For a long time, much of what we knew about the human body came from research that did not adequately represent half of the population.
This is not to say that women were completely absent from medical research. They were not. Nor is it accurate to suggest that modern mental-health research simply ignores biological sex. It does not. Over the last several decades, there has been substantial progress in including women in clinical research and examining sex and gender differences.
But there is a more complicated problem.
Being included in a study is not the same as being studied adequately.
A woman can be present in a clinical trial while the researchers never ask where she is in her menstrual cycle. She can be included in a psychiatric study without tracking whether her symptoms change predictably across the month. She can be included in a study of depression without examining whether reproductive transitions such as pregnancy, postpartum, perimenopause, or menopause alter her vulnerability to symptoms or her response to treatment.
And this distinction matters.
Because the female body is not hormonally static.
For someone who menstruates, the biological environment changes repeatedly across the lifespan and, on a shorter timescale, across the menstrual cycle itself. Estrogen and progesterone fluctuate. Ovulation occurs. Hormonal contraception can alter that hormonal environment. Pregnancy produces enormous endocrine changes. The postpartum period involves an equally dramatic transition. Later, perimenopause brings another period of substantial hormonal variability before menopause.
For decades, much of psychiatry has been built around diagnoses that describe symptoms without necessarily accounting for the biological rhythms in which those symptoms occur.
The question, then, is not whether hormones “cause” women’s mental health problems.
The more interesting question is:
How much of women’s psychological experience have we failed to understand because hormonal context was treated as a complication rather than as part of the phenomenon itself?
The History of the “Standard” Research Participant
The history is important because the problem did not emerge from nowhere.
Women of reproductive age were historically excluded from many areas of clinical research, particularly because of concerns about pregnancy and potential fetal harm. Early drug development was especially affected by this approach. Historical reviews describe the systematic exclusion of women of childbearing potential from early phases of drug research, alongside broader concerns about inadequate representation of women in clinical studies.
There was a logic to some of this caution.
There was also a consequence.
When research repeatedly uses one population as its default, findings from that population can quietly become treated as universal.
The problem is not simply that women were missing from individual studies. It is that an evidence base can accumulate around a body that is assumed to represent the human body in general.
This became particularly visible in medicine, where women were underrepresented in research across several disease areas and where sex-specific differences in treatment response were not always adequately examined. A systematic review of randomized controlled trials published in 2017 found a median enrollment of women of only 41% across several major medical fields, with representation declining further in studies involving older participants.
The story is therefore not one of complete exclusion followed by complete correction.
It is a story of partial inclusion without always achieving complete sex-sensitive research.
But Mental Health Is Not Just About Hormones
This is where the conversation needs some caution.
Women experience higher rates of several common mental-health conditions, including depression and anxiety, but these differences cannot simply be attributed to estrogen, progesterone, or the menstrual cycle.
Mental health is produced through an interaction between biology, psychology, relationships, culture, socioeconomic conditions, trauma, discrimination, caregiving responsibilities, reproductive experiences, and countless other factors.
A review of women’s mental-health research has pointed out precisely this complexity: differences in mental disorders between men and women raise questions not only about biological mechanisms but also about social and cultural influences.
The body and the social world are not competing explanations.
They interact.
A woman experiencing chronic stress, for example, is not experiencing “psychology” separately from her physiology. Stress affects biological systems. Hormonal states can influence emotional processing. Social circumstances influence stress exposure. And the meaning a person makes of those experiences can influence how symptoms develop and persist.
This is why a genuinely sex- and gender-informed approach to mental health cannot simply ask whether women are more depressed or anxious.
It has to ask why, when, under what conditions, and in relation to what biological and social changes.
The Menstrual Cycle Is Not Just a Reproductive Event
The menstrual cycle has historically been treated primarily as a reproductive phenomenon.
But the brain is also a hormone-sensitive organ.
Research increasingly suggests that psychiatric symptoms can fluctuate across the menstrual cycle, although the effects are not identical across people or disorders. A comprehensive review of psychiatric symptoms across the menstrual cycle found that the premenstrual and menstrual phases were the periods most consistently associated with symptom exacerbation across several psychiatric conditions. Depression, anxiety, stress, psychosis, mania, alcohol use, and severe outcomes such as suicide attempts have all been examined in relation to menstrual timing.
