Overview
Quick answer
If you are searching PCOS after irregular periods, acne, extra hair, thinning hair, or a scan that mentioned "cysts," the first job is to stop treating the name as a finding on an ovary. The 2023 international evidence-based guideline, led and primarily funded from Australia and approved by the NHMRC, treats PCOS as a chronic endocrine condition with insulin resistance and hyperandrogenism. It is not an isolated ovarian disease, and it does not involve true ovarian cysts.
The Australian MJA summary of that guideline (Teede et al., 2024) describes PCOS as affecting one in eight Australian women. The guideline process itself confirmed a global prevalence of 10% to 13% using international / Rotterdam-derived criteria. Both figures are epidemiology, not a diagnosis of you.
In adults, diagnosis needs two of three: clinical or biochemical hyperandrogenism, ovulatory dysfunction or irregular cycles, and polycystic ovarian morphology on ultrasound or a raised anti-Mullerian hormone (AMH). If you already have irregular cycles and hyperandrogenism (about 70% of cases), ultrasound and AMH are not required. Blood is a filter and a metabolic baseline. It is not the syndrome.
Overview
Key takeaways
PCOS is a misnomer. The follicles counted on ultrasound are not pathological cysts. This page keeps the name people still search.
Adult diagnosis is two of three features after other causes are excluded. Adolescents need both hyperandrogenism and ovulatory dysfunction. Ultrasound and AMH are not used in that age group.
Biochemical hyperandrogenism is total testosterone plus calculated free testosterone, preferably by LC-MS/MS. Direct free-testosterone immunoassays are not preferred. Androstenedione and DHEA-S are second-line and less specific.
You cannot read androgens reliably on combined hormonal contraception. The 2023 algorithm asks for a withdrawal of at least three months, with alternative contraception, before those bloods.
Insulin resistance is the metabolic core. Clinically available insulin assays are still not recommended for routine PCOS care. The guideline's glycaemic test is a 75 g oral glucose tolerance test (OGTT). Fasting glucose and HbA1c are second-line.
AMH can stand in for adult ultrasound morphology. Use one or the other, not both, or you over-diagnose. RCPA classifies AMH as non-MBS-rebatable.
A normal total testosterone does not rule PCOS out. Clinical hyperandrogenism counts on its own, and low SHBG can raise free testosterone while the total looks ordinary.
Blood cannot image an ovary, time fertility, or diagnose from a single flagged line. A membership panel is not a PCOS diagnosis.
The frame
PCOS is not ovarian cysts
Most people arrive here because a webpage, a scan report, or a relative used the word "cysts." That is the wrong picture, and it is why the condition is underdiagnosed and misunderstood.
A chronic endocrine condition, not an ovary problem
The MJA 2024 Australian perspective is explicit: PCOS is neither an isolated disease of the ovaries nor a condition of true ovarian cysts. It is a genetic and lifestyle-related chronic endocrine condition with insulin resistance and hyperandrogenism, and with reproductive, psychological, cardiometabolic, and dermatological features.
One in eight is the Australian line, not a headcount
Teede et al. in the MJA (2024) write that the guideline addresses a condition that affects one in eight Australian women and about 140 million women globally. The 2023 guideline process confirmed 10% to 13% globally. Indigenous Australians are noted in that paper as having higher prevalence and more limited access to care outside the cities. None of those figures is a blood-survey census of every Australian woman.
The 2026 rename, mentioned once
In May 2026 a global consensus, published in The Lancet and welcomed by RANZCOG, renamed PCOS to polyendocrine metabolic ovarian syndrome (PMOS). The new name is being phased in over three years and is due to land in the 2028 international guideline. People still search PCOS. This page uses that name.
If you came here hoping a blood test would confirm "cysts on my ovaries," this is the honest frame: blood measures androgens, the hormones that bind them, the pituitary lines that can mimic PCOS, and the metabolic risk that travels with the syndrome. Morphology, when it is needed at all, is ultrasound or AMH. Neither is a cyst count.
