HRT — The Complete Clinical Picture

Here’s a pattern I see often. This is a composite of several women, not one person. She is 47, with fatigue, poor sleep, brain fog, weight gain around the middle and mood swings. Her GP, after a ten-minute appointment, says her symptoms suggest perimenopause and offers HRT. No blood tests.

That GP is following the guideline. NICE says that over 45, perimenopause and menopause are diagnosed from symptoms, and blood tests “should not be used” to do it (NICE NG23, updated November 2024). HRT is often the right answer for women like her.

What a ten-minute appointment often doesn’t leave room for is the rest of the picture. Fatigue, poor sleep and brain fog are shared with other common, treatable things: an underactive thyroid, low iron, low B12, poor sleep for other reasons, depression, blood sugar problems. Checking for those isn’t instead of HRT. It’s so that if she starts HRT and still feels exhausted, nobody has missed the low ferritin.

This post goes through the questions worth asking, what the different hormone tests can and can’t tell you, the delivery options, and the risks, including breast cancer, which a page calling itself “the complete clinical picture” has to include.

Questions worth asking before or alongside HRT

Questions worth asking

Is this perimenopause or menopause, and how is that decided? Over 45, NICE diagnoses it from symptoms without blood tests. Between 40 and 45 with symptoms, including a change in periods, an FSH test may be used. Under 40, menopause is diagnosed as premature ovarian insufficiency: the 2024 international guideline needs one raised FSH above 25 IU/L, with a repeat FSH or an AMH test if there’s doubt (Panay 2024). The old rule of “FSH above 40 on two tests a year apart” that this page used to give was wrong, and I’ve removed it.

What else could be causing the symptoms? Thyroid function, ferritin and B12 are worth checking when fatigue and brain fog lead. Hot flushes, night sweats and vaginal dryness point more specifically to menopause. Grade: standard practice. (This page used to say an underactive thyroid raises SHBG. It’s the other way round: thyroid hormone raises SHBG (established physiology), so I’ve removed that line.)

Is stress suppressing ovulation? Severe stress, under-eating or heavy training can turn down the brain signals that drive ovulation. Functional hypothalamic amenorrhoea is the clearest example. Grade: established at the severe end; less clear for everyday stress. (The popular “pregnenolone steal” explanation doesn’t hold: the hormones are made in different tissues.)

Does oestrogen breakdown matter? A DUTCH urine test shows which routes oestrogen is broken down by. The idea that this pattern predicts risk is appealing but not well supported. More below. Grade: observational, with mixed findings.

What about the gut? Gut bacteria make beta-glucuronidase, an enzyme that can re-activate oestrogen the liver has packaged for removal. It’s measured on the GI-MAP. Grade: mechanistic. A link between that stool result and blood oestrogen levels hasn’t been established before menopause.

The oestrogen breakdown pathways

The liver breaks oestrogen down along three main routes. A DUTCH test reports how much goes down each.

Oestrogen breakdown pathways, as reported on DUTCH

2-OH pathway

2-OH-E1

Weakly oestrogenic. Often called the “protective” pathway. Methylation converts it to 2-methoxyoestrone, and DUTCH reports that step too.

4-OH pathway

4-OH-E1

Can form quinones that damage DNA in laboratory studies (mechanistic). Cleared by methylation through the COMT enzyme.

16-OH pathway

16-OH-E1

More strongly oestrogenic. Body weight and alcohol are among the things that influence it.

How sure is this? The 2-OH:16-OH ratio has been studied as a breast cancer risk marker. In a pooled analysis of five studies, a higher ratio was not linked to lower risk after menopause. Before menopause the trend pointed that way, but it wasn’t statistically clear (Dallal 2013). Grade: observational, mixed. The “optimal” percentages printed on reports are the lab’s reference ranges, not cut-offs shown to predict outcomes. Supplements such as DIM are sold to shift the ratio; whether shifting it changes anyone’s risk hasn’t been shown.

Blood, saliva and urine compared

Each sample type has genuine strengths, and none is complete on its own.

