Hormone Testing for Women: Which Test, Which Timing
A hormone value is only as good as the day on which you measure it. Which test is good for what, why the cycle day often matters more than the number, and why normal does not always mean well.
Many women come to me with a stack of lab values and a sentence already lodged in their head: „Everything is in the green zone, so I must be imagining it." I see it differently. A hormone test measures a snapshot on a particular day, not how you feel. Hormones fluctuate across the cycle, they act in relation to one another, and a value without the right timing says almost nothing. Before we test, I therefore always ask: what would this result change?
Perhaps you know this. For months you have not felt like yourself. Your mood drops before your period, your sleep is shallow, your energy gone. So you have your hormones tested, hoping for a number that finally explains how you feel. And then the result arrives: all normal. You sit there with a finding that clarifies nothing, wondering whether you are imagining things.
You are not. The problem often does not lie in your body, but in how the testing and the reading were done. In this article we clarify which values make sense when, on which cycle day which hormone belongs, what blood, saliva and the DUTCH urine test can really do, and why a reference range is something different from a functional optimum. This is not about testing more. It is about testing more wisely.
Why the timing often matters more than the value
Imagine you photograph the sea and want to capture the height of a wave. If you take the photo at the wrong moment, you see stillness, even though a big wave was there just a second ago. Hormones across the cycle work much the same way. They are not fixed numbers but a curve that changes day by day.
This is especially true for progesterone. It is only produced in meaningful amounts after ovulation, by the corpus luteum in the second half of the cycle. In the first half it is low, and that is entirely normal. Measuring progesterone on cycle day 3 gives a low value that says absolutely nothing about whether the second half of the cycle is sound. The right time is about one week after presumed ovulation. In a 28-day cycle this is often around cycle day 21, and correspondingly later in longer cycles.
Why good cycle diagnostics look at ovulation and the second half of the cycle together
Systematic review, 60 studies According to PubMed, Bernadette Taim and colleagues in 2023 in Sports Medicine evaluated 60 studies with 6,380 women on how cycle disorders are assessed. A central finding: 90 percent of studies had captured cycle function only retrospectively by self-report, which is unreliable. The authors instead recommend a combination of calendar counting, urinary ovulation tests and a progesterone measurement in the middle of the second half of the cycle, in order to judge ovulation and the function of the second half of the cycle cleanly. This shows that a single value without the right timing is not enough.
Taim BC, Ó Catháin C, Renard M, et al. Sports Med. 2023;53(10):1963-1984. doi:10.1007/s40279-023-01871-8 · PMID: 37389782
Other hormones have other windows. Estrogen, FSH and LH are frequently measured in the early first half of the cycle, often around cycle day 3, because the baseline regulation can be read most clearly there. Androgens such as testosterone and fasting insulin are less cycle-dependent and can be measured more flexibly. And now you know why your doctor asks about the first day of your last bleed before planning the blood draw.
A hormone test without the right cycle day is like a time of day without a date. The number may be correct, but it floats in a vacuum. Before you get upset or reassured about a value, it is worth asking: on which cycle day was it measured, and does that day fit the hormone? Often a seemingly abnormal finding dissolves once the timing is right.
Blood, saliva or DUTCH: what the three routes really measure
The most common question in my consultation is: saliva test or blood test, which is better? The honest answer is uncomfortable. Neither is better across the board. They measure different things, and which one fits depends on the question.
The blood test is the established standard. It usually measures the total hormone in the blood, that is the bound plus the free fraction. For mapping the cycle, ovulation and fertility it is well studied and delivers reliable, comparable values. The saliva test captures more of the free, not protein-bound fraction of a hormone. That can be practical when the binding proteins are altered or when you want to map a daily course, for example with cortisol.
Where saliva can be useful for cortisol
Review According to PubMed, Joanne Blair and colleagues in 2017 in Current Opinion in Endocrinology, Diabetes and Obesity summarised the state of measuring cortisol in saliva. Saliva captures only the free, that is the active fraction. Several samples can be collected conveniently in everyday life, and sampling causes less stress than a venepuncture, which can itself release cortisol. With altered binding behaviour, saliva could even reflect the adrenal reserve more accurately than serum cortisol. For the cortisol rhythm, saliva can therefore be a useful tool.
