A practical guide for femtech and digital health teams, clinics and laboratories, pharma-supported programmes, and insurers and employer health programmes on the data decisions behind a condition that is common, often missed and newly renamed.
Polycystic ovary syndrome (PCOS) is one of the most common hormonal conditions in women, and one of the most often missed. The World Health Organization (WHO) estimates that it affects 10% to 13% of women of reproductive age, and that up to 70% of those affected worldwide do not know they have it.1 It shapes far more than fertility: it is linked to irregular cycles, excess androgens, metabolic risk and mental health, often over decades.
It also has a new name. In May 2026, after a process involving more than 50 organisations, PCOS was formally renamed polyendocrine metabolic ovarian syndrome (PMOS), with a three-year transition leading to full adoption in the 2028 international guideline update.2 The new name reflects what clinicians and researchers have long argued: this is a condition of hormones and metabolism, not only of the ovaries. We use "PCOS" in this guide because it is still the name most people search for and most records carry, but the change itself is one of the data points that teams now need to plan for.
Source: World Health Organization, polycystic ovary syndrome fact sheet; Endocrine Society announcement of the PMOS name (May 2026).
This guide draws on the 2023 International Evidence-based Guideline for the Assessment and Management of PCOS, led by Monash University and endorsed internationally, which has since been updated to use the new name.3, 4 We wrote it for four groups of readers. If you are building a femtech or digital health product, it sets out the data decisions to get right before you scale. If you run a clinic or laboratory, it covers what it takes to make results comparable and complete. If you lead a pharma-supported programme, it explains what your data needs in order to show outcomes. If you are an insurer or run an employer health programme, it shows how definitions and data quality shape who you reach and what you can demonstrate.
A woman with PCOS may track her cycles in one app, receive hormone and glucose results through a laboratory portal, and hold ultrasound reports as PDFs. Her prescriptions (often including hormonal contraception, which changes both her cycle data and her hormone results) come from more than one pharmacy. Mental health questionnaires, if she has been offered them, sit in yet another system, and her gynaecologist, endocrinologist, fertility specialist and general practitioner (GP) each keep their own notes.
Each source holds part of the story. The value comes from bringing them together in a way that stays accurate over years of care.
The twelve points below are grouped into four themes: getting the clinical picture right, making the data usable, connecting the longer story, and building for trust and scale.
In adults, the 2023 international guideline diagnoses PCOS when two of three features are present: irregular cycles or ovulatory dysfunction, clinical or biochemical hyperandrogenism (excess androgen hormones), and polycystic ovarian morphology on ultrasound or a raised anti-Müllerian hormone (AMH) level. Other conditions must be excluded first. Adolescents are assessed differently: they need both irregular cycles and hyperandrogenism, and ultrasound and AMH are not recommended within eight years of the first period.4
Based on the 2023 International Evidence-based Guideline for the Assessment and Management of PCOS.
Older definitions are still in use, and they change who counts. In a large meta-analysis, prevalence was 6% using the US National Institutes of Health (NIH) 1990 criteria, and 10% using the Rotterdam 2003 or Androgen Excess and PCOS Society criteria.5 Records will also carry both the old and new names for years, so systems need to recognise "PCOS" and "PMOS" as the same diagnosis.
What good looks like: the criteria you apply written down and recorded with each classification, adult and adolescent logic kept separate, and both disease names mapped to the same coded diagnosis.
Matters most to: Startups, Insurers and employers, Pharma programmes
"Irregular" has a precise meaning in the guideline, and it depends on time since the first period. Irregular cycles are normal in the first year. From one to three years after the first period, cycles shorter than 21 days or longer than 45 days are irregular. From three years onwards until perimenopause, cycles shorter than 21 days or longer than 35 days, or fewer than eight cycles a year, are irregular. At any point more than a year after the first period, a single cycle longer than 90 days counts.3, 4
Cycle tracking apps collect exactly the data needed to apply these rules, but many use their own thresholds, and most cannot tell whether hormonal contraception is shaping the pattern. A woman on the combined pill has withdrawal bleeds, not natural cycles, and her data should not be read as if she did.