Method	What it measures	Strengths	Limitations	Best used for
Blood serum	Total circulating hormones (bound and unbound): FSH, LH, oestradiol, progesterone, testosterone, DHEA-S, SHBG, TSH, free T3 and T4.	Large evidence base. NHS standard. FSH where NICE uses it (under 45). Total testosterone and SHBG allow free testosterone to be calculated. Thyroid panel.	Single time point. No metabolite information. Perimenopausal hormone levels swing from day to day, which is why NICE doesn’t use them to diagnose over 45.	Thyroid, iron and B12 checks; FSH under 45; testosterone and SHBG; checking levels if symptoms don’t respond to treatment.
Saliva	Free (unbound) hormone fraction, particularly cortisol, oestradiol, progesterone, testosterone, DHEA.	Non-invasive. Cortisol can be measured at several points in the day.	Less validation than blood or urine. Easily contaminated (food, bleeding gums). Unreliable for sex hormones when using creams or gels.	Cortisol across the day. Limited use for sex hormones.
Urine (DUTCH)	Hormone metabolites from dried urine samples over one day: oestrogen, progesterone and androgen breakdown products, cortisol pattern. DUTCH Plus adds saliva for the cortisol rise after waking.	Shows breakdown products, not just levels. Collected at home.	Not on the NHS. One day’s pattern. Doesn’t replace FSH where FSH is needed. No trial shows that using it improves HRT outcomes.	Context when symptoms are unexplained, or don’t settle on treatment. Not needed to start HRT.

Delivery methods

How a hormone is delivered changes how it’s absorbed, whether it passes through the liver first, and its risks.

Method	First pass through liver	Advantages	Limitations and risks	Follow-up
Oral tablet or capsule	Yes	Convenient. Micronised progesterone (Utrogestan) is taken orally and is widely used for womb protection.	Oral oestrogen raises clotting factors and SHBG. Oral HRT was associated with more blood clots (VTE) in a large UK study; the risk was lower with oestradiol than with conjugated equine oestrogen (Vinogradova 2019, observational).	Symptom review with your prescriber.
Transdermal patch	Bypassed	Avoids the liver first pass. In the same UK study, patches and gels were not associated with more VTE (Vinogradova 2019). Once or twice-weekly application.	Skin reactions in some women. Absorption varies with skin and site.	Symptom review with your prescriber.
Transdermal gel or spray	Bypassed	Flexible dosing. Oestradiol gels are widely available on the NHS. Testosterone gel is sometimes added for low libido.	Absorption varies. Can transfer to partners or children by skin contact. Progesterone cream is poorly absorbed and shouldn’t be relied on to protect the womb.	Symptom review; blood levels if symptoms don’t respond, or for testosterone.
Sublingual drops or troches	Partly bypassed	Usually compounded “bioidentical” preparations.	Compounded hormones aren’t licensed products in the UK, so dose and quality can vary. Less evidence than patches or gels.	Blood levels and symptom review.
Pellet implants	Bypassed	Release over months. No daily routine. Used mainly in US private practice.	Can’t be removed if side effects occur. Risk of levels going too high, particularly testosterone. Not widely available in the UK.	Blood levels before each new implant.

“Bioidentical”, the WHI trial, and what changed

The Women’s Health Initiative trial, stopped early in 2002, gave 16,608 women aged 50 to 79 oral conjugated equine oestrogen with medroxyprogesterone acetate, or placebo. After about five years, the hormone group had more breast cancer, heart disease, stroke and blood clots, and fewer hip fractures and bowel cancers. Overall, the risks outweighed the benefits for preventing chronic disease (Rossouw 2002). Grade: strong, high-quality randomised trial.

Those results apply most directly to that combination, taken by mouth, by women who were mostly well past menopause. Applying them unchanged to oestradiol patches or gels in women in their late 40s and early 50s isn’t straightforward. The clearest evidence for a difference is on blood clots. Transdermal HRT wasn’t associated with VTE (adjusted odds ratio 0.93, 95% CI 0.87–1.01), while oral HRT was (1.58, 1.52–1.64) (Vinogradova 2019). Grade: observational, from 80,396 cases in UK GP records.

Micronised progesterone is widely thought to carry less breast risk than synthetic progestogens. Grade: observational, not verified here. I haven’t checked the source studies for this page, so I’m not putting a figure on it.

The timing hypothesis: both halves

A Cochrane review of 19 trials with 40,410 women found that hormone therapy overall gave no protection against heart disease or death and increased stroke (RR 1.24) and VTE (RR 1.92) (high-quality evidence). In women who started within 10 years of menopause, there was lower mortality (RR 0.70) and less coronary heart disease (RR 0.52), but still more VTE (moderate-quality evidence, from a subgroup analysis) (Boardman 2015). So HRT started within 10 years may lower heart disease and mortality. That’s not a reason to start HRT for heart protection, and started later it adds stroke and clot risk without that benefit.

Breast cancer

The largest analysis pooled the worldwide prospective studies: 108,647 women with breast cancer (Collaborative Group on Hormonal Factors in Breast Cancer, 2019).

