The relationship runs in both directions. That’s the part that keeps catching researchers off guard: not merely that premenstrual disorders and psychiatric conditions tend to cluster in the same women, but that each one predicts the other, forwards and backwards across time. A woman with depression is roughly twice as likely to later develop a premenstrual disorder. A woman with a premenstrual disorder is roughly twice as likely to develop depression. The arrow, it turns out, points both ways simultaneously.
A large Swedish study published this week in JAMA Network Open puts the most robust numbers yet to what clinicians have long suspected. Following more than 100,000 women with clinically diagnosed premenstrual disorders (PMD) across an average of 8 years, researchers at Karolinska Institutet found bidirectional associations between PMD and nearly every psychiatric condition they examined, from anxiety to ADHD to bipolar disorder to personality disorders.
PMD is an umbrella category covering premenstrual syndrome and the more severe premenstrual dysphoric disorder, both characterised by mood swings, anxiety, and irritability that emerge and recede with the luteal phase of the menstrual cycle. Somewhere between 20 and 30 percent of women of reproductive age experience moderate or severe PMS; premenstrual dysphoric disorder, which is disabling enough to seriously impair daily functioning, affects roughly 2 to 6 percent. What makes the condition biologically strange is that the hormonal fluctuations driving the symptoms aren’t abnormal in themselves. Women with PMD appear to have an atypical sensitivity to normal hormonal changes, a kind of signal amplification in the brain’s response to the monthly ebb and flow of oestrogen and progesterone.
That sensitivity, the study suggests, may be part of something larger. A shared vulnerability, written partly in genetics and partly in the biology of stress.
The Swedish team drew on national registers covering 3.6 million women, cross-linking health records across psychiatry, gynaecology, and primary care to track who developed what, and in what order. Nearly half (47.8%) of women with PMD had received a psychiatric diagnosis before their PMD diagnosis, compared with about 30 percent of matched controls. Looking forward from the point of PMD diagnosis, 36.6% went on to develop a psychiatric condition during follow-up, versus 21.1% of unaffected women. Both figures translated to roughly doubled risk after adjusting for demographics and socioeconomic factors.
A Connection That Survives Family Controls
Crucially, the team also ran what are called sibling analyses, comparing women with PMD against their unaffected full sisters, raised in the same households with the same parents. This design strips out a lot of confounding: shared genetics, shared childhood environment, shared socioeconomic trajectories. The associations attenuated, but held. That they survived the sibling comparison is perhaps the study’s most telling finding. It suggests the link isn’t merely a byproduct of lifestyle or social circumstance.
When the researchers drilled into specific diagnoses, the numbers became striking in their own right. The forward risk from PMD to ADHD showed a hazard ratio of 3.55, the highest in the dataset. Bipolar disorder came in at 3.36. Personality disorders at 3.34. Anxiety disorders at 2.43. These are not trivial elevations. Women with ADHD and autism are often diagnosed late, partly because the conditions present differently in women than in the male-dominated study populations that originally defined diagnostic criteria, and the researchers note this probably inflates the apparent risk in the PMD-to-ADHD direction: a PMD diagnosis brings women into more frequent clinical contact, which in turn increases the chance that an existing but undetected neurodevelopmental condition gets picked up. So some portion of that 3.55 is, in a way, diagnostic access rather than new disease. Still. The underlying connection almost certainly isn’t an artefact.
The one conspicuous exception was schizophrenia, which showed no association in either direction.
Researchers offer a few possible explanations, none of them definitive. Diagnostic overshadowing is one: schizophrenia’s severe symptoms may simply eclipse other clinical concerns, meaning PMD goes undetected in women who have it. Antipsychotic medications also frequently disrupt menstruation, which could mask or eliminate the cyclic symptom pattern that defines PMD in the first place. The genetic overlap between schizophrenia and PMD that some earlier work has proposed apparently doesn’t translate into the clinical record, at least not in these data.