What good looks like: cycle data stored with dates rather than summary labels, the guideline definitions applied by age at first period, and hormonal contraception recorded so affected cycles are excluded or flagged.
Matters most to: Startups, Clinics and laboratories
Several other conditions can look like PCOS. The guideline lists tests to exclude them, including thyroid-stimulating hormone (TSH), prolactin, 17-hydroxyprogesterone (for non-classic congenital adrenal hyperplasia) and follicle-stimulating hormone (FSH), with further tests for Cushing's syndrome and adrenal tumours where indicated.4 A PCOS label without these exclusions recorded is weaker evidence, for clinicians and for research.
What good looks like: exclusion tests stored alongside the diagnosis with their dates and results, so anyone reading the record can see how the diagnosis was reached.
Matters most to: Clinics and laboratories, Pharma programmes
Biochemical hyperandrogenism is assessed with total and free testosterone. The guideline recommends that laboratories measure total testosterone using liquid chromatography tandem mass spectrometry (LC-MS/MS) rather than direct immunoassays, and that free testosterone is calculated (for example as the free androgen index) or measured by equilibrium dialysis. If testosterone is not raised, androstenedione and dehydroepiandrosterone sulfate (DHEAS) may be considered. Because the combined pill changes androgen levels, it should be stopped for at least three months before testing.3, 4
This means an androgen result without its method, reference range and contraception status can be misleading. Units also vary between laboratories and countries, adding another layer when results are combined.
What good looks like: every androgen result stored with its assay method, units, reference range and the laboratory that produced it, with hormonal contraception at the time of testing recorded.
Matters most to: Clinics and laboratories, Startups, Pharma programmes
Polycystic ovarian morphology can be shown on ultrasound or through a raised AMH level, but the guideline advises using one or the other, not both, to limit over-diagnosis.4 On transvaginal ultrasound, the threshold in adults is 20 or more follicles in at least one ovary, with an ovarian volume of 10 mL or more as an alternative, and reports should record the transducer's frequency because image quality affects the count.4, 6 AMH cut-offs are specific to the population and the assay used,4 so a single universal threshold in an app or dashboard can misclassify people.
Ultrasound findings usually arrive as free-text reports, which is where much of this detail gets lost.
What good looks like: ultrasound reports turned into structured findings (follicle count, ovarian volume, probe type) with the original kept for reference, and AMH stored with its assay and the cut-off that was applied.
Matters most to: Clinics and laboratories, Startups
Clinical hyperandrogenism is often assessed with the modified Ferriman-Gallwey score for hirsutism, where the guideline sets a threshold of 4 to 6 depending on ethnicity.3 Acne, hair loss and other symptoms are recorded in many different ways, from clinician scores to self-reports in apps. A single universal threshold, or symptoms captured only as free text, makes results hard to compare across people and over time.
What good looks like: symptom scores stored with the scale used, who recorded them (clinician or self-report) and the threshold applied, with ethnicity-appropriate thresholds where the guideline allows.
Matters most to: Startups, Pharma programmes
PCOS carries a higher risk of type 2 diabetes, gestational diabetes and impaired glucose tolerance. The guideline recommends assessing glycaemic status at diagnosis in all adults and adolescents and then every one to three years, and notes that the 75 g oral glucose tolerance test (OGTT) is the most accurate test regardless of body mass index. It also recommends a lipid profile at diagnosis and blood pressure checks every year.3, 4
These results often sit with a different clinician from the one managing cycles or fertility, which is why they are so easily lost. Our diabetes and hypertension guides cover the data issues in more depth.
What good looks like: glucose, HbA1c (glycated haemoglobin), lipids and blood pressure held in the same longitudinal record as cycle and hormone data, with reminders when metabolic reviews are due.
