A new longitudinal study indicates that the amount of dried cannabis flower consumed by naturally cycling individuals with premenstrual symptoms fluctuates across the menstrual cycle—peaking in the days just before menstruation and rising again around ovulation. The research, published in Addictive Behaviors, found these premenstrual increases in cannabis use were especially pronounced on days when individuals reported worse emotional sensitivity or greater interpersonal conflict.
The study situates these findings within the broader biology of the menstrual cycle, noting that ovarian hormones—primarily estradiol and progesterone—drive the cycle’s two main phases: the follicular phase, from the start of bleeding to ovulation, and the luteal phase, from ovulation until the next period. It also points to the heightened mood and behavioural shifts seen in those with premenstrual disorders in the days leading up to menstruation.
Historically, men have used cannabis at higher rates than women, but consumption among women has risen and the gender gap has narrowed. Women tend to meet criteria for cannabis use disorder after fewer years of use than men and experience more severe withdrawal symptoms, underscoring the need to identify female‑specific factors in substance use. Two emotion‑based theories are used to explain cycle‑related shifts in substance use: the self‑medication theory, which links use to relief of negative moods in the luteal and perimenstrual phases, and the reward‑sensitivity theory, which suggests pre‑ovulatory estradiol surges heighten responsiveness to pleasurable stimuli.
Evidence linking these theories to cannabis use has been inconsistent. A 2021 study of nearly daily cannabis users with premenstrual dysphoric disorder found increases in consumption during the premenstrual and menstrual phases, with a peak in the follicular phase, but the study did not biologically verify ovulation and used mood alone to diagnose the disorder rather than the full emotional symptom spectrum.
Leading the new work were Gianna Zorzini and Jennifer L. Gordon from the Reproductive Mental Health Research Unit in the Department of Psychology at the University of Regina, who designed the study to address these limitations and to explore how cannabis use relates to emotional symptoms across the cycle. Zorzini said: “Cannabis use among women has increased in recent years, yet we still know relatively little about why and when women use cannabis. We wanted to better understand whether cannabis use changes throughout the menstrual cycle and how these changes relate to emotional symptoms, particularly those associated with premenstrual disorders.”
Researchers analysed pre‑treatment data from a registered clinical trial evaluating a self‑help programme for premenstrual disorders; the raw dataset remained confidential. Sixty‑four naturally cycling individuals, aged 18 to 42, in Canada and the United States, were recruited via social media channels run by a premenstrual disorder advocacy organisation. All had regular 21‑to‑35‑day cycles and either a diagnosed or suspected premenstrual disorder, and they were not pregnant, breastfeeding, using hormonal contraception or hormone medications, nor undergoing certain therapies. Participants completed daily surveys each evening over two to three cycles.
At baseline, participants underwent clinical interviews and questionnaires assessing depression, anxiety, recent stressful life events, and lifetime trauma. Roughly 83 percent of the sample met criteria for another psychiatric diagnosis and about 44 percent were taking antidepressants. Each evening, participants logged whether they had used cannabis the previous day and reported the amount by product type, including grams of dried flower, milligrams of THC in edibles or oils, or hits of concentrate. They also rated daily symptoms on a six‑point scale covering depressed mood, anxiety, mood swings, anger or irritability, emotional or rejection sensitivity, and interpersonal conflict.
To account for cycles varying in length, researchers combined self‑reported menstruation dates with urine tests detecting luteinising hormone surges, which typically precede ovulation by about a day. They employed phase‑aligned cycle time scaling to map each cycle onto a continuous timeline anchored to both menstruation and ovulation, and used generalized additive mixed models to test for non‑linear cyclical patterns in cannabis use and quantity, followed by exploratory models assessing whether daily symptoms or baseline traits shaped those patterns.
Of the 64 participants, 25 used cannabis at least once across 47 to 87 tracked days (averaging 61.70 days and totalling 1,449 observations). Ovulation was confirmed by urine test for 49 cycles and estimated for five cycles; excluding those five cycles did not alter the results. Compared with 39 non‑users, cannabis users reported more recent stressful life events (averaging 4.8 versus 2.7) and had higher odds of a confirmed premenstrual disorder via daily tracking (about 80 percent versus 54 percent, roughly four times the odds, or nearly five times after adjusting for age, country and stress).
Users consumed cannabis on around 27.48 days on average, or 44.4 percent of tracked days. Eighty per cent of users consumed edibles or oils at least once (178 observations, averaging 3.56 milligrams of THC, ranging from 0.02 to 50.00 mg), and 68 per cent used dried flower (341 observations, averaging 0.95 grams, with a range of 0.02 to 5.00 grams). A further 32 per cent used concentrates (95 observations, averaging 3.91 hits). Some participants used other THC products, but those data were not included in the cycle modelling due to small numbers.
