, “Transfer of Inhaled Cannabis into Human Breast Milk.”

Introduction to the lecture

Dr. Teresa Baker and Dr. Thomas Hale traveled from the Texas Tech Health Sciences Center in Amarillo, Texas to Denver, Colorado to present their paper, “Transfer of Inhaled Cannabis Into Human Breast Milk.”  The lecture was co-hosted by the Colorado Breastfeeding Coalition and Elephant Circle, with a big shout out and thanks to Katie Halverstadt of COBFC and Laurel Wilson of Mother Journey.  They were both essential to the planning and facilitation of the event.

I was honored to have the chance to introduce Drs. Baker and Hale and I took the opportunity to thank them for the thoughtful ways in which they conducted their research. 

I said something along these lines,

When the InfantRisk Center decided to pursue doing research in CO, they reached out to lactation groups as well as cannabis groups, to include the perspective of the potential subjects as early as the study planning and theorizing stage.  Constituent-led research is rare, and it was an honor to collaborate with all the groups involved.

It turned out to be a good idea to have involved local organizations with previously cultivated relationships with parents who use cannabis, as they ended up recruiting nearly all the subjects who participated.  We quickly learned that there were some significant barriers to health care and lactation professionals recruiting subjects.  Those folks who were board-certified, registered or licensed expressed two main obstacles: 1) the fear that by recruiting subjects they would look like cannabis advocates and that would damage their relationships with colleagues and clients, and 2) the fear of the tension created by recruiting subjects while being a mandatory reporter.  As a result, subject recruitment became a grassroots effort and was slower going than was originally hoped - 9 subjects participated over the course of 10 months of active subject recruitment. 

There is a saying that was popularized in this country by the disability justice movement -

“Nothing about us without us.”

Drs. Baker and Hale showed what it looked like to put that saying into action, and we at Elephant Circle were grateful to be part of such a collaboration.  Thank you to all the organizations involved: CannaMama Clinic, Badass Breastfeeders of Denver, COBFC, the Colorado Lactation Consultant Association, and Mother Journey as well as those individual practitioners who also helped recruit subjects.  Thank you!

The lecture itself, starting with background from Dr. Thomas Hale:

Dr. Hale presented the first half of the lecture with an overview of pharmacokinetics, the endocannabinoid system, what we know about the pharmacokinetics of cannabis in human milk, and the reason for doing this research in the first place.

Before he started, he wanted to be very clear about his and Dr. Baker’s stance on cannabis use during lactation:  THEY DO NOT SUPPORT IT.  Let me say that one more time - Neither Dr. Hale or Dr. Baker support or advocate for the use of cannabis during breastfeeding.  They do not believe it to be “safe.”

They do not support the use of cannabis during breastfeeding in part because we are ignorant about it.  The perspective of the InfantRisk Center is to provide information on the pharamacokinetics of any drug that may be used by nursing parents.  In this way, individual consumers, along with their practitioners, can make data-driven decisions about the risk of a particular drug in their particular situation.

(A quick note about safety and risk.  They inform each other but are not interchangeable concepts.  Describing the risk of something as low does not mean you believe it to be safe.  Safety is the lack of harm, but harm is a subjective idea.  What looks like harm to me may not look like that to you.  Risk - sometimes talked about as risk factors - can be low or accumulate to high levels, but risk is not an absolute concept either.  Something that might be low risk for one family, might present more risk in another family, based on their circumstances.  Safety is the evaluation of the known risks with the known benefits in the context of an individual’s life to decide if it is something to pursue during breastfeeding.  Now, sometimes, the risk is so high that it is never advisable for anyone - it cannot be used safely under any circumstances - but that is a rare phenomenon in the metabolism of drugs in breastmilk.  In describing the risks associated with drugs that nursing parents use, the InfantRisk Center is providing information to be used in the evaluation of the use of those drugs, not commenting on their inherent safety.)

Dr. Hale reminded us that he produces information for folks to use in their individual lives.  He wants to meet them where they are and help them make good decisions for them.  He reminded us that we can never know the journey of another person, and it is important not to judge them for the decisions they make.

He also reflected that when we are in the “ignorant phase” of a drug’s use (like we are currently with cannabis), there can be hysteria about its use.  He recalled that despite the deep fears about the effects of in utero exposure to LSD and crack cocaine, the reality showed that the childhood environment and access to early education were the most impactful factors for childhood development, much more than drug exposure. He assured us that solid data collection would move us out of ignorance and would allow us to “work through the hysteria.”  Good reminders!