Blair J, Adaway J, Keevil B, Ross R. Curr Opin Endocrinol Diabetes Obes. 2017;24(3):161-168. doi:10.1097/MED.0000000000000328 · PMID: 28375882
The DUTCH test works with dried urine collected across the day. Its real appeal lies not in the hormones themselves but in the breakdown products. It can show which pathways the body uses to further process estrogen, for example. That is an interesting additional view, especially when it comes to the liver and estrogen clearance. But the evidence for hard clinical decisions is thinner than for blood.
What a dried-urine test can map and what remains open
Real-world analysis According to PubMed, Mark Newman and colleagues in 2022 in Steroids examined whether a test using dried urine can reflect estrogen uptake under an estradiol patch. In women after menopause, estradiol and further estrogen metabolites in urine rose measurably with increasing patch dose. The authors themselves stress, however, that prospective studies linked to clinical outcomes are still missing. The test can therefore make uptake and breakdown visible, but it does not replace established standard diagnostics for treatment decisions.
Newman MS, Mayfield BP, Saltiel D, Stanczyk FZ. Steroids. 2022;189:109149. doi:10.1016/j.steroids.2022.109149 · PMID: 36414155
„The DUTCH test shows me more than the blood test." It is not that simple. The DUTCH test shows something different, namely breakdown pathways, not something better. For the question of whether you ovulate or whether your thyroid is involved, the blood test remains the more reliable route. The tests are not in competition with one another; they open different windows onto different questions.
Which value answers which question
Before a tube of blood is drawn, it should be clear which question it is meant to answer. Otherwise numbers without consequence arise. Here are the four question fields that count most often in practice, and which values belong to them.
Does ovulation occur?
Here progesterone in the second half of the cycle is the key. It is only produced by the corpus luteum after ovulation. A clearly raised value about one week after presumed ovulation speaks for an ovulation having taken place. If progesterone stays low even then, that can point to an absent or weak ovulation. This is often what sits behind the picture of a relative progesterone deficiency.
Is there an androgen excess?
When PCOS is suspected, free testosterone is more sensitive than total testosterone for detecting androgen excess. Sex hormone-binding globulin belongs in the panel, as it co-determines how much hormone is freely available, along with 17-hydroxyprogesterone to distinguish other causes. These values, however, only form a picture in the context of the cycle and the symptoms.
Is the thyroid involved?
Thyroid and sex hormones are closely interwoven. For cycle and mood complaints, TSH and the free thyroid hormones belong in the picture, often complemented by thyroid antibodies. A borderline thyroid can amplify complaints that look like a pure hormone problem. Measuring only the sex hormones easily overlooks this player.
Is iron or reserve lacking?
Exhaustion and cycle disorders often lead to a pure hormone focus, yet it is worth looking at ferritin as the iron store. For questions about egg reserve and the transition into menopause, anti-Mullerian hormone can be a complementary building block. On their own each says little, but in context they say much.
Why free testosterone counts in PCOS, not total testosterone
Consensus Guideline According to PubMed, the joint guideline of the American endocrinology societies and the Androgen Excess and PCOS Society from 2015 holds that free testosterone is more sensitive than total testosterone for demonstrating an androgen excess. The diagnosis of PCOS, accordingly, never rests on a single value but on at least two of three criteria: cycle disorder, signs of androgen excess and a typical ultrasound finding. This underlines that a lab value alone does not make a diagnosis.
Goodman NF, Cobin RH, Futterweit W, et al. Endocr Pract. 2015;21(11):1291-1300. doi:10.4158/EP15748.DSC · PMID: 26509855
With anti-Mullerian hormone an honest framing is worthwhile, because it is marketed as a „fertility test." According to PubMed, a review by Marcelle Cedars in 2022 in the Journal of Clinical Endocrinology and Metabolism describes that anti-Mullerian hormone estimates the quantity of maturing follicles, but says nothing about the quality of the eggs or the chance of pregnancy (doi:10.1210/clinem/dgac039, PMID: 35100616). Hormonal contraception can also lower the value without the reserve actually being smaller. And now you know why a single AMH number is not a verdict on your fertility.
Reference range versus functional optimum: what normal really means
Here lies one of the biggest stumbling blocks. A reference range describes where the middle 95 percent of a comparison group fall. It tells you what is statistically common. It does not tell you where you feel best. These two things are constantly confused.
A good example is the thyroid. The reference range for TSH is broad, and its upper end is drawn differently depending on the lab and the available evidence.
Why the normal range for TSH is not a fixed truth
Population study, n≈360,000 According to PubMed, Qiang Su and colleagues in 2018 in Biological Trace Element Research determined the TSH reference range in around 360,000 women of reproductive age in rural regions of China. The result: the range lay at about 0.39 to 5.13 and shifted noticeably by age and region. The mean rose with age. This shows that a reference range fluctuates from group to group and is not a universal boundary. A value near the upper limit can be unremarkable for one woman and relevant for another.
