Marram

A private companion through perimenopause and menopause.

Everything you write stays on your phone. No account. No server.

The name

Marram grass grows up through moving sand. Its roots hold the dune together while the ground shifts underneath.

Question prompt list

Six questions, one side of one page

A question prompt list is a studied intervention with systematic reviews behind it, not a nice idea: a short written set of questions the patient hands over at the start of the consultation. Ten minutes with a GP goes quickly, and it is common to come out having forgotten the thing you went in for.

They have been tested in randomised trials. They increase how much a patient takes part in the decision, and how many questions she asks. One result matters more than the rest: the list increased the number of questions patients asked without increasing consultation time. More asked, same appointment.

Where they fail is length. Across the published lists that have been reviewed, the number of questions had a mean of 51.0 and a median of 39.0, ranging from 9 to 191. More than half ran to more than one page. Nobody hands over four pages.

What works is short. Five to eight questions on one side of one page, a line under each to write the answer on, and an instruction at the top saying when to hand it over — the version carrying the instruction did better than the list on its own. This one is six. That is the finding, not modesty.

Question prompt list · perimenopause and menopause

Hand this to your doctor at the start of the appointment, before you begin — not at the end.

Are these symptoms perimenopause, or is there something else we should rule out?
What are my options — including HRT, and including doing nothing for now?
Given my history, what are the risks for me specifically?
How long before I know whether it is working, and what should I watch for?
What can I do about the sleep and the brain fog?
If this does not help, what do we try next, and when should I come back?

Tick what matters most today. Nothing here is sent anywhere, nothing is saved, and closing this page forgets it.

Six questions, and the instruction line, follow two systematic reviews of question prompt lists: PMC12391282 and PMC10349246.

The app

Built on menopause-specific CBT

Thirty-four sessions across twelve weeks. Every session is sourced, and every session can be repeated.

The programme follows menopause-specific cognitive behavioural therapy — the approach the NICE menopause guideline, NG23, tells clinicians to consider … as an option for vasomotor symptoms associated with menopause. Marram is not that therapy and does not stand in for it. It is a self-guided programme built to its shape, to use alongside whatever you and your doctor decide.

“it is the extent to which the VMS are perceived as problematic, rather than their frequency, that is associated with quality of life (QOL) and help-seeking”
Cognitive behavioural therapy for menopausal symptoms, PMC8453849

That sentence decides the design. It is why nothing here counts your hot flushes, and why the sessions are about how a symptom lands rather than how often it arrives.

Where each piece comes from

Every session carries its sources. Forty-one citations across the thirty-four sessions, drawn from seven sources. Each citation holds a URL, a verbatim quote, and the date it was fetched. A build gate rejects any draft that cites something outside that closed set. The rule the corpus is written under is: a source you did not open is not a source.
The breathing runs at six cycles a minute. That is the pace the slow-paced breathing literature uses, against a spontaneous rate the same paper puts between 12 and 20. Six is a number off a page, not a feel.
Plans are written if-then. When this happens … I will … The implementation-intentions research found effect sizes were larger when plans had a contingent if-then format, and when they were rehearsed.
Pacing, and reducing avoidance. Not general advice — both are named as helpful behavioural strategies in the CBT review quoted above.
Every session stays repeatable, because this is not a short thing. The NHS puts the span of symptoms at seven to nine years, sometimes longer, and says they change as they go. Twelve weeks is where the programme is written to, not a claim about how long any of this lasts.

The sessions are made of eight kinds of thing: breathe, plan, log, note, voice note, pick, review, and a dossier you can export and take to an appointment. Any of them can be repeated as many times as you like, and going slowly is fine.

Nothing is scored. Nothing is tracked. There are no streaks.

That is a position, not a gap. A programme that scores you becomes one more thing to be behind on, and the paper above is explicit that what tracks with quality of life is not the count.

The seven sources the programme is written from

NICE NG23 — Menopause: identification and management. Consider menopause-specific cognitive behavioural therapy (CBT) as an option for vasomotor symptoms associated with menopause nice.org.uk/guidance/ng23
Cognitive behavioural therapy for menopausal symptoms. Encouraging helpful behavioural strategies, for example, reducing avoidance, pacing activities PMC8453849
Slow-paced breathing, Psychophysiology. SPB is usually realized at a pace of around six cycles per minute (cpm), while the spontaneous breathing frequency generally comprises between 12 and [20] doi.org/10.1111/psyp.13952
Implementation intentions — University of Konstanz repository. Effect sizes were larger when plans had a contingent (if-then) format, participants were highly motivated to pursue the goal, and plans were rehearsed kops.uni-konstanz.de
NHS — Menopause and perimenopause: symptoms. Symptoms usually last for 7 to 9 years, sometimes longer, and during this time they can change nhs.uk
NHS — Menopause and perimenopause: things you can do. The behavioural strategies the programme practises are the ones named here. nhs.uk
NHS — Menopause and perimenopause: treatment. Including that CBT is among what a doctor may offer. nhs.uk

The two reviews behind the question prompt list are further up the page. The forty-one citations themselves sit inside the sessions, where the claims are.

Privacy

It does not leave the phone

This page holds itself to the same rule. It loads no fonts, scripts, styles or images from anywhere else, has no analytics, and sets no cookies.

No account, and no server. There is nothing to sign up for and nothing to upload.
Everything is written to an encrypted database on the device. The key lives in the platform keystore — Keychain on iOS, Keystore on Android.
The Android build removes the internet permission entirely. The app cannot open a network connection, even in error.
No analytics, no crash reporting, no advertising identifier, no third-party SDK.
Voice notes are encrypted at rest. Transcription runs on the device, and where a phone cannot do it, transcription is turned off rather than sending your voice away.
Marram opts out of iCloud Backup, and of Android cloud backup and device transfer. A new phone starts empty, on purpose.
Two things open outside the app, and you start both: tapping a crisis number opens your dialler, and the helpline directory opens in your browser.

If things are bad

Marram is not an emergency service. It offers reflection and gentle practices — not diagnosis, and not medical treatment.

You can name one person to call, and choose your region so the crisis line shown is a local one. Marram never substitutes another country's emergency number. A number that does not connect is worse than none.