Every type of HRT except vaginal oestrogen was associated with some extra breast cancer risk. The risk rose with longer use, and was greater for oestrogen plus a progestogen than for oestrogen alone.
In absolute terms: five years of HRT from age 50 was estimated to cause about one extra breast cancer between ages 50 and 69 for every 50 women on oestrogen plus a daily progestogen, one in 70 on oestrogen plus an intermittent progestogen, and one in 200 on oestrogen alone.
After stopping: some extra risk lasted more than 10 years, depending on how long HRT had been used. There was little extra risk after less than a year of use.

Grade: observational (if the associations are largely causal, as the authors note), very large, and consistent with the WHI trial’s finding of more breast cancer on combined HRT.

This is part of the decision, not a reason to refuse HRT. For many women, the benefits for symptoms, sleep, bone and quality of life are worth it. But it belongs in the conversation, and no urine or saliva test changes these figures.

Progesterone

Progesterone’s breakdown product allopregnanolone acts on GABA receptors, the brain’s main calming system. That is one likely reason progesterone affects sleep and anxiety, and why low progesterone in cycles where ovulation doesn’t happen can show up as poor sleep and low mood. Grade: mechanistic, with some clinical support. In HRT, the main reason for a progestogen is to protect the womb lining if you still have a womb.

What monitoring looks like

On the NHS, HRT is reviewed by symptoms, with bleeding problems investigated. Blood levels are checked when symptoms don’t respond or for testosterone. That’s reasonable.

Some of the women I work with choose to add a DUTCH test before starting HRT and again a few months in, to see the breakdown pattern and cortisol. Grade: not trial-tested. It can add context; it doesn’t tell you whether HRT is safe for you, and it isn’t a minimum standard. Earlier versions of this page said it was, and recommended MTHFR testing and “methylation support” before HRT. There’s no evidence for either, so I’ve removed them.

Doing nothing, and the case against it

HRT isn’t right for everyone. Some women find their symptoms are mostly explained by something else (thyroid, iron, sleep, stress), and do well once that’s addressed.

For many others, menopause is causing real harm: hot flushes and night sweats that wreck sleep, vaginal and urinary symptoms, and bone loss. For them HRT is effective, and withholding it on vague, population-level fears does real harm. The decision belongs to the woman and her prescriber, weighing her symptoms against her own risks, including breast cancer and clots.

“The question isn’t simply whether to use HRT. It’s which hormone, in which form, by which route, for which woman, with the risks set out honestly.”

Sources

NICE. Menopause: identification and management (NG23), updated 7 November 2024. Quoted via British Menopause Society, Menopause: identification and management – from guideline to practice, March 2025.
Panay N et al. Evidence-based guideline: premature ovarian insufficiency (ESHRE, ASRM, CRE-WHiRL, IMS). Hum Reprod Open 2024;2024(4):hoae065. PMID 39660328. doi:10.1093/hropen/hoae065
Dallal CM et al. Urinary estrogen metabolites and breast cancer: a combined analysis of individual level data. Int J Biol Markers 2013;28(1):3–16. PMID 22865302. doi:10.5301/JBM.2012.9353
Rossouw JE et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288(3):321–33. PMID 12117397. doi:10.1001/jama.288.3.321
Vinogradova Y et al. Use of hormone replacement therapy and risk of venous thromboembolism: nested case-control studies using the QResearch and CPRD databases. BMJ 2019;364:k4810. PMID 30626577. doi:10.1136/bmj.k4810
Boardman HMP et al. Hormone therapy for preventing cardiovascular disease in post-menopausal women. Cochrane Database Syst Rev 2015;(3):CD002229. PMID 25754617. doi:10.1002/14651858.CD002229.pub4
Collaborative Group on Hormonal Factors in Breast Cancer. Type and timing of menopausal hormone therapy and breast cancer risk: individual participant meta-analysis of the worldwide epidemiological evidence. Lancet 2019;394(10204):1159–68. PMID 31474332. doi:10.1016/S0140-6736(19)31709-X

Updated 5 October 2026. Corrected: the FSH rule for diagnosing menopause, the thyroid–SHBG direction, and the NICE date. Added: breast cancer, and both halves of the timing evidence. Removed: claims that DUTCH testing is essential before HRT, MTHFR-based advice, and unsourced “optimal” percentages. The opening example is a composite.

Want the rest of the picture alongside your GP’s advice?

An Initial Consultation goes through your symptoms, history and any results you already have, and sets out which checks are worth doing (thyroid, iron, B12, and, where it adds something, a DUTCH test) so you can take a clearer picture to your prescriber.

Book an Initial Consultation (£145) →