27% With OCD Recall Worsening at Menopause in 8 Small Studies

A woman who has kept her checking rituals to a manageable half hour for twenty years notices, somewhere in her forties, that the half hour is stretching. Her sleep has gone strange, her cycles have gone irregular, and the intrusive thoughts have acquired a volume she thought she had negotiated away. Whether those things are related is the sort of question that has historically been answered in a clinic room by a shrug, because the literature to answer it with did not obviously exist.

Carmela Melina Albanese, Gabriella Antaya and Jennifer L. Gordon set out to find whether it did. Their scoping review, published online in November 2025 and carried in the March 2026 issue of Menopause, searched MEDLINE, Embase, PsycINFO and CINAHL from inception through April 15, 2025, plus conference proceedings and dissertation databases, for anything examining OCD in relation to menopause. The yield was eight studies — seven full papers and one conference abstract — from the United States, the Netherlands, Turkey, Mexico, India, South Africa and Spain. All eight were cross-sectional. Seven asked women who already carried an investigator-confirmed DSM-IV diagnosis of OCD to recall how their symptoms had behaved around menopause; one compared premenopausal and postmenopausal groups. Sample ages varied by study — one South African sample spanned 18 to 75, another Mexican comparison was restricted to 45 to 55. The two studies that reported a mean age at menopause put it at 46.5 and 46.9.

Of 265 participants across the studies that asked about symptom change, 72 — 27.2% — reported that their OCD got worse with the onset of menopause. That is the figure the headlines took. Two others belong beside it. Thirty participants, 11.3%, reported that their symptoms improved. And in the three studies that reported a no-change category, it was often the largest group: 46.7% in Guglielmi and colleagues' 171 participants with pre-existing OCD (their full postmenopausal sample was 187, but only 171 had OCD before menopause), 47.4% in Vulink and colleagues' sample of 19. In the third, a Turkish sample of 17, the largest group instead reported improvement — 47.1%, against 23.5% who reported no change.

The spread across studies is wide enough to be its own finding. Worsening ranged from 2.2% in a Turkish sample of 46 to 47.4% in that Dutch sample of 19. Improvement ranged from zero to 47.1%. On new onset, 17 of 373 participants — 4.6% — said their OCD symptoms first appeared as menopause began, with study-specific rates from 2.2% to 27.8%. Separately, one Indian study of 92 women found a single case (1.08%) in which menopause occurred in the year before OCD onset — a different construct, and a reminder of how differently these eight studies asked their question. The single study that compared groups rather than memories, by Carranza-Lira and Pablo-Cruz, found higher median obsessive-compulsive scores among 164 postmenopausal women than among 202 premenopausal women, all aged 45 to 55 (P<0.001) — a significance value with no effect size attached to it.

Two structural things about those percentages deserve stating plainly. The denominators differ — 373 for onset, 265 for change — because different studies asked different questions, so these are not two views of one cohort. And a scoping review counts; it does not pool. The authors did not weight studies by size, model heterogeneity, or formally grade risk of bias — by design; the authors note critical appraisal is not required in a scoping review and state plainly that they did not perform one. Which means 27.2% is an arithmetic summary of a very uneven pile, dominated by the single study contributing 171 of those 265 participants.

The mechanistic story on offer is that estradiol and progesterone modulate neurotransmission in the serotonergic, dopaminergic and glutamatergic systems implicated in OCD, and that they may ease symptoms by enhancing serotonin signaling. Menopause removes them. The review describes the worsening finding as consistent with a neuroprotective effect of those hormones, and notes that administering estradiol has been shown to reduce compulsive behaviors in rats.

The same body of work, however, supplies the opposite prediction. OCD symptoms have long been observed to spike in the perimenstrual phase, when hormones swing hardest — which implies that ending the swings could settle things down for some women. The review offers this as the reading of the 11% who improved. A hypothesis flexible enough to predict both worsening and improvement is not thereby wrong, but it is not being tested here either. The decisive detail is one the authors state outright: none of the eight studies measured sex hormone levels in relation to obsessive-compulsive symptoms. The mechanism is imported from adjacent literature, not demonstrated by anything in this review.

The review's most important sentence is a negative one. Symptom changes during the menopause transition were not assessed in any included study — and the transition, as the authors point out, is the window with the most erratic ovarian hormone patterns, which is to say the window where the hormonal hypothesis would predict the most turbulence. What the studies captured instead was women on the far side of menopause, asked to remember. That is retrospective self-report, and the authors name recall bias as the consequence.

The gaps compound. Menopause status was self-reported throughout; five studies defined it as at least twelve months without menses and three offered no clear definition. No study examined age at natural menopause, whether menopause was natural or surgical, reproductive stage, or vasomotor symptoms. None accounted for hormone therapy, which stabilizes hormone levels and could plausibly move symptoms in either direction. Medications and comorbid mood and anxiety disorders were rarely reported in enough detail to adjust for. Differences by race, ethnicity or cultural subgroup were not considered anywhere in the prior literature, and the search was limited to English.

The seven retrospective studies had no comparison group — no similarly aged women who had not gone through menopause, asked the same recall questions. Only Carranza-Lira and Pablo-Cruz compared groups, and that study measured symptom scores at a point in time rather than remembered change. Midlife delivers disrupted sleep, aging parents, and its own elevated rate of depression, all of which can inflame OCD without any help from estradiol. Nothing in this design separates those from the hormonal account. And because every participant already had OCD, 27.2% is the share of already-diagnosed women who remembered getting worse — not the odds that menopause will produce OCD in someone who does not have it.

The honest summary is that this review found less evidence than the question deserves and reported it carefully, which is a more useful contribution than it sounds. Before it, a clinician asked whether menopause worsens OCD had anecdote. After it, that clinician has a defensible order of magnitude — roughly a quarter of patients report worsening, roughly one in ten report the reverse, and in most of the studies that asked, the largest group reported neither — plus a precise map of which studies would need to exist for the number to firm up.

The authors' own recommendations are correspondingly modest: increased symptom monitoring of midlife patients with OCD, an acknowledgment that changes in treatment intensity or dosage may be warranted for some women as menopause begins, and a call for prospective longitudinal studies that track symptoms across premenopause, perimenopause and postmenopause using objective measures of reproductive staging. The authors note first-line OCD treatment remains cognitive behavior therapy and SSRIs, and that nothing in this review alters that. What changes is whether a patient who says her rituals got worse at 49 is describing something the literature recognizes.

There is something clarifying about a review whose most quotable finding is a list of things nobody has measured. The 27.2% is real, in the narrow sense that 72 women out of 265 said it; it is also a count of memories, gathered after the fact, from eight small cross-sectional studies with no hormone assays, almost no comparison groups, and no observation of the transition itself. Coverage that reduced this to menopause worsening OCD skipped the 11.3% who improved and the large no-change group — roughly half in two of the three studies that measured it, under a quarter in the third — and skipped the word the authors leaned on hardest, which is scoping: a survey of what exists, not a verdict on what is true. A careful reader should come away with a plausible association worth watching in a patient who already has OCD, an untested but coherent estradiol hypothesis, and no basis at all for telling a woman what menopause will do to her symptoms. The authors frame the next step as prospective studies, and they are right that the field does not yet have one.