Matters most to: Pharma programmes, Insurers and employers, Clinics and laboratories
The guideline recommends screening all adults and adolescents with PCOS for depression, and all adults for anxiety, using regionally validated tools. It also asks clinicians to consider eating disorders and disordered eating regardless of weight, and recognises that features of PCOS can affect body image.4
For a product, this is both important and delicate. Screening results are highly sensitive, the tools used differ by country, and weight-focused features can do harm if they are not designed with this in mind.
What good looks like: validated screening tools recorded by name and version, results protected with stricter access than other data, clear pathways to support when a screen is positive, and weight features designed with clinicians.
Matters most to: Pharma programmes, Insurers and employers, Startups
The guideline recommends the combined oral contraceptive pill as first-line medicine for irregular cycles and hyperandrogenism, metformin primarily for metabolic features, and letrozole as first-line medicine for fertility treatment. Inositol offers limited clinical benefit, and anti-obesity medicines may be considered in line with general population guidance.3
Because the pill affects both cycle data and androgen results, its start and stop dates are essential context for almost every other data point in the record.
What good looks like: each medication recorded with active ingredients, dose and start and stop dates, mapped to medical coding standards, and linked to the cycle and laboratory data it affects.
Matters most to: Clinics and laboratories, Pharma programmes
A useful summary brings together cycle patterns against the guideline definitions, androgen and ultrasound or AMH findings with their methods, the exclusion tests, metabolic results, current treatment and screening outcomes. Some findings need a clear route to action rather than simply being logged. Rapidly worsening or severe androgen symptoms can point to other causes the guideline says should be excluded, such as adrenal tumours,4 and screening results that suggest depression, anxiety or disordered eating need a defined pathway to care.
What good looks like: an exportable clinician summary in a consistent format, escalation rules agreed with clinicians and tested, and a record that can move into a clinic's own system.
Matters most to: Clinics and laboratories, Startups
Software intended for a medical purpose, without being part of a hardware device, is described internationally as Software as a Medical Device.8 Many PCOS products begin as cycle tracking or coaching tools, and cross into this territory when they suggest a diagnosis, predict an individual's risk or recommend treatment. Regulators in Singapore, the UK, the European Union, Australia, the United States and many other markets each apply their own rules.
What good looks like: an intended purpose written down early, claims reviewed against the rules in each market you serve, and validation evidence gathered from the start rather than retrofitted.
Matters most to: Startups, Pharma programmes
Cycle, fertility and hormone data is among the most sensitive health information a person can share. A 2025 review of reproductive health apps found that many collect identifiable information and share data with third parties for advertising, and concluded that users' privacy expectations are not being met.7 This data is protected under laws such as the EU General Data Protection Regulation (GDPR), Singapore's Personal Data Protection Act (PDPA) and, where it applies, the Health Insurance Portability and Accountability Act (HIPAA) in the United States, and people need confidence that it will not be used against them by insurers or employers.
PCOS is also long-term, while products change. Structuring data to interoperability standards such as FHIR (Fast Healthcare Interoperability Resources), with results and medications mapped to medical coding standards, keeps it useful when people change apps, clinics or countries.
What good looks like: granular consent recorded alongside the data, no advertising use of health data, pseudonymisation for secondary use, clear separation from employment or underwriting decisions, and standard-format export with provenance intact.
Matters most to: Startups, Insurers and employers, Pharma programmes, Clinics and laboratories
You do not need to solve all twelve at once. The right starting point depends on what you are building and what your data needs to do.
Some of the twelve points are data problems, and some are questions of design, strategy and delivery. We support both, so you can come to us with the whole picture rather than splitting it across several partners.
Jonda Health is ISO 27001 certified, and our platform is designed to comply with HIPAA, GDPR and PDPA.
This guide is for teams designing and running digital health tools and programmes. It is not clinical advice for individuals.
