Saliva vs. Urine: The ‘Elephant in the Room’ Debate

In fall of 2023, Menopause published the first 3-way comparison of estrogen exposure from FDA-approved estradiol patches and gels and compounded estradiol creams. This data was published by the research team at Precision Analytical, makers of the DUTCH Test, and highlighted the utility of dried urine testing for monitoring estrogen replacement therapy.

Shortly after publication, Pam Smith wrote a Letter to the Editor, and in April 2024, this letter, and the subsequent response from Mark Newman was published by the journal. The Letter clearly shines a light on a few of the common misconceptions about salivary monitoring, dried urine testing and HRT in general, and is well worth the read.

This blog will also summarize the key points.

Clinicians newer to HRT can be challenged by hearing conflicting voices in education on whether saliva or urine offers a more clinically useful approach to testing in the setting of hormone replacement therapy. It may be tempting to simply trust someone’s confident statement one way or the other. With this particular clinical decision, however, it is recommended that you take the time to explore and critically evaluate the published literature and your own patients’ outcomes!

How to classify low, medium, and high dose estrogen therapy

The classification of low, medium, and high dose estrogen therapy is well-established in published literature, particularly with FDA-approved bioidentical estradiol patches and gels.  These classifications are based upon clinical outcomes studies and show partial improvement in symptoms (such as VMS) at low doses, with improved outcomes at medium and high doses.

Those dose ranges are:

  • Low Dose ERT: <0.5 mg for cream and gel; 0.025- 0.0375 mg for patch
  • Mid Dose ERT: 0.5-1.0 mg for cream and gel; 0.05 mg for patch
  • High Dose ERT: 1.0-1.5 mg for cream and gel; 0.1 mg for patch

A few questions may arise when you review these values if you are not familiar with the published literature. First, why are doses for patches so much lower than for gels and creams? This was the first criticism raised by Smith, and points to a very common misunderstanding of prescribed estradiol therapies.

The labeled dose for estradiol patches is quite different in approach compared to gels and creams. Gels and creams are labeled according to the amount of estradiol contained in one pump/application of gel. With patches, labeling describes the daily delivered dose, which effectively represents how much should be absorbed daily. As an example, let’s look at one FDA-approved patch labeled 0.05 mg. If you read the packaging on the product, you will find that the patch actually contains 3.8 mg of estradiol. It is intended to be applied for 7 days, so that works out to be 0.54 mg per day. This daily dose contained in the patch would be a better comparison to the 0.5 mg of estradiol in a cream or gel (which would be labeled 0.5 mg). With all applications, it is important to also recognize that there is not 100% absorption. This brings us to the second misconception.

Applied dose is not the delivered dose

If we consider the patch mentioned above, you can see that the patch applies a dose of approximately 0.54 mg daily, yet the patch labeling clarifies that 0.05 mg of estradiol is delivered. So for this particular patch, women, on average, absorb only 9.2% of the estradiol in the patch. Gels and creams will also have different absorption rates. With compounded creams, different base ingredients may impact absorption.

This is a critical understanding in hormone replacement therapy that is commonly missed. The applied dose and the delivered dose of a hormone are different.

This is often misunderstood and leads to providers thinking that low-to-mid dose estradiol application (such as a 0.27 mg gel) is a massive dose. It is true that if 0.27 mg of estradiol was added and was 100% bioavailable in a woman’s system, that would be a massive dose.  However, only approximately 10% is bioavailable. So, you are only introducing approximately 0.027 mg of estradiol into the system (after absorption and bioavailability are considered). Clinical outcomes studies are consistent with this assessment.

For example, 0.27mg EstroGel failed to relieve VMS at 4 weeks (success at 12 weeks) and did not relief vaginal atrophy. Divigel 0.25mg is reported to deliver only a fraction of the administered dose, and this product’s clinical efficacy is in alignment with this assessment as it also failed to relieve VMS.

It is true that saliva test results elevate substantially with estradiol gels and creams, and may suggest that these are high doses, but serum and urine tests as well as clinical outcