Among a sample of pregnant and breastfeeding women who used cannabis to address a mental health concern, only a quarter used a psychiatric medication. Mirroring these findings are those from the general population where mental illness is often not treated with medications or left untreated. According to the National Alliance on Mental Illness, 1 in 5 US adults experience some type of mental illness. ²⁷ Yet, only 46% of those affected received treatment in 2020. ²⁷ In the United States, 19% (48 million people) have anxiety, 8% (21 million people) have depression, and 3.6% (9 million people) suffer from both of these conditions in a given year. ²⁷ Prevalence among perinatal individuals is similar, with 1 in 5 women experiencing mental health problems like depression, anxiety, PTSD, or obsessive-compulsive disorder. ²⁸ The numbers rise considerably if women experience violence or are from a high-risk group. ²⁹ Byatt et al. ³⁰ reported that 75% of women with mental health problems do not receive treatment. This is consistent with our findings.

The high proportion of reported mental distress is particularly concerning because of the increased risks to both mother and infant after perinatal women screen positive for depression and/or anxiety. ³¹ However, only 22% of women who screen positive for depression actually follow through with the referral.^30,32 In our study, only 25% of the pregnant and breastfeeding women who used cannabis to treat depression or anxiety also used a psychiatric medication. This meant that either mothers were untreated, they used non-pharmacologic treatment (such as psychotherapy), or they self-medicated with cannabis.

For women taking medications, stopping during pregnancy is risky. Womersley et al. ³³ concluded that women who discontinued their antidepressants while pregnant were three times more likely to struggle with a subsequent episode of depression compared to those who continued their treatment throughout their pregnancy. Untreated depression during pregnancy is associated with increased risk of preeclampsia, preterm birth, low birth weight, inadequate diet, and failure to seek prenatal care. ¹⁸ Furthermore, untreated mental illness has also been associated with increased risk of fetal harm due to increased use of substances like caffeine, tobacco, alcohol, and recreational drugs. The effects of untreated mental illness can be so devastating that a 20% increase in suicide attempts has been noted among adolescent mothers.^12,18,33

Major medical organizations provide little direction regarding the pharmacological treatment of mental health disorders, with statements supporting psychopharmacology for “some pregnant patients” or upon an individualized consideration of risks and benefits. ³⁴ This lack of concrete direction may contribute to physician uncertainty about recommending pharmacologic options when offering interventions for mental illnesses perinatally. ³⁵

A 2012 study of postpartum depression found that women were more likely to be prescribed medication treatment (with or without counseling) if their medical practice received specialized training. ³⁶ A 2022 review on the knowledge, attitudes, and decision-making process of patients and providers regarding antidepressant use during pregnancy found that while physicians were confident in these skills, many overestimated the teratogenic risk, and between 20% and 74% avoided starting an antidepressant during pregnancy. ³⁵ Furthermore, between 24% and 47% would recommend discontinuing antidepressant use once pregnant. ³⁵ Clearly, there is a wide range of opinions among physicians regarding this topic. Ultimately, such variations in practices based on provider beliefs, comfort levels, and a lack of guidance of high-quality recommendations leave many women untreated for depression. ³⁵

All mothers in our sample reported using cannabis to address mental health concerns, yet less than a quarter screened positive for anxiety or depression, and 88% were classified as having mild or no psychological distress in the composite PHQ-4. Although substantial differentially missing data for PHQ-4 limits our ability to draw conclusions about psychological distress between those who did and did not use psychiatric medications, the rate of self-reported symptoms (100% of the sample) is higher than it appeared on the PHQ-4. This mismatch between PHQ-4 results and self-report can be easily explained. First, the PHQ-4 asks about symptoms in the past 2 weeks. The questions about health issues that mothers treated with cannabis could have occurred at any time. That number, by definition, will be higher than a scale that focuses only on current symptoms. Second, this mismatch may result from the way that the word “depression” is used by laypeople to describe feeling down or sad, and “anxiety” may be used to describe stress rather than pathological depression or anxiety which have clearly defined criteria according to the DSM-5.^37,38 Third, their treatments (pharmaceutical, cannabis, or other) may have been effective, meaning that they were not depressed or anxious when they completed the survey.

Regarding the low rate of use of medications, many mothers and providers fear the effects of psychiatric medication on the fetus or infant, or they are concerned about pressure to take these medications. ³⁹ Womersley et al. ³³ found that only 33% of women thought it was acceptable to use SSRI medications during pregnancy. Colaceci et al. ⁴⁰ found that attitudes about psychiatric medications among breastfeeding mothers fell into one of the three categories: to discontinue breastfeeding to take their medication, to “endure the pain,” or to use natural products that they believe are safer than prescribed pharmacologic agents. ⁴⁰ In a qualitative study, many women described choosing to use cannabis over acetaminophen or psychiatric medications, in an effort to avoid “chemicals” or side effects from a medication. ⁹

In contrast, some pregnant or lactating women perceive cannabis as a natural alternative to medications.^9,40 This perception was prevalent in our sample and could help explain why the majority of respondents were not taking psychiatric medications despite reporting mental health concerns. Chang et al. ⁹ found that many mothers were aware of the risks and, thus, avoided smoking tobacco during their pregnancy, even though it is natural; conversely, they continued cannabis use, reporting limited knowledge or no knowledge of the potential risks. ⁹ Women in their study believed that cannabis “elevates your mood” while psychiatric medications “suppress your mood” and that the “chemicals [in antidepressant medications] mess with you.” ⁹ Similarly, a Swedish study by Kvillemo et al. ⁴¹ reported that individuals who used cannabis used it to reduce anxiety, associated its use with a “calm and cozy” effect, and believed public health information about its harms were not credible. ⁴¹ In our study, some respondents also used other natural medications such as St. John’s Wort (n = 10). St. John’s wort is a flowering plant that is native to Europe and Asia (Hypericum perforatum), and is the most widely prescribed antidepressant in the world. ⁴²

Many societal stressors may contribute to the development or worsening of mental illness perinatally. Risk factors for peripartum depression include lack of social support, financial stress, food insecurity, poor physical health, stressful life events, and intimate partner violence.^{43 bibr44-17455057231199391–45} While we may advocate for and aim to address improvements in economic, health, and societal equity, such situations are often inadequately addressed, and pregnant and postpartum individuals experience substantial challenges. If such hardships cause mental distress or illness to the point of self-medicating with cannabis, a conversation about risks and benefits of medications versus cannabis use is warranted.

Healthcare providers should inquire about and address maternal use of cannabis to self-treat mental illness as evidence suggests that it may exacerbate symptoms. ⁴⁶ Our findings showed that half of the mothers used cannabis daily, with 32.8% using cannabis up to three times daily and 17.6% using cannabis four or more times daily. The National Academies of Sciences, Engineering, and Medicine (NASEM) report that frequent cannabis use, especially with high doses, may cause feelings of anxiety and paranoia. ⁴⁷ Furthermore, heavy cannabis use is associated with an increased risk of the development of depressive disorders as well as thoughts of suicide, suicide attempts, and death by suicide. ⁴⁷

Cannabis use can cause neuropsychological impairment, particularly when used during adolescence. Its use can decrease motivation, and heavy use is associated with apathy, and a reduced ability to concentrate, follow routines, or successfully master new content. ⁴⁸ Notably, these symptoms are similar to those of depression, so it may be difficult to distinguish the symptoms of cannabis use from those of depression. Cannabis use is also associated with cognitive impairment, which may put both the mother and child at risk. Beyond depression, cannabis use is considered a preventable risk factor for psychosis. ⁴⁹ The age of onset of psychosis appears to be correlated with the age of onset of cannabis use. ⁴⁸

The effect of in utero exposure to cannabis on the cognitive development of the offspring remains unclear as methodological challenges of studies to date limit the conclusions which can be drawn.^{50 bibr51-17455057231199391–52} Importantly, a lack of clear data in humans does not equate to a lack of risk. Animal studies, which tend to use higher doses than would be consumed by humans, have demonstrated that prenatal exposure to cannabis increases the excitability of the dopamine neurons in the offspring which may lead to enhanced impulsivity, hyperactivity, loss of sustained attention, and increased sensitivity to drugs of abuse. ⁵³ It is unclear how such findings relate to humans. ⁵⁴

Some mothers, including in the present study, report using cannabis while pregnant to address anxiety. Unfortunately, some research reports higher anxiety among children exposed to cannabis prenatally. Rompala et al. ⁵⁵ reported greater anxiety, aggression, and hyperactivity levels in young children exposed to cannabis, as well as alterations in gene expression which may suggest possible mechanisms for these phenotypic differences.^55,56 These findings suggest the potential for long-term neurodevelopmental impacts on the next generation, effects that are not widely known by the public. ⁵⁶

On average, individuals in our sample used cannabis for the first time around the age of 16. This is very similar to the average age of first cannabis use in females across the United States which is 16.3 (SD 3.2). ⁵⁷ This age may be important because many parts of the brain are still developing during this time and into early adulthood. ⁵⁸ Research suggests that when comparing the amygdala size of 16- to 18-year-old females, the amygdala volume of those who used cannabis was larger than that of those who did not use cannabis.^58,59 This finding is associated with more self-reported depression and anxiety. ⁵⁸ Similarly, cannabis use can impact the development of the prefrontal cortex, an area of the brain which has been implicated with major depression.^58,60

Cannabis–drug interactions have been established in vitro that involve cytochrome P450 (CYP) enzymes, including CYP1A2 and CYP3A4. ⁶¹ Based on these data, several drug interactions may exist between cannabis and the medications that our study respondents reported using. Cannabis is a substrate of CYP3A4, and St. John’s Wort, which was used by 10 survey respondents, is a CYP3A4 inducer. The induction of this enzyme may decrease the serum concentration of cannabis, thereby reducing the effects of cannabis in individuals. Cannabis is also a CYP1A2 inducer and may decrease the serum concentration of CYP1A2 substrates, such as duloxetine, clozapine, and olanzapine, potentially decreasing the therapeutic effects of these medications. Though in vivo data on these drug interactions are limited, cannabis smoking cessation is among the implications of two cases of intoxication with CYP1A2 substrates (clozapine and olanzapine).^62,63 The subsistence of adverse reactions from cannabis–drug interactions remain unknown, and merit future study.

Although the differences in psychiatric medication use by race/ethnicity in our sample did not reach significance, our sample may not have been large enough to capture possible differences. An analysis of the National Ambulatory Medical Care Survey (NAMCS, 2006–2010) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) demonstrated marked differences in rates of provider-reported maternal depression diagnoses by race during prenatal visits. ⁶⁴ They found that non-Hispanic White women were more likely to be diagnosed with depression than non-White women (5.7% vs 2.9%). Furthermore, among those with a depression diagnosis, non-Hispanic White women were over three times more likely to report antidepressant use during pregnancy compared to non-White women (adjusted odds ratio [aOR] 3.3, 95% CI 2.1, 5.3). ⁶⁴ As there are clear influences of race and ethnicity on antidepressant use during pregnancy, the inclusion of cannabis into the discussion also merits further investigation.