This refers to the way that pharmacologic agents (drugs or medications) are metabolized by the body.  Pharmacokinetics involve parameters that describe how quickly a drug enters the bloodstream after consumption, what the accumulation of that drug looks like, how quickly the drug is broken down and then removed from the blood stream and where that drug goes after it leaves the bloodstream.  Pharmacokinetics is also concerned with the bioavailability of the drug, both to the nursing mother and the baby.  In other words, how much of the dose is experienced by the person exposed to the dose?  In the case of this study, how much THC is available to reach the mother’s brain after smoking or reach the infant’s brain via the oral intake of breastmilk?

The Endocannabinoid System

This system is part of the human body - we are all born with it -  and it consists of molecules called cannabinoids that affect the human body through their interaction with cannabinoid receptors (the two most well-researched receptors are CB1 and CB2). We have only known about this system for the past 15 years or so.  The term “endo” means internal and the fact that humans have an endocannabinoid system is the reason that cannabis exerts any effect on the human body (cannabis is a plant full of cannabinoids - called phytocannabinoids). It is not known what happens to the endocannabinoid system when exposed to an exocannabinoid (external cannabinoid) like THC.

Dr. Hale went on to tell us that 2-AG (2-Arachidonoylglycerol, an endocannabinoid) is found in human milk.  Animal studies suggest the endocannabinoid system is essential for the suckling reflex at birth (pups will die without their endocannabinoid system), but we do not know the role of 2-AG in human milk.  This is one of Dr. Hale’s next research projects.

He also summarized the important details to know about the molecules involved in cannabis consumption, metabolism and detection - if you need a refresher, please check out our Molecules 101 fact sheet.  And, if you would like a review or a prep for what you need to know in background to understand the paper, please check out our 10 concepts to understand when reading or thinking about cannabis and breastfeeding.

What we know about the pharmacokinetics of Cannabis

This summarizes what you need to know to understand their paper:

  1. The bioavailability (availability to the body) of THC when smoked is 5-50%.  This wide range is due to individual variation in smoking efficiency and personal metabolism.
  2. THC is not absorbed well orally.  90-95% of the THC taken orally is broken down by the liver, so only 1-5% is bioavailable when consumed orally (through human milk).
  3. THC enters the human system and is quickly redistributed - broken down and re-routed to other tissues.
  4. Brain is not one of the primary tissues for THC to reside long-term.  Adrenals, adipose and the thymus are the sites for long-term detection of THC.
  5. In plasma, smoking results in a peak THC of 150-250 ng/mL.
  6. Consuming cannabis orally results in a peak at 1.0 ng/mL, more than 100 times lower than smoking.
  7. In animal models (chimps) fetal blood shows 1/6 of the maternal blood concentration of THC.  The placenta is a good filter for THC exposure.
  8. Regular users (daily or near daily) tend to have plasma levels of THC that remain between 1-5 ng/mL for at least 30 days.  Occasional users do not show this static baseline.

Why study this in Colorado?

The InfantRisk Center studies the pharmacokinetics of drugs used by nursing parents.  They receive hundreds of calls a month about cannabis, so they decided to do a study that would improve upon what was in the literature.  The only other investigation to describe the movement of THC into breastmilk was published in 1982, had only two subjects and did not control for the dose or the timing of the consumed cannabis.  Good pharmacokinetic studies must control for the dose and timing of the drug being studied, so there was definitely room for improvement.

Drs. Baker and Hale decided to come to CO because it was already legal for breastfeeding parents to use cannabis.  It was important that the study design not appear to promote cannabis use, so subject recruitment focused on folks who were already using cannabis.

An important - and unique - element of their study design was that participants could give consent for participation just by participation.  This allowed subjects to participate anonymously and confidentially and protect them from legal retribution.  The Institutional Review Board (the IRB) in Texas reflected the way research was done in the early days of HIV - it could be risky for a person to sign their name admitting they had HIV (or were gay), so consent was given via participation, not through the traditional signature route.  Women who participated in Colorado were protected by this nuanced approach by the IRB of Texas Tech.  Another example of ethical research approach executed by Drs. Baker and Hale.

The second half of the lecture, details about the paper itself, was presented by Dr. Teresa Baker.

Dr. Teresa Baker, primary author of the paper, gave the second half of the lecture and delivered pure, delightful science!  I have broken the information down into sections for ease of reading.

The Study Design

Goal - Evaluate the transfer of delta-9-THC, 11-OH-THC and COOH-THC in human milk

Dose - 0.1g of Prezidential Kush cannabis flower (23.18 g THC)

Consumption - Smoked in an unused glass pipe

Timing of milk collection - Evaluate milk prior to cannabis consumption and at 20 minutes, 1, 2 and 4 hours after consumption

Subjects - Term baby, exclusively breastfed, 1-6 months old, abstain from all cannabis for 24 hours prior to participation, already cannabis users, able to pump and not nurse baby for 6 hours while participating, 4 reported they used cannabis infrequently (not all participants quantified their use) and one reported using cannabis 7-10 times in the past week.

