Menopause Is a Major Health Event. Medicine Barely Records It.

There is a familiar sentence in women’s health: we need more research.

We do.

But a large analysis from the NIH All of Us Research Program points to a problem that comes even earlier. Before researchers can study an event properly, healthcare systems have to record that it happened.

Menopause is one of the most consequential physiological transitions in a woman’s life. It can influence symptoms, bone health, cardiometabolic risk and the interpretation of health changes that occur across midlife.

Yet in a dataset of roughly 396,000 female participants, menopause appeared in structured electronic health records far less often than women reported it themselves.

Researchers found about 193,000 menopause observations in survey responses. Structured electronic health records contained only about 28,000 menopause diagnoses.

Nearly seven times as many menopause events appeared when women were directly asked.

“What we fail to measure becomes harder to understand.”

The study, published in the journal Menopause, compared survey responses, electronic health record diagnoses and genomic data within the All of Us cohort.

Only around 7 percent of women in the dataset had menopause documented in their electronic health record. Among those who reported menopause in survey data, only a minority had the corresponding structured EHR diagnosis.

The direction of the mismatch was striking. More than 99 percent of women who did have an EHR menopause diagnosis also reported menopause in the survey. The problem was not that medical records were routinely inventing menopause that women did not report. The problem was that the record often contained nothing.

Age at menopause, another variable that can matter to future health research, was also frequently unavailable.

This is not simply an administrative inconvenience.

Large health datasets are increasingly used to identify risk patterns, build predictive models, study disease and inform clinical decisions. If a major female health transition is missing from the record, the research built on that record inherits the absence.

Menopause timing matters.

Earlier or later menopause can be associated with different long-term health patterns, and researchers want to understand how menopause type, timing and symptoms intersect with cardiovascular, metabolic, bone and cognitive health.

But those questions become harder to answer when the underlying event is inconsistently documented.

The All of Us programme is specifically valuable because it combines electronic health records, surveys and genomic information across a very large and diverse population. The new analysis found about 22,000 participants with menopause-related information across EHR, survey and genomic data.

That is useful.

It is also a reminder of how much potential information disappears when routine clinical documentation is incomplete.

Women do not necessarily enter a clinic and announce, ‘Please code my menopause status for future population research.’

Healthcare systems have to decide what they routinely ask, what they structure, where the information lives and whether it survives transfers between clinicians and institutions.

Menopause can be discussed through symptoms rather than recorded as a transition. It may sit in free-text notes rather than structured fields. A clinician may focus on the immediate reason for the appointment. A patient may never be asked directly.

The study also found a notable concentration of EHR menopause codes after age 45 and a spike at age 65. That pattern is a useful warning that what appears in a health record can reflect healthcare contact and documentation practices as well as biology.

Data does not simply exist. Systems decide what becomes data.

We have spent years asking women to track more.

Track your cycle. Track your sleep. Track your symptoms. Track your hormones. Track your mood.

There is something slightly absurd about placing all of that responsibility on the individual while the formal medical record may fail to capture one of the largest reproductive transitions of her life.

Better women’s health research requires better trials, better funding and better representation.

It also requires something less glamorous: better records.

Medicine needs to get better at recording the events that shape women’s health. What we fail to measure becomes harder to understand.