Australian practice
How PCOS is diagnosed in 2023
The 2023 international evidence-based guideline updated the old consensus Rotterdam criteria into an evidence-based algorithm. Australia led and primarily funded the work through the NHMRC Centre for Research Excellence in Women's Health in Reproductive Life. NHMRC approved the guideline in July 2023.
Adults: two of three, after exclusion
The required features are (i) clinical or biochemical hyperandrogenism, (ii) ovulatory dysfunction or irregular cycles, and (iii) polycystic ovarian morphology on ultrasound or a raised AMH. Other causes must be excluded first: TSH, prolactin, 17-hydroxyprogesterone, and FSH, plus anything the history actually suggests (Cushing's, an androgen-secreting tumour, hypogonadotrophic hypogonadism from low body fat or extreme exercise).
About 70% do not need a scan
If irregular cycles and hyperandrogenism are both present, ultrasound and AMH are not required. That is a practice point in the MJA summary and Algorithm 1. The scan is not the diagnosis. It is the third criterion, used when you only have one of the first two.
Adolescents are a different rule
WHO adolescence here is 10 to 19 years. Diagnosis needs both hyperandrogenism and ovulatory dysfunction. Ultrasound and AMH are not recommended, because multi-follicular ovaries and a swinging cycle are common in puberty and are not specific. One criterion only means "at risk of PCOS" and follow-up, not a label.
What "irregular" actually means
The first year after menarche is the pubertal transition and is not scored. From one to three years: cycles shorter than 21 or longer than 45 days. From three years after menarche to perimenopause: shorter than 21 or longer than 35 days, or fewer than eight cycles a year. Any cycle longer than 90 days after the first year, or primary amenorrhoea by 15, also counts. Ovulatory dysfunction can still exist with regular bleeding; that is a progesterone question, not a webpage call.
A historical diagnosis is considered enduring after menopause. New, severe, or worsening hyperandrogenism after menopause is a different problem and needs investigation for a tumour or ovarian hyperthecosis, not a recycled teenage label.
The panel
What blood tests actually do
Blood has three jobs in a PCOS workup: support or refute biochemical hyperandrogenism, exclude the conditions that look like PCOS, and set a metabolic baseline. It does not replace the history or the examination.
Androgens, when clinical signs are unclear
The 2023 algorithm uses total testosterone and free testosterone. Free testosterone should be calculated (or measured by equilibrium dialysis or ammonium sulfate precipitation), not read from a direct immunoassay. LC-MS/MS is the recommended assay. If testosterone is not raised, androstenedione and DHEA-S can be added. They are less specific and have a limited role in the diagnosis.
The exclusion set
TSH, prolactin, 17-hydroxyprogesterone, and FSH are there to catch thyroid disease, hyperprolactinaemia, non-classical congenital adrenal hyperplasia, and premature ovarian insufficiency. They are not "PCOS markers." A very high androgen, or a suddenly progressive story, is a reason to look for a tumour rather than to tick the PCOS box.
Glycaemic status, in every adult with the diagnosis
Impaired fasting glucose, impaired glucose tolerance, and type 2 diabetes are increased in PCOS independent of, and worsened by, body mass index. Assess glycaemic status at baseline, then every one to three years depending on other risk. The most accurate test is a 75 g OGTT, regardless of BMI. Fasting glucose and HbA1c are second-line. Offer an OGTT when planning pregnancy.
Lipids and blood pressure sit beside the hormones
The algorithm asks for a fasting lipid profile at diagnosis in every woman with PCOS, then follow-up guided by the result and global cardiovascular risk. Blood pressure is annual, and again when planning pregnancy. That is why this hub links the high-cholesterol guide rather than reprinting it.
LH and FSH are useful when you need to exclude hypogonadotrophic hypogonadism. An LH:FSH ratio above 2 is a pattern some older pages still treat as diagnostic. It is not a 2023 criterion.
