When in Your Cycle to Get PCOS Blood Tests — and What Day 3 Actually Means

Why does the day you’re tested on actually matter?

Testosterone, oestradiol and sex-hormone-binding globulin all rise substantially between the early follicular phase and midcycle in healthy women, according to mass-spectrometry data from 31 premenopausal volunteers — which means the same blood, drawn on two different days, can produce two genuinely different results.

That study measured testosterone, dihydrotestosterone, oestradiol and oestrone using liquid chromatography-tandem mass spectrometry, a far more precise method than the older immunoassays most labs historically used. It found testosterone, free testosterone, oestradiol, oestrone and SHBG all peaked at midcycle and stayed elevated through the mid-luteal phase, compared with the early follicular phase. Dihydrotestosterone was the exception and did not change across the cycle. This is why a testosterone level means little without knowing where in the cycle it was drawn — the reference range a lab applies typically assumes an early-follicular sample, not a midcycle one.

Which tests actually need to be timed to a cycle day?

Four hormones are cycle-sensitive enough to need the early-follicular window, one needs a completely different window, and several do not depend on timing at all — the table below is the full breakdown, test by test.

The LH:FSH ratio is a special case worth calling out on its own: it is only meaningful as an early-follicular reading, because that is the only point in the cycle where both hormones sit at a comparable baseline before either one moves. A ratio calculated from a midcycle sample, when LH naturally surges ahead of FSH, tells you nothing about PCOS — it tells you a surge happened. The same timing sensitivity applies to free versus total testosterone: both numbers shift together across the cycle, so comparing a free-testosterone result from one day against a total-testosterone result from another is not a meaningful comparison either.

What does “day 3” mean if your cycle isn’t 28 days?

Day 1 is always the first day of a full bleed, not a proportional point in your cycle — so day 3 in a 35-day cycle and day 3 in a 24-day cycle are the exact same three days after bleeding started, not three-tenths of the way through.

The early-follicular window is anchored to the start of menstruation because that is when gonadotropin and ovarian hormone levels reset to their lowest point, regardless of how long the rest of the cycle runs. Spotting before a full flow does not count as day 1; if you are unsure whether what you are seeing is the real start, most clinics ask you to wait for a day of proper flow before counting.

Why is progesterone tested on a completely different day?

Progesterone only rises after ovulation, so it has to be drawn roughly a week before your next period is due — for anyone using it to confirm that a cycle was ovulatory, the international guideline is explicit that progesterone, not a cycle-day-2 hormone panel, is the test that answers that specific question.

In a standard 28-day cycle that lands around day 21, but the number that matters is “about a week before the next period,” not a fixed calendar date. Test it too early and progesterone will still be low regardless of whether ovulation later occurred; that is a timing failure, not evidence against ovulation.

What if your periods are too irregular to count a day?

With no predictable cycle to count from, a random-day blood draw, interpreted cautiously and sometimes repeated, is the practical standard — the guideline’s own position is that assessment timing in people with irregular cycles is a judgment call to make with a clinician, not a fixed formula.

A randomly timed test is not equivalent to a properly timed one, and results should be read with that caveat rather than treated as definitive on their own. It is still informative — a markedly high androgen level drawn on any day is still a meaningful finding — but a borderline result on an untimed sample is a reason to repeat it correctly, not to rule PCOS out.

Which tests genuinely don’t care what day it is?

AMH stayed within normal cycle-to-cycle variation across the menstrual cycle in a study of 24 healthy volunteers, which is why it does not need the same day-2-to-5 scheduling that LH, FSH and androgens do — though the international guideline still lists cycle day among the factors labs should be aware can introduce some variability when interpreting results.

That is not a contradiction so much as two different claims: the dedicated timing study found AMH’s natural fluctuation across a cycle stays modest, while the guideline’s broader caution list includes cycle day alongside age, BMI and recent hormonal contraception as things that can shift an individual reading. In practice, this is why AMH can be booked on a convenient day without engineering your schedule around it, while a genuinely borderline result is still worth discussing in context. TSH and prolactin are similarly untethered from the ovarian cycle, although prolactin is best drawn without recent breast stimulation, strenuous exercise or high stress beforehand, since all three can transiently raise it regardless of cycle day.

What good timing does not fix

Correct timing improves how comparable your result is to a reference range; it does not fix a result run on a low-accuracy assay, and it does not replace the exclusion tests a diagnosis depends on. The guideline is explicit that laboratories should use tandem mass spectrometry rather than direct immunoassays for testosterone, because immunoassays lack precision at female concentrations regardless of what day the sample was drawn.

Getting the day right also does not turn a single panel into a diagnosis on its own. Timing affects whether your androgen and gonadotropin numbers are read correctly against a reference range — the actual diagnostic framework, including which findings count and how many are needed, is set out on what actually gets tested for a PCOS diagnosis. And a family history does not change any of this timing advice — it changes what a relative might want to watch for, not when blood should be drawn.

Your next step, this week

Write down the date your last period started and how long your cycles have run over the past three months, even approximately. That single piece of information is what lets a phlebotomist or clinic book the draw for the right window instead of guessing, and it is the same information an adolescent’s assessment is timed against, if the question is about a teenager rather than an adult.

If your cycles are too irregular to count from, say so explicitly when booking rather than picking a day at random and hoping — ask directly whether the clinic wants a random-day draw with a planned repeat, or a progestogen challenge first to establish a new day 1. The full diagnostic work-up, test by test, covers what each result actually means once it comes back.

One naming note, briefly: PCOS was renamed polyendocrine metabolic ovarian syndrome, or PMOS, by global consensus in May 2026. None of the testing or timing recommendations above changed — only the name did. This article uses PCOS because that is still what most people search.