How to write screening criteria for menopause studies

Menopause studies fail at the screener more often than at any other point. The reason is that there is no single test to point at, so the screener has to reconstruct the stage from what a woman can observe about her own body.

Key points

• Do not rely on participants to label their own menopause stage.
• Ask about observable cycle patterns, symptoms and treatment.
• Pilot the screener before recruitment begins.

Do not ask women to self-classify

The most common mistake is a question like "Are you perimenopausal, menopausal or postmenopausal?" Women use these words differently from researchers. Plenty of women who meet every criterion for perimenopause will answer no, because nobody has ever told them that is what is happening.

Ask about observable facts instead, and do the classification yourself.

The questions that actually work

Periods. Ask when her last period was, and whether her cycle length has changed by a week or more in the last year. That pair separates most of the stages on its own. Twelve consecutive months without a period, where there is no other cause, is the usual clinical definition of postmenopause. [1]

Age. Useful as a cross-check, not as a criterion. Menopause before age 45 is considered early, while premature ovarian insufficiency refers to loss of ovarian function before age 40. [2]

Surgery and treatment. Ask directly whether she has had her ovaries or womb removed, and whether she has had chemotherapy or radiotherapy affecting the ovaries. Surgical and medical menopause behave differently and often need their own arm.

Hormonal contraception and hormonal coils. These mask the bleeding pattern your staging depends on. If you do not ask, you will misclassify people.

HRT. Ask whether she currently uses it, what form it takes, and roughly how long she has been on it. If your study is about symptoms, current HRT use changes the answer entirely.

Symptoms. Use a validated instrument if you plan to publish. If you are running commercial insight work, a plain checklist is fine, but keep the list broad. NICE guidance recognises vasomotor, musculoskeletal, mood, sleep, cognitive and genitourinary symptoms associated with menopause. [1]

Common mistakes

• Screening on blood tests that most participants have never had.
• Using a hard age band that quietly excludes early menopause.
• Forgetting that hysterectomy without oophorectomy removes the bleeding signal but not the hormonal timeline.
• Writing an "other" box and then having no plan for what to do with the answers.

A worked screener

1. What was the date of the start of your last period? (Date, or "more than twelve months ago", or "I do not have periods")
2. If you still have periods, has your cycle length changed by a week or more in the last twelve months? (Yes / No / Not sure)
3. Do you use hormonal contraception or a hormonal coil? (Yes / No)
4. Have you had your womb or ovaries removed? (Womb only / Ovaries / Both / Neither)
5. Do you currently take HRT? (Yes, and what form / No / I used to)

Five questions, all answerable without notes, and enough to stage almost everyone.

Read the full picture in the pillar guide, Recruiting for women's health research, or see how much to pay research participants in the UK.

References

1. National Institute for Health and Care Excellence. Menopause: identification and management. NICE guideline NG23.
2. NHS. Early or premature menopause.