This does not mean that every woman’s mood changes significantly with her menstrual cycle.
It does not mean that every episode of irritability before menstruation is a hormonal disorder.
And it certainly does not mean that psychological suffering in women can be dismissed as “just hormones.”
In fact, the opposite may be more appropriate.
If hormonal fluctuations meaningfully affect symptoms in some people, dismissing those fluctuations as irrelevant is itself a failure of good assessment.
A systematic review of anxiety disorders, PTSD, and OCD found evidence of symptom exacerbation at particular points in the menstrual cycle, while also highlighting major methodological problems in the existing literature, including small samples, inconsistent methods for identifying menstrual phases, and limited representativeness.
That last part is important.
We are beginning to see the pattern.
But we are still learning how to measure it properly.
The Problem With Treating the Cycle as “Noise”
For researchers, variability can be inconvenient.
If one participant’s hormone levels are different from another’s, and if the same participant’s hormonal state changes across the month, the data become more complicated.
Historically, one response to biological variability has been to control for it.
But what happens when the variability is part of the biology we are trying to understand?
This is one of the central problems in menstrual-cycle research.
If hormonal fluctuations influence emotional processing, stress responses, cognition, or psychiatric symptoms, then controlling away those fluctuations may make the experiment cleaner while simultaneously making it less representative of real life.
The question changes from:
“How much variability can we eliminate?”
to:
“What does the variability tell us?”
Modern research has increasingly moved toward the latter question. NIH policies introduced the consideration of sex as a biological variable in research involving vertebrate animals and humans, reflecting growing recognition that sex can influence physiology, disease expression, and treatment response.
But changing research policy does not instantly change the accumulated evidence.
There is still considerable work to do.
And the Problem Begins Before Human Clinical Trials
The gap is not confined to human participants.
For years, basic neuroscience research disproportionately used male animals. An analysis of thousands of neuroscience articles found persistent sex bias, including studies using only male animals and studies that included both sexes without actually analysing sex as an experimental variable.
This matters because clinical research does not begin with the human clinical trial.
There is a chain.
Animal models inform biological theories. Biological theories influence clinical hypotheses. Clinical trials test interventions based on those hypotheses. Guidelines are then developed from the accumulated evidence.
If one sex is underrepresented at earlier stages, the resulting knowledge base can inherit that imbalance.
This does not mean that research conducted in male animals is automatically irrelevant to women.
It means that we should be careful about assuming that biological mechanisms discovered predominantly in one sex are automatically universal.
PMDD Shows Us What Happens When Timing Matters
Perhaps one of the clearest examples comes from premenstrual dysphoric disorder (PMDD).
PMDD is not simply “bad PMS.”
It is a clinically significant condition in which severe emotional and behavioural symptoms occur in a cyclical pattern, typically during the luteal phase, and improve around the onset of menstruation.
What makes PMDD scientifically interesting is that the problem is not necessarily the presence of abnormal hormone levels.
Many researchers now understand PMDD in terms of heightened sensitivity to normal hormonal fluctuations in susceptible individuals.
In other words, the question may not be:
“Does this person have too much or too little estrogen?”
It may be:
“How does this person’s brain respond to the normal hormonal changes occurring across the cycle?”
That distinction is enormous.
It suggests that two people can experience the same biological fluctuation and have very different psychological responses to it.
And it provides a useful model for thinking about women’s mental health more broadly: biological change does not affect everyone in the same way.
There Is Another Pattern We Often Miss: Premenstrual Exacerbation
PMDD is only one piece of the puzzle.
Another phenomenon is premenstrual exacerbation (PME), where an existing psychiatric condition becomes worse during particular phases of the menstrual cycle.
Someone may already have depression, anxiety, PTSD, OCD, or another psychiatric condition.
Then, at certain points in the cycle, their symptoms intensify.
This is clinically important because the person may not have a menstrual-specific disorder at all.
Their underlying disorder may simply be cycle-sensitive.
Recent reviews have argued that premenstrual exacerbation remains underexplored and poorly defined despite evidence that hormonal fluctuations can coincide with worsening across multiple psychiatric disorders.
This distinction matters for diagnosis.
If clinicians only ask, “Are you depressed?”, they may miss the pattern.