Across the group, the daily probability of cannabis use showed no statistically significant cyclical pattern, whether considering overall use, dried flower, or edibles and oils. Individual differences were large, and some participants exhibited distinct cycle‑specific patterns. However, the quantity of dried flower used on days of use did follow a clear cyclical pattern when anchored to the onset of menstruation, peaking premenstrually and rising again around ovulation. The authors caution that anchoring to ovulation weakened the signal at the group level, with substantial individual variation around ovulation.
Among the roughly 16 dried‑flower users, about half showed only a premenstrual increase, three showed only an ovulatory increase, and five showed both. The researchers describe the ovulatory rise as weaker than the premenstrual one, given the smaller number of individuals showing it. For THC‑infused edibles and oils, the group‑level pattern in milligrams consumed on use days was not statistically significant, though there was a slight rise shortly after menstruation and notable individual differences.
“We were surprised that increases in cannabis use before menstruation were more common than increases around ovulation,” Zorzini remarked. “Although we observed both patterns, they differed considerably between women. This raises interesting questions about why some women show different consumption patterns across the cycle.”
“Not everyone used more cannabis at the same points in their menstrual cycle or to the same extent,” she added. “This suggests that individual differences are important and that there is no single pattern that applies to everyone. Larger studies are needed to better understand what explains these differences.”
In exploratory within‑person analyses, daily fluctuations in core emotional symptoms altered the shape of the dried‑flower cycle: higher‑than‑usual daily emotional symptoms were linked to greater premenstrual dried‑flower consumption, while lower‑than‑usual symptoms corresponded with lower premenstrual intake. Around ovulation, lower‑than‑usual emotional symptoms were tied to greater dried‑flower consumption than higher‑than‑usual symptoms.
After correcting for multiple comparisons, higher‑than‑usual emotional or rejection sensitivity and interpersonal conflicts emerged as the specific daily symptoms shaping the premenstrual rise in dried‑flower quantity. Zorzini noted that “women tended to use more cannabis on days when they experienced greater emotional sensitivity or more conflicts with others than usual,” adding that the study cannot determine causality or whether cannabis was used specifically to cope with these symptoms.
Preliminary exploratory analyses suggested that baseline characteristics also influenced consumption curves, though the authors stressed that these tests had low statistical power due to small, uneven subgroups. Stronger premenstrual increases in dried‑flower quantity appeared among those with confirmed premenstrual disorder, older participants, those with more past traumatic events, and users not taking antidepressants. For edibles and oils, older age was linked to a steeper increase early in the luteal phase, in the days shortly after ovulation.
The authors propose that the rise in dried‑flower use around ovulation aligns with the reward‑sensitivity theory, as pre‑ovulatory estradiol surges may heighten responsiveness to rewarding effects when emotional symptoms are relatively low. They also suggest dried flower’s fast action may make it preferable for rapid relief of premenstrual symptoms, while ingested edibles and oils act more slowly, though the study did not record how cannabis was consumed. It also appears that those not on antidepressants may use cannabis premenstrually to cope with mood changes, whereas those on antidepressants might already have some symptom buffering pharmacologically.
Caution is warranted in interpreting the results. The study’s observational design cannot determine whether premenstrual symptoms drive higher cannabis use or whether cannabis alters subsequent symptom ratings. Dried cannabis flower often contains cannabidiol (CBD), a non‑intoxicating compound that some research suggests may ease menstrual symptoms, so cannabis itself may have affected symptom reporting, and the study did not measure CBD or THC concentrations.
As Zorzini explained, the generalisability is limited because all participants reportedly suspected they had a premenstrual disorder due to their emotional changes across the cycle. The small sample of 25 cannabis users further limits broad applicability, particularly for exploratory comparisons, and the researchers did not assess motives for use or histories of substance use disorders, leaving open whether the patterns reflect casual use, symptom relief, or dependence.
Future studies involving larger, more diverse samples are needed to confirm whether the observed cyclical peaks in dried‑flower use replicate and whether differences in reward‑seeking traits help explain why some users increase intake around ovulation. The researchers also emphasise the value of incorporating measures of product potency, ingestion methods, and daily motives to better explain cannabis use across the cycle and to inform female‑tailored prevention and treatment approaches.
“Our next step is to study larger and more diverse groups of women,” Zorzini told PsyPost. “We hope to better understand why and when cannabis use changes throughout the menstrual cycle for some women but not for others, and whether these patterns are related to emotional symptoms. Ultimately, this research could help inform more individualized prevention and treatment approaches.”
She added that women have historically been underrepresented in substance use research and that the menstrual cycle has often received limited attention, hoping the study encourages more research that recognises women’s unique experiences and factors influencing their substance use.
The study, Menstrual cyclicity of cannabis use: a longitudinal study using phase‑aligned cycle time scaling (PACTS), was authored by Gianna Zorzini, May Ly, Taryn Wahl, and Jennifer L. Gordon.