Reading your results
How to read an Australian PCOS printout
Australian pathology reports print each hormone with the lab's own name, unit, and interval. Hormone intervals are method, age, and cycle-day dependent. Match the row and the printed interval, not a remembered US unit or a single webpage range.
"Testosterone" or "Total testosterone"
In nmol/L. RCPA (last reviewed 2 January 2024) says the interval is method, age, gender, and Tanner-stage dependent. A result inside the printed interval does not rule PCOS out. Clinical hyperandrogenism still counts, and a total in range with a low SHBG can still mean a high free fraction.
"Free testosterone" or "Calculated free testosterone"
Usually in pmol/L when it is calculated from total testosterone and SHBG. The guideline prefers calculation over a direct free-testosterone immunoassay. If the report says "free" without "calculated," ask which method was used.
"SHBG" or "Sex hormone binding globulin"
In nmol/L. RCPA: produced in the liver, binds testosterone and oestradiol. About 66% of testosterone is SHBG-bound in healthy women. Insulin resistance and PCOS lower SHBG, which raises the free fraction. Oral oestrogens raise it. Read testosterone with this row, not against it.
"AMH" or "Anti-Mullerian hormone"
In pmol/L on Australian reports. Made by small ovarian follicles. The 2023 guideline allows it as an adult stand-in for ultrasound morphology. There is no single national PCOS cut-off on the RCPA Manual. Use the lab's age-specific interval. A high AMH is not a fertility promise and is not used in adolescents.
"LH", "FSH", "Oestradiol", "Prolactin"
LH and FSH in IU/L, oestradiol in pmol/L, prolactin in mIU/L. These help exclude other causes and, for oestradiol and gonadotrophins, describe the cycle you were in. They do not diagnose PCOS on their own. Cycle-day matters for LH, FSH, and oestradiol. AMH does not need cycle timing.
"DHEA-S" and "17-OH progesterone"
DHEA-S in µmol/L. 17-hydroxyprogesterone is the non-classical CAH screen. Neither is a first-line PCOS diagnostic androgen. Androstenedione, when it appears, is the other second-line androgen. RCPA notes it is raised in about 60% of hirsutism workups and does not separate adrenal from ovarian sources.
A result "well above" the laboratory reference range is a reason to think about other causes, not to celebrate a more definite PCOS. History of onset and speed of change matters more than a single high line.
Mechanism
Insulin resistance is the mechanism
Insulin resistance is why PCOS is a metabolic condition that can show up as skin, hair, cycles, or later diabetes. The mechanism is settled enough to state. The clinic test for it is not a fasting insulin.
Hyperinsulinaemia drives ovarian androgens
Insulin acts on ovarian theca cells, alone and with luteinising hormone, to increase androgen synthesis through CYP17 (17-alpha-hydroxylase / 17,20-lyase). Theca cells from women with PCOS make more testosterone in response to insulin than control theca cells (Nestler et al., 1998; Munir et al., 2004). That steroidogenic pathway stays sensitive while the metabolic actions of insulin in muscle and fat are impaired. That split is called selective insulin resistance.
The same insulin lowers hepatic SHBG
Hyperinsulinaemia suppresses liver production of SHBG (Endocrine Society scientific statement, Dumesic et al., 2015). Less SHBG means a larger free-testosterone fraction from the same total. That is why a "normal testosterone" next to a low SHBG is still an androgen story, and why the calculated free value exists.
Androgens then worsen the insulin picture
Houston and Templeman (Journal of Endocrinology, 2025) describe a loop: androgens can increase insulin secretion and worsen insulin resistance. The sequence is not always insulin resistance first. The 2023 guideline still treats the metabolic features as independent of, and exacerbated by, excess weight. Lean PCOS is still PCOS.
The guideline still does not want a routine insulin assay
Practice points 1.9.12 and 3.1.10 say insulin resistance is a pathophysiological factor, and that clinically available insulin assays have limited clinical relevance and are not recommended in routine care. Assay methods vary. A single fasting insulin is not a diagnostic criterion. The test the guideline wants for glycaemia is the 75 g OGTT.