A total of 9 subjects met the criteria and sent their milk to Texas for analysis.

Data from eight of those subjects are reported in the paper.

Technical note: It should be noted that instrumentation used by the Texas Tech team is the best available and their expertise is important to this project.  Extracting THC and its metabolites from human milk can be a tricky task.  1.9 ng/mL THC was determined to be the lower detection limit.

The Results

The paper reports:

  1. “An exclusively breastfeeding infant ingests an estimated 2.5% of the maternal dose (range 0.4-8.7%).”
  2. “Our findings demonstrate breast milk concentrations of delta-9-tetrahydrocannabinol peaked at 1 hours with a peak of 94 ng/mL (range = 12.2-420.3), and receded slowly over the subsequent 4 hours.”
  3. “The results at zero hour were exceedingly low in 6/8 participants, suggesting that most participants were able to abstain from using cannabis for 24 hours before the collection of the samples.  A relatively low but measureable concentration of delta-9-tetrahydrocannabinol was found at zero time in two of the participants, which suggests some residual accumulations of delta-9-tetrahydrocannabinol either from prior heavy use or use close to the start of breast milk collection.”
  4. Neither of the metabolites (11-OH-delta-9-tetrahydrocannabinol nor 11-nor-9-carboxy- delta-9-tetrahydrocannabinol) were detected in milk (their detection limit was even lower than for delta-9-tetrahydrocannabinol at 0.097 ng/mL).

Note:  The paper refers to the full scientific names of the molecules studied and refers to delta-9-tetrahydrocannabinol, which I have called “THC” throughout this document.  The paper calls the metabolites 11-OH-delta-9-tetrahydrocannabinol and 11-nor-9-carboxy- delta-9-tetrahydrocannabinol.

What are the takeaways from these results, point by point?

1. The data collected allowed the authors to calculate a relative infant dose (RID) of 2.5%.  This means that these infants were exposed to 2.5% of their mother’s dose.  Of that 2.5%, only 1-5% is biologically available to the baby orally (see above for more about the bioavailability of various methods of consumption).

The RID is a way to talk about how much would an infant receive over the course of the day if exposed to a known maternal dose.  It is based on this calculation:

infant dose mg/kg/day  /  maternal dose mg/kg/day

This resulting ratio is then multiplied by 100 to express RID as a percent. 

In the case of this paper,

Infant dose = 53.5 ng/mL x *150 mL/kg/day = 0.008 mg/kg/day

Maternal dose = 23.18 mg/*72kg/day = 0.32 mg/kg/day

0.008/0.32 = 0.025 x 100 = 2.5


the terms with an * are generalized assumptions that babies consume 150 mL/kg/day and the average maternal weight is 72kg

So, in this study, infant exposure would have been 2.5% of 0.32 mg/kg/day, or .008 mg/kg/day.  But because oral bioavailability of THC is so low, 1-5%, the amount available would have been 1-5% of that amount, or 0.00008 to 0.00004 mg/kg/day.

2. Time zero - the milk collection prior to cannabis consumption - represents the baseline THC concentration for each subject.  Subjects were asked to abstain from any cannabis for 24 hours prior to participation.  For all but two of the subjects, there was no THC detected in their milk prior to their consumption in this study. 

The detection of THC in two of the subjects could be the result of not abstaining for 24 hours, having a slower metabolism of THC than other subjects or may be indicative of consistent use.  Interestingly, the subject who reported the highest use (7-10 times in the past week) did not have any detectable THC at baseline.

3. THC levels measured at their peak one hour after smoking in all subjects, though absolute peak levels varied among subjects. 

Note that this is just the measured peak - there were no measurements taken between 20 minutes and 60 minutes (because human milk would be hard to pump at 20, 40 and 60 minutes) - so the peak could have been somewhere between 20 and 60 minutes.  These peak levels declined over the next 3 hours and were approaching the levels seen prior to smoking by 4 hours after consumption.

4. Neither metabolite was detected in the milk at any time during the study.  (For a refresher on the molecules involved in cannabis consumption and detection, check out our Molecules 101 fact sheet.) 

This may be because 1) these two molecules are more water soluble and more polar than the parent molecule THC, restricting their entry into the breastmilk compartment or 2) the relatively short timecourse of the study did not allow for the detection of these metabolites.


Next Steps for the InfantRisk Center:

Drs. Baker and Hale’s rich research program has lots of plans for the future of this topic.  They shared two main topics for immediate future research:

  1. They want to study regular users.  Several observations from this pilot study suggested the pharmacokinetics may differ between occasional and regular users.  The InfantRisk Center would like to look at this variable.
  2. Endocannabinoids in human milk!  Why are they there? What do they do??

Please find a full, free copy of the paper here if you have not yet downloaded it.

Questions?  Please contact Heather at 720-504-8206 or