If they ask, “When do your depressive symptoms become worse?”, they may begin to see something different.
The Menstrual Cycle May Matter for Risk, Not Just Mood
This is where the subject becomes more serious.
Research has found associations between menstrual timing and severe psychiatric outcomes.
A meta-analysis of 32 studies involving 3,791 participants reported higher risks of suicide death, suicide attempts, and psychiatric admission around menstruation, although the authors also noted limitations in the available evidence.
These findings should not be interpreted as meaning that menstruation itself causes suicidal behaviour.
That would go far beyond the evidence.
Mental-health risk is multifactorial.
But if psychiatric symptoms and risk can change systematically across the menstrual cycle for some people, cycle timing may become clinically relevant information rather than background information.
A patient saying,
“I don’t know why, but every month there are five or six days when everything becomes unbearable,”
is telling us something potentially important.
That pattern deserves investigation.
The Same Story Appears Again During Pregnancy and After Birth
The reproductive story does not stop with menstruation.
Pregnancy involves profound physiological and hormonal changes. So does the transition immediately following childbirth.
Postpartum mental health has received considerable research attention, but the broader point remains important: reproductive transitions can represent periods of altered psychological vulnerability.
The challenge is separating hormonal effects from everything else happening simultaneously.
Childbirth itself is enormous.
Sleep deprivation can be extreme.
Identity changes.
Relationships change.
Work and financial circumstances may change.
The demands of caregiving increase.
There may be pain, breastfeeding difficulties, birth trauma, or previous psychiatric vulnerability.
Hormonal shifts are therefore not an isolated explanation.
They are one component of a much larger biological and psychosocial transition.
This is precisely why reproductive mental health requires interdisciplinary thinking rather than a simplistic “hormones versus psychology” framework.
And Then Comes Perimenopause
Perhaps one of the clearest examples of a reproductive transition that has historically received less psychological attention is perimenopause.
Perimenopause is not simply the moment menstruation stops.
It is a period of transition characterised by changing and increasingly variable ovarian hormone patterns.
And emerging research suggests that this period can be psychologically significant.
A 2024 systematic review and meta-analysis of prospective studies found that women in perimenopause had a significantly higher risk of depressive symptoms and diagnoses compared with premenopausal women, with a pooled odds ratio of 1.40.
A newer 2026 systematic review and meta-analysis involving more than one million participants likewise found substantial burdens of depressive, anxiety, and insomnia symptoms during the perimenopausal and postmenopausal periods, although estimates varied considerably across studies.
Again, this does not mean menopause causes depression in every woman.
But it does suggest that reproductive stage belongs in the clinical conversation.
If a 45-year-old woman presents with new-onset anxiety, disrupted sleep, irritability, low mood, cognitive complaints, and changes in functioning, asking only about her psychological history may provide an incomplete picture.
The reproductive history may matter too.
What Have We Been Calling “Psychological” That Might Be Biological, Social, or Both?
This is perhaps the most important question.
A woman experiencing worsening concentration before menstruation might be experiencing ADHD-related difficulties that fluctuate with hormonal changes.
A woman experiencing intense irritability and hopelessness during the luteal phase might be experiencing PMDD or premenstrual exacerbation.
A woman in perimenopause may experience sleep disruption, anxiety, and depressive symptoms alongside profound hormonal variability.
A woman with PTSD may notice that certain trauma-related symptoms intensify at predictable points in her cycle.
None of these possibilities eliminates psychological explanations.
They complicate them.
And clinical psychology is at its best when complexity is not treated as an inconvenience.
We Also Need to Be Careful With the Word “Women”
There is an important scientific distinction between sex and gender.
Not every woman menstruates. Not everyone who menstruates identifies as a woman. Some women have undergone hysterectomy or menopause; some use hormonal contraception; some have endocrine or reproductive conditions that alter their hormonal patterns.
So when researchers talk about “women’s mental health,” they can sometimes be referring to a mixture of biological sex, gender identity, reproductive status, social experience, and cultural expectations.
These are related but not interchangeable variables.
Good research needs to distinguish them rather than collapsing them into a single category.
This matters particularly when studying hormonal mechanisms.
If we want to understand how ovarian hormones influence mental health, we need biological measures and reproductive histories.
If we want to understand how gendered social experiences influence mental health, we need measures of those experiences too.