Hemexa still measures fasting insulin and HOMA-IR on the preventative panel, because they can move while HbA1c is still in the reference interval. That is a tracking choice, not a claim that the 2023 guideline has made fasting insulin a PCOS test. For ranges, cadence, and how to order those assays, use the fasting insulin and HOMA-IR guides.
For the assays themselves, see the fasting insulin guide, the HOMA-IR guide, and the HbA1c guide. For total versus calculated free testosterone, see the testosterone guide.
Next test
AMH versus ultrasound
The 2023 update that most Australian consumer pages still miss is this: in adults, serum AMH can define polycystic ovarian morphology instead of a scan.
One marker, not two morphology tests
Use AMH or ultrasound, not both. The algorithm is explicit: using both over-diagnoses. If you already meet the first two criteria, you need neither.
What the scan is actually counting
With a transducer whose frequency bandwidth includes 8 MHz, PCOM is 20 or more follicles per ovary and/or an ovarian volume of 10 ml or more on either ovary, avoiding corpora lutea, cysts, and dominant follicles. Those follicles are the source of AMH. They are not the "cysts" in the old name.
AMH is not Medicare-rebated
The RCPA Manual (Anti-Mullerian hormone, last reviewed 2 January 2024) classifies the immunoassay as non-MBS-rebatable and lists PCOS support as one application, alongside ovarian reserve and granulosa-cell tumour monitoring. Australian labs (including Healius brands such as QML) bill it as a private add-on. Ask the collection site for the current fee. Do not use a US ng/mL page against a pmol/L report.
AMH falls with the oral contraceptive pill, GnRH agonists, hypothalamic amenorrhoea, and high-dose biotin (RCPA). A low result on the pill is not a reserve diagnosis. A high result is not a baby count.
Timing
Adolescents and the pill
Two timing problems wreck otherwise careful bloods: puberty, and contraception.
Do not diagnose from a teenage scan
Multi-follicular ovaries are common in the pubertal transition. That is why ultrasound and AMH are excluded in adolescents. Delayed adult diagnosis often started as a missed adolescent one. The guideline's answer is follow-up of the "at risk" group, not an early morphology label.
The pill masks the androgen criterion
Reliable biochemical hyperandrogenism is not possible on hormonal contraception. Algorithm 1: consider withdrawal for at least three months, with alternative contraception. Combined oral contraceptives also suppress AMH. Stopping contraception is a clinical decision. This page is not asking you to come off it.
Androgen bloods still have a cycle day
When you are cycling and off hormonal contraception, the useful androgen draw is in the early follicular phase. The hair-loss guide already cites AJGP 2018 days 4 to 7. A random mid-cycle testosterone on the pill is a hard result to read.
If you cannot stop contraception, the diagnosis can still be made on irregular cycles plus clinical hyperandrogenism, or on a historical diagnosis. The bloods can wait.
Thinning hair is one way hyperandrogenism shows up. The hair-loss blood tests guide covers the AJGP androgen set without repeating this diagnosis algorithm.
Limits
What blood tests cannot show
A PCOS-related printout reports circulating hormones and, if you add them, glucose handling. It does not examine an ovary or forecast a life.
Ovarian morphology
Blood does not count follicles. AMH correlates with that count in adults well enough to stand in for a scan. It is still not a picture, and it is the wrong test in adolescents.
Whether you will conceive, or when
The 2023 guideline tells clinicians to reassure that pregnancy is often achieved naturally or with oral ovulation induction, especially with early diagnosis. That is a clinical conversation. AMH, LH, and a flagged testosterone do not give you a date.
Whether you already have diabetes or fatty liver
HbA1c and fasting glucose miss cases the OGTT would catch. Liver enzymes miss a large share of metabolic liver disease. The guideline's glycaemic test is the OGTT. Imaging the liver is imaging.
