Information for Clinicians
Stabilize Health · Last updated 27 August 2026
Your patient may bring you a summary produced by Stabilize Health. This page explains what is in it, how each figure is derived, and what it deliberately does not claim.
In one line: everything in the report is patient-reported, entered by hand, and summarised arithmetically. Nothing in it is measured by a sensor, inferred by a model, or validated against a clinical instrument.
What the report contains
- Symptoms — per day, each rated by the patient on a 0–3 scale (none / mild / moderate / severe). Symptom names come from a built-in library or are free text the patient added.
- Energy and pacing — a daily 0–10 self-rating, an optional self-reported "spoon" count, and a patient-marked flare day flag.
- Context tags — a fixed set of self-selected daily tags (for example poor sleep, stress, overexertion, weather change, skipped meal, new food, alcohol).
- Treatments and adherence — medications and supplements as the patient entered them, with their schedules, and doses logged as taken, missed, or skipped.
- Laboratory values — transcribed by the patient, with the reference interval they entered from their own laboratory report. Not imported from any laboratory system.
- Measurements — weight, waist, blood pressure, glucose, sleep duration, where the patient chose to record them.
- Cycle entries — only if the patient enabled the optional cycle module.
- Free-text notes — if the patient chose to include them.
The patient chooses the date range and which sections to include, so a report may be partial by design. A CSV of the raw entries can be produced alongside the PDF if you would prefer the underlying rows.
How the derived figures are calculated
| Figure | What it is | What it is not |
|---|---|---|
| Pattern Map | A layout of the patient's own 0–3 ratings across seven areas of daily life, taken from a self-report questionnaire in the app. | Not a diagnosis, not a screening tool, and not derived from any published instrument. Labelled in-app as a tracking profile. |
| Symptom Burden Score (0–100) | An arithmetic rescaling of those same 0–3 answers into a single number. The formula is displayed inside the app. | Not a disease severity score. Not comparable to any validated severity measure, and not comparable between patients. |
| Doctor Readiness | A completeness measure of the record itself — how many days were logged, whether treatments and labs were entered. | Not a health measure of any kind. A high score means good record-keeping, nothing more. |
| Co-occurrence descriptions | Statements that two things appear in the log around the same time, phrased associatively. | Not a causal claim, not a trigger identification, and not statistically tested. |
| Cycle summaries | Descriptive statistics over cycles the patient actually recorded: shortest, longest, typical, and range. | Not a prediction. The app does not forecast ovulation, fertile windows, or next period, and suppresses cycle interpretation where the patient reports pregnancy, hormonal contraception, or absence of menstrual bleeding. |
What the app never does
- It does not diagnose, and never tells a patient they do or do not have a condition. Conditions shown in the report are patient-declared and only set tracking defaults.
- It does not interpret laboratory values, flag them, or comment on them.
- It does not recommend starting, stopping, changing, or dosing any medication or supplement, and never suggests a dose.
- It does not predict flares or disease course.
- It contains no AI assistant and no generated clinical advice.
- It is not a medical device and has not been reviewed or cleared by any regulator.
Reliability caveats worth knowing
- Recall and adherence bias. Entries are made when the patient remembers. Gaps are common and are not marked as missing data in every view.
- Self-anchored scales. The 0–3 and 0–10 scales are anchored to the individual, not to a population norm. Trends within one patient are more meaningful than absolute values.
- Transcription. Laboratory values and reference ranges are typed in by hand and are not verified against a source system.
- Selective reporting. The patient chooses the range and sections, so absence from the report does not mean absence from their record.
Privacy
The patient's data is stored on their own device. There is no account and no server-side copy, and we do not receive it. A report exists only when the patient generates and shares it. Details are in the Privacy Policy.