One cannot substitute for the other.
What We Still Don’t Know
Perhaps the most interesting part of this field is how much remains unanswered.
We still do not fully understand why some people are highly sensitive to normal hormonal fluctuations while others are not.
We do not yet have sufficiently precise models for predicting who will experience premenstrual exacerbation.
Research methods for determining menstrual phase have varied substantially, making studies difficult to compare. Reviews have specifically highlighted problems with small samples, inconsistent phase definitions, and inadequate representativeness.
Even when menstrual cycle information is clinically relevant, it is not routinely assessed in many mental-health settings.
A recent clinical review described the menstrual cycle as an overlooked potential factor in psychiatric care and argued for more systematic cycle-informed assessment.
And there is a practical problem.
Gold-standard assessment of menstrual-related psychiatric symptoms often involves prospective daily symptom tracking across multiple cycles. Research and clinical guidance increasingly support this approach, but collecting that information in routine clinical settings remains difficult.
So the gap is not simply a lack of awareness.
It is also a methodological challenge.
What Would Better Mental-Health Research Look Like?
Better research would not mean conducting every study exclusively in women.
It would mean asking better questions.
When sex matters, researchers should measure it and analyse it.
When reproductive stage matters, it should be recorded.
When menstrual-cycle phase could influence the outcome, it should be measured rather than assumed away.
When studying psychiatric medication, researchers should examine whether sex and reproductive status influence pharmacokinetics, side effects, or therapeutic response.
When studying mental-health symptoms, researchers should consider whether the symptoms are stable or cyclical.
And when researchers find a difference between men and women, they should not immediately assume that biology is the explanation.
Social experience deserves equal scrutiny.
The goal is not to replace one simplistic model with another.
It is to stop pretending that one model can explain everyone.
What This Means for Clinical Psychology
For clinicians, this does not mean attributing every emotional change to hormones.
That would be another form of reductionism.
It means becoming curious about timing.
When did the symptoms begin?
Are they continuous or episodic?
Do they change across the menstrual cycle?
Did they emerge during pregnancy, postpartum, or perimenopause?
Did the person have similar symptoms before beginning hormonal contraception?
Do symptoms worsen predictably before menstruation?
Does an existing psychiatric condition become more severe at particular points in the cycle?
These questions can add another dimension to assessment without invalidating the psychological experience.
A woman does not become “less psychological” because hormones are involved.
Hormones are part of the biological context in which psychological experience occurs.
The Larger Problem Is Not That Women Are “Too Hormonal”
There is a danger in having this conversation badly.
Women have historically been dismissed as irrational, emotional, unstable, or “hormonal.” A renewed emphasis on biological sex should never become another way of reducing women’s experiences to their reproductive systems.
The point is not that hormones explain women.
The point is that hormonal biology is one part of the person that mental-health science has sometimes failed to measure adequately.
There is an enormous difference between saying,
“She’s behaving this way because she’s hormonal,”
and asking,
“Could predictable hormonal changes be interacting with her psychological symptoms in a clinically meaningful way?”
The first dismisses.
The second investigates.
The Bottom Line
Mental-health research has changed considerably.
Women are no longer simply absent from clinical research. Sex is increasingly recognised as a biological variable. Reproductive psychiatry has become a legitimate and growing field. PMDD is formally recognised, and research into menstrual-cycle effects, pregnancy, postpartum mental health, and menopause continues to expand.
But inclusion is not the same as understanding.
We are still learning how much hormonal variability matters, for whom it matters, and how it interacts with psychological vulnerability, social experience, trauma, stress, and psychiatric illness.
Perhaps the most important shift is therefore not from “women were ignored” to “now we understand women.”
We are not there yet.
The more honest conclusion is that we are finally asking better questions.
What if the timing of symptoms matters?
What if reproductive transitions are periods of psychological vulnerability for some people?
What if a disorder that appears constant is actually fluctuating?
What if the same diagnosis looks different at different hormonal stages?
What if the biological variability we once tried to control for is actually part of the phenomenon?
And what if understanding mental health properly requires us to stop searching for a single explanation—biological, psychological, or social—and instead study how all three continually shape one another?
For women, the question is no longer simply whether they are represented in mental-health research.
It is whether the research has been designed to see them clearly.