For the conditions that did show the bidirectional pattern, the proposed biological scaffold involves several overlapping systems. The hypothalamic-pituitary-adrenal (HPA) axis, the body’s core stress-response machinery, is dysregulated in both PMD and mood disorders. Oestrogen and progesterone modulate serotonin, dopamine, and GABA signalling, all neurotransmitter systems already implicated in depression, anxiety, and bipolar disorder. There’s also a genetic component: twin and family studies estimate the heritability of PMD at somewhere between 35 and 56 percent, and earlier work has found genetic overlap between PMD and major psychiatric conditions including depression and ADHD. The picture is of systems that are not fully separable, where vulnerability in one tends to predict vulnerability in others.
Rethinking the Menstrual Cycle in Mental Health
The clinical implications are perhaps more immediate than the mechanistic ones. Psychiatry has historically been organised around conditions rather than the patients who have them, which tends to leave the menstrual cycle somewhere outside the frame. The study authors argue this needs to change. A woman presenting with depression should routinely be asked about her cycle; a woman presenting with severe premenstrual symptoms should be assessed for psychiatric comorbidity. What looks like sequential bad luck, one condition following another, may in many cases be the same underlying biology expressing itself in different registers at different times.
What the study cannot do is tell us why some women develop both and others don’t, or which comes first in any individual case. But the bidirectional pattern itself is information. It means the menstrual cycle isn’t a footnote in women’s mental health. It might be rather more central than medicine has tended to assume.
Frequently Asked Questions
If I have PMDD, does that mean I’m going to develop a psychiatric condition?
Not necessarily, but the risk is meaningfully elevated compared with women who don’t have PMD. The Swedish study found that roughly 37 percent of women with a PMD diagnosis went on to develop a psychiatric condition over an average of 8 years, versus about 21 percent of unaffected women. That’s a real difference, not a small one, but it also means the majority of women with PMD did not develop a subsequent psychiatric diagnosis during follow-up. The takeaway is probably less about inevitability and more about the value of monitoring: women with PMD may benefit from routine screening for anxiety, depression, and ADHD.
Why would a psychiatric condition make premenstrual symptoms worse?
The connection runs through shared biology rather than one condition causing the other in any simple sense. Both PMD and many psychiatric disorders involve dysregulation of the same underlying systems, including the brain’s stress-response axis and neurotransmitters like serotonin, dopamine, and GABA. Oestrogen and progesterone fluctuations across the menstrual cycle directly modulate these systems, so a brain that is already sensitised by depression or anxiety may respond more intensely to normal hormonal shifts. It’s less a case of one condition triggering another and more that they share a common vulnerability.
Is the link between PMD and ADHD real, or just because women get diagnosed more once they see doctors more?
Probably both, in varying proportions. The researchers themselves acknowledge that increased clinical contact following a PMD diagnosis likely accounts for some of the elevated ADHD detection, since women with ADHD are often diagnosed late and a PMD diagnosis can bring them into the healthcare system more regularly. But the association also survived comparisons between sisters with and without PMD, which controls for many confounders. The underlying biological connection, involving dopamine signalling and sex hormone effects on neurodevelopment, is supported by independent evidence and is unlikely to be entirely artefactual.
Why does schizophrenia appear to be the exception?
The study found no association between PMD and schizophrenia in either direction, which surprised the researchers given some earlier evidence of genetic overlap between the conditions. One plausible explanation is diagnostic overshadowing: schizophrenia’s severe symptoms may mean that premenstrual symptoms simply don’t get noticed or recorded. Another is that antipsychotic medications frequently disrupt menstruation, which could effectively mask the cyclic symptom pattern that defines PMD. Whether a real biological dissociation exists, or whether the null finding is a measurement problem, remains an open question.
ScienceBlog.com has no paywalls, no sponsored content, and no agenda beyond getting the science right. Every story here is written to inform, not to impress an advertiser or push a point of view.
Good science journalism takes time — reading the papers, checking the claims, finding researchers who can put findings in context. We do that work because we think it matters.
If you find this site useful, consider supporting it with a donation. Even a few dollars a month helps keep the coverage independent and free for everyone.