Su Q, Zhang S, Hu M, et al. Biol Trace Elem Res. 2018;189(2):336-343. doi:10.1007/s12011-018-1480-1 · PMID: 30143915
That context counts is also clear from life phases. According to PubMed, Sun Lee and Elizabeth Pearce in 2021 in the Journal of Clinical Endocrinology and Metabolism describe that narrower TSH target values apply in pregnancy than otherwise, because thyroid hormone plays a special role there (doi:10.1210/clinem/dgaa945, PMID: 33349844). The same value is therefore judged differently depending on the situation. From the perspective of functional and integrative medicine this is an important point: we look not only at whether a value lies within the limits, but where it lies within the span and whether it fits your symptom picture. This is not a dismissal of the reference range, but a complement.
Normal means common, not optimal. A value at the lower edge of the normal range is statistically unremarkable and can still be too low for you. That is no licence to treat every borderline value. But it is a reason not to play your complaints off against a number in the green zone. Both can be true at the same time.
Three levers for a hormone test that truly helps you
Before you take the next test, a clear plan is worthwhile. These three levers are a direction, not a recipe. You will find the individual path with medical guidance.
Clarify the question first, then the test
Ask yourself before every test: what would this result change? Is it about confirming an ovulation, placing the thyroid in context or assessing PCOS? A test with a clear question leads to a consequence. A test without a question just produces numbers. This order saves you money, uncertainty and superfluous findings.
Plan in the right cycle day
Note the first day of your last bleed and align the blood draw with it. Progesterone belongs in the second half of the cycle, about one week after presumed ovulation. Estrogen, FSH and LH usually in the early first half. A cycle diary over two to three months can help narrow down ovulation and makes every measurement more meaningful.
Have the result placed in context
A value alone is a number. Only in the context of your complaints, your life phase and the other values does it become a statement. Insist that not just a single hormone but the picture of cycle, thyroid, iron and metabolism is considered. Good interpretation can also mean that no test is needed.
Sometimes the most honest advice is not to test at all. When complaints fit the symptom picture well and the consequence would be the same anyway, a cycle diary can do more than a lab value. Iron is a good counterexample, because here a measurement often pays off directly. According to PubMed, a small clinical study by Ruchika Sharma and colleagues in 2016 in the American Journal of Hematology showed that in young women with low ferritin and fatigue, but without anaemia, treating the iron deficiency could markedly improve the fatigue in self- and observer ratings (doi:10.1002/ajh.24461, PMID: 27351586). Here the measurement answers a clear question and leads to a clear consequence.
Do not test more, test more wisely
A good hormone test begins not with the tube but with a question. The right timing, the right medium and interpretation in context turn a number into an answer. How you feel is not a disturbance alongside the values. It is the actual yardstick against which any diagnostics must be measured.
Frequently asked questions about hormone testing for women
Which hormones should I have tested as a woman?
It depends on your question. For cycle complaints and suspected progesterone deficiency, progesterone in the second half of the cycle is central. For suspected PCOS, free testosterone, sex hormone-binding globulin and often 17-hydroxyprogesterone belong in the panel. Estrogen, FSH and LH are usually measured in the early first half of the cycle. For exhaustion and cycle disorders, thyroid values such as TSH and free thyroid hormones plus ferritin are worth adding. A single value without reference to your symptoms and your cycle phase says little. That is why the choice of values belongs in medical hands that know your story.
On which cycle day should I test hormones?
Timing decides how meaningful the result is. Progesterone is only informative in the second half of the cycle, ideally about one week after presumed ovulation, often around cycle day 21 in a 28-day cycle, and correspondingly later in longer cycles. Estrogen, FSH and LH are frequently measured in the early first half of the cycle, often around cycle day 3. Androgens and fasting insulin are less cycle-dependent. Measuring progesterone in the first half of the cycle gives a low value that says nothing about the second half. The wrong day can make a healthy system look unwell.
Saliva test or blood test, which is better for hormones?
Both measure different things, and neither is better across the board. The blood test usually measures total hormone, that is bound plus free. The saliva test captures more of the free, unbound fraction. For mapping the cycle and fertility, the blood test is well established and the usual standard. For the daily cortisol rhythm, saliva can be practical, because several samples can be collected conveniently across the day and sampling causes less stress than a blood draw. The choice follows the question, not the trend. When in doubt, the blood test in medical hands is the most reliable starting point.
What is the DUTCH test and is it useful?
DUTCH stands for dried urine collected across the day. The test measures hormones and their breakdown products in urine, for example estrogen metabolites or cortisol metabolites. Its strength is that it can make breakdown pathways visible, that is how the body further processes a hormone. Studies show that it can, for example, reflect hormone uptake under an estrogen patch. Its value for mapping the cycle or for hard decisions is more limited than the blood test, and large studies on clinical endpoints are missing. The DUTCH test can be a complementary piece of the puzzle, but it does not replace standard medical diagnostics.
Why are my hormone values normal even though I feel unwell?
A reference range describes where the middle 95 percent of a comparison group fall, not where you feel best. Your value can sit at the lower edge of the normal range and still be too low for you. Timing adds to this: a progesterone value in the first half of the cycle is low without anything being wrong. And hormones act in relation to one another, not as a single number. That is why normal values and feeling unwell can both be true at the same time. Good diagnostics look at cycle phase, ratio and your symptom picture, not just a single number in the green zone.
How reliable are at-home hormone self-tests?
Home tests can give a first impression, but they have limits. How meaningful they are depends heavily on whether the sample was taken at the right time, whether the lab works cleanly, and above all on who interprets the result. A single value without reference to cycle phase, history and symptoms can confuse more than it clarifies. On top of this, some providers compare values against desired ranges that are not well supported scientifically. A self-test therefore does not replace a medical assessment. Used sensibly, it can start a conversation. Left alone, it easily leads to false conclusions.
Do I even need a hormone test if I have complaints?
A test is not always the first step. Many cycle-related complaints can be assessed well from the symptom picture and a cycle diary, because hormones fluctuate strongly anyway and a single measurement is only a snapshot. A test makes sense when it answers a concrete question, for example whether ovulation occurs, whether the thyroid is involved or whether PCOS is present. A test without a clear question often just produces numbers without consequence. That is why the conversation comes first, along with the question of what the measurement would change.
What does the ratio of estrogen to progesterone in a test mean?
Estrogen and progesterone work as counterparts. For many complaints, what matters is not the absolute value of one hormone but their ratio in the second half of the cycle. When progesterone is too low relative to estrogen, for example in cycles without ovulation or under chronic stress, that can favour complaints. This is often what sits behind the term estrogen dominance. A single estrogen value therefore says little. More informative is the question of whether enough progesterone is produced in the second half of the cycle and how it relates to estrogen. This interpretation belongs in medical hands.
Which values belong in a panel when PCOS is suspected?
When polycystic ovary syndrome is suspected, free testosterone is more sensitive than total testosterone for detecting androgen excess. Sex hormone-binding globulin belongs in the panel, often 17-hydroxyprogesterone to distinguish other causes, and anti-Mullerian hormone as a complementary building block. Because insulin plays a central role in PCOS, fasting insulin and blood sugar are also important. The diagnosis, however, does not rest on a single lab value but on a combination of cycle behaviour, signs of androgen excess and ultrasound. That is why PCOS diagnostics belong in specialist hands that bring all the pieces together.
When should I see a doctor about hormone complaints?
Persistent or new complaints should generally be assessed. You should seek prompt medical assessment for suddenly changed or very heavy bleeding, bleeding after menopause, an absent period without pregnancy over several months, strongly increased body hair with voice changes, and severe premenstrual mood lows. Treatable causes can sit behind hormone complaints, for example thyroid disorders, PCOS or iron deficiency. A test here is a tool, not an end in itself. Good assessment takes your complaints seriously and places the values in your context, rather than treating a number in isolation.
All topics in the cluster „Hormone Guide"
This spoke belongs to the cluster around female hormones as a connected system. Here it continues into greater depth.
- Hormonal imbalance in women (overview/pillar)
- Estrogen dominance: recognise symptoms and address them naturally
- Xenoestrogens: hormone disruptors in everyday life
- Coming off the pill: what happens in the body
- Progesterone deficiency: symptoms and testing
- PMS: Symptome und was helfen kann
- PMDD: when PMS hits the mind
- Perimenopause: symptoms and when it begins
- Wechseljahre: Symptome und was helfen kann
- PCOS: causes and symptoms
- Hormonal acne from within
- Endometriosis: an integrative view
- Hormone-free contraception compared
- Loss of libido in women
- Hormone testing: which test, when
- Lowering estrogen naturally (the liver)
- Cycle-based nutrition
- Thyroid and female hormones
- Insulin resistance and hormones
- Cortisol, stress and female hormones
- Chasteberry and herbal hormone helpers
Connections to other topics
Why a TSH at the upper edge of the normal range can still cause complaints and how the reference range reaches its limits here.
Why cortisol is often better read across the day than in a single value, and what that means for the choice of test medium.
Why a measurement of ferritin often pays off directly, and how iron deficiency amplifies complaints that look like a hormone problem.
Why for many complaints the ratio of estrogen to progesterone counts, and not a single value in the report.
Why the gut co-determines, via estrogen metabolism, how much hormone stays in the body, and what a DUTCH test can make visible of this.
Why the hormone landscape and metabolism change across the cycle, and why that also influences how values are interpreted.
Sources and further reading
- Taim BC, Ó Catháin C, Renard M, et al. The Prevalence of Menstrual Cycle Disorders and Menstrual Cycle-Related Symptoms in Female Athletes: A Systematic Literature Review. Sports Med. 2023;53(10):1963-1984. doi:10.1007/s40279-023-01871-8 · PMID: 37389782 [Systematic Review]
- Arce JC, Balen A, Platteau P, et al. Mid-luteal progesterone concentrations are associated with live birth rates during ovulation induction. Reprod Biomed Online. 2011;22(5):449-456. doi:10.1016/j.rbmo.2011.01.006 · PMID: 21397560 [Cohort, n=335]
- Çinar M, Tokmak A, Kuru Pekcan M, et al. Does mid-luteal progesterone predict pregnancy in intrauterine insemination cycles? Medicine (Baltimore). 2023;102(35):e34754. doi:10.1097/MD.0000000000034754 · PMID: 37657005 [Cohort, n=107]
- Blair J, Adaway J, Keevil B, Ross R. Salivary cortisol and cortisone in the clinical setting. Curr Opin Endocrinol Diabetes Obes. 2017;24(3):161-168. doi:10.1097/MED.0000000000000328 · PMID: 28375882 [Review]
- Newman MS, Mayfield BP, Saltiel D, Stanczyk FZ. Assessing estrogen exposure from transdermal estradiol patch therapy using a dried urine collection and a GC-MS/MS assay. Steroids. 2022;189:109149. doi:10.1016/j.steroids.2022.109149 · PMID: 36414155 [Real-World]
- Goodman NF, Cobin RH, Futterweit W, et al. AACE, ACE, and AES Disease State Clinical Review: Best Practices in the Evaluation and Treatment of Polycystic Ovary Syndrome, Part 1. Endocr Pract. 2015;21(11):1291-1300. doi:10.4158/EP15748.DSC · PMID: 26509855 [Consensus Guideline]
- Cedars MI. Evaluation of Female Fertility, AMH and Ovarian Reserve Testing. J Clin Endocrinol Metab. 2022;107(6):1510-1519. doi:10.1210/clinem/dgac039 · PMID: 35100616 [Review]
- Su Q, Zhang S, Hu M, et al. Reference Range and Sociodemographic Characteristics of TSH among Reproductive Age Women in Rural China. Biol Trace Elem Res. 2018;189(2):336-343. doi:10.1007/s12011-018-1480-1 · PMID: 30143915 [Cohort, n≈360,000]
- Lee SY, Pearce EN. Testing, Monitoring, and Treatment of Thyroid Dysfunction in Pregnancy. J Clin Endocrinol Metab. 2021;106(3):883-892. doi:10.1210/clinem/dgaa945 · PMID: 33349844 [Review]
- Sharma R, Stanek JR, Koch TL, et al. Intravenous iron therapy in non-anemic iron-deficient menstruating adolescent females with fatigue. Am J Hematol. 2016;91(10):973-977. doi:10.1002/ajh.24461 · PMID: 27351586 [Clinical Trial, n=20]
- Wolfram M, Bellingrath S, Kudielka BM. The cortisol awakening response (CAR) across the female menstrual cycle. Psychoneuroendocrinology. 2011;36(6):905-912. doi:10.1016/j.psyneuen.2010.12.006 · PMID: 21237574 [Cohort, n=29]
- Di Lorenzo M, Cacciapuoti N, Lonardo MS, et al. Pathophysiology and Nutritional Approaches in Polycystic Ovary Syndrome (PCOS): A Comprehensive Review. Curr Nutr Rep. 2023;12(3):527-544. doi:10.1007/s13668-023-00479-8 · PMID: 37213054 [Review]