Mood, sleep, and endometrial risk
Depression, anxiety, eating disorders, and obstructive sleep apnoea are more common in PCOS. Endometrial hyperplasia and cancer risk is higher (about two- to six-fold in premenopausal women), though the absolute risk stays low. None of those is a hormone-panel diagnosis. Persistent bleeding or a cycle longer than 90 days is a GP visit, not a membership question.
Sudden virilisation, a rapidly growing mass of hair, or postmenopausal new hyperandrogenism is not a preventative-panel problem. Those belong with a GP or endocrinologist this week.
How to order
How Australians get these tests
| Approach | Best for | Typical cost | What it covers |
|---|---|---|---|
| GP-ordered Medicare hormone set | Irregular cycles, clinical hyperandrogenism, or a first suspected diagnosis | Often bulk-billed or low gap when clinically indicated. MBS item 66695 covers one hormone or binding-protein assay (schedule fee $30.50), including total or free testosterone, SHBG, FSH, LH, oestradiol, prolactin, DHEA-S, insulin, androstenedione, and 17-hydroxyprogesterone. Extra assays on the same request use the related P2 items. | The androgen and exclusion set your GP actually writes. Not AMH. Not a 75 g OGTT unless that is ordered separately. |
| GP-ordered OGTT and lipids | Anyone with a PCOS diagnosis, and anyone planning pregnancy | Usually Medicare-rebated when indicated. Confirm with the collection site. | 75 g oral glucose tolerance test, plus the fasting lipid profile the 2023 algorithm asks for at diagnosis. HbA1c and fasting glucose if OGTT is not the first test. |
| GP-ordered private AMH | An adult who has only one of the first two diagnostic features and prefers blood to a scan | Private. RCPA classifies AMH as non-MBS-rebatable. Ask the lab. Do not use a US price list. | AMH in pmol/L. Do not also book a morphology ultrasound for the same criterion. |
| Membership platforms (e.g. Hemexa) | Androgens, SHBG, the female hormone set, and a metabolic trend on one record | ~$1,199/year (full membership) | Total testosterone, calculated free testosterone, SHBG, DHEA-S, oestradiol, FSH, LH, and progesterone on the female annual baseline, plus fasting insulin, HOMA-IR, fasting glucose, HbA1c, and lipids among 76–80 signature markers. AMH and prolactin are add-ons. Hemexa does not run a 75 g OGTT. |
Pathology in Australia requires an authorised request from a registered medical practitioner. Collection for Hemexa members is through Healius Pathology, with regional brands that differ by state. A request form is not a diagnosis, and a membership is not a substitute for a GP visit about sudden virilisation, persistent bleeding, or infertility.
FAQ
Frequently asked questions
- Can a blood test diagnose PCOS?
- No single assay can. Adult diagnosis needs two of three features after other causes are excluded. Blood can show biochemical hyperandrogenism and can, in adults, stand in for ultrasound morphology via AMH. If you already have irregular cycles and clinical hyperandrogenism, you may not need either extra test.
- My ultrasound said I have cysts. Do I have PCOS?
- Not from the scan alone. Polycystic ovarian morphology is one of three adult criteria, and the things counted are follicles, not pathological cysts. You still need a second feature, and other causes still need excluding. In adolescents the scan is the wrong test.
- Why is my testosterone normal if I have acne and extra hair?
- Clinical hyperandrogenism counts without a raised blood androgen. Biochemical testing is for when the clinical picture is unclear. Low SHBG can also raise free testosterone while the total stays inside the printed interval. Direct free-testosterone immunoassays are not the preferred method.
- Is fasting insulin the PCOS test?
- No. Insulin resistance is central to the biology. The 2023 guideline still says clinically available insulin assays have limited clinical relevance and should not be used in routine care. The glycaemic test it wants is a 75 g OGTT. Fasting insulin and HOMA-IR can still be useful on a preventative record. They are not a diagnostic criterion.
- Do I need an AMH test?
- Only if you are an adult, you have just one of the first two diagnostic features, and you are using AMH instead of a morphology ultrasound. Do not order both. Do not use AMH in adolescents. RCPA classifies it as non-MBS-rebatable.
- Can I be tested while I am on the pill?
- You can have a TSH, prolactin, and a metabolic set. You cannot reliably assess biochemical hyperandrogenism, and AMH is suppressed. The 2023 algorithm asks for at least three months off hormonal contraception, with an alternative method, before those androgen bloods. Do not stop contraception because a webpage said to.
- How much do PCOS blood tests cost in Australia?
- Androgens, SHBG, gonadotrophins, oestradiol, prolactin, DHEA-S, insulin, and 17-hydroxyprogesterone sit on MBS item 66695 (schedule fee $30.50 for one assay) and the related P2 items when a GP orders them for a clinical reason. A 75 g OGTT and a standard lipid panel are usually rebated when indicated. AMH is non-MBS-rebatable. Memberships that include the hormone and metabolic markers on a panel of 76–80 signature markers start around AU$1,199 per year.
- Does Hemexa diagnose PCOS?
- No. Hemexa's female annual baseline includes total testosterone, calculated free testosterone, SHBG, DHEA-S, oestradiol, FSH, LH, and progesterone, among 76–80 signature markers. Fasting insulin, HOMA-IR, fasting glucose, HbA1c, and lipids repeat on the included six-month retest where those markers are biannual. AMH and prolactin are add-ons. Collection is through Healius Pathology. A membership does not replace a GP or the 75 g OGTT the 2023 guideline asks for.
How Hemexa fits
How Hemexa can help
PCOS is diagnosed in a clinic. The useful product move is to put the androgens, the binding protein, the female hormone set, and the metabolic markers on one record, then watch the ones that move.
Androgens and SHBG each baseline
Hemexa includes total testosterone, calculated free testosterone, SHBG, and DHEA-S on the annual baseline. That matches the 2023 first-line androgen set (total plus calculated free) and keeps DHEA-S available as the second-line adrenal androgen. 17-hydroxyprogesterone and androstenedione are not on the panel.
Female hormones on the female baseline
Oestradiol, FSH, LH, and progesterone are the four female-only baseline markers. They help exclude other causes and describe the cycle. They are not themselves a PCOS diagnosis. AMH and prolactin are add-ons, not silent inclusions.
Metabolic markers twice a year
Fasting insulin, HOMA-IR, fasting glucose, and HbA1c sit on the annual baseline and the included six-month retest. That is Hemexa's preventative cadence, not a substitute for the 75 g OGTT the 2023 guideline prefers for glycaemic status. Lipids are on the same panel.
Dashboard bands are Hemexa policy
The member app applies Hemexa preventative bands to these markers. Hormone policy rows in the live database conflict and several cite a single-state lab brand, so this page does not quote a single Hemexa "optimal" testosterone or SHBG. Use the interval on your own report, then the dashboard as company policy, not as the 2023 diagnostic rule.
Collection through Healius Pathology
Healius Pathology is Hemexa's collection partner, with regional brands that differ by state. Results from other Australian labs can be imported after your first baseline. The partner is the network, not one state brand.
Sources
References
Teede, H. J., et al. (2024). Summary of the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome: an Australian perspective. Medical Journal of Australia, 221(7). View source ↗
Teede, H. J., et al. (2023). Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome. Journal of Clinical Endocrinology & Metabolism, 108(10), 2447-2469. View source ↗
Teede, H. J., et al. (2023). Recommendations from the 2023 international evidence-based guideline for the assessment and management of polycystic ovary syndrome. Fertility and Sterility, 120(4), 767-793. View source ↗
Monash Centre for Health Research and Implementation. 2023 International evidence-based guideline for the assessment and management of polycystic ovary syndrome. Full guideline and Algorithm 1. View source ↗
Royal College of Pathologists of Australasia. Testosterone total (RCPA Manual, last reviewed 2 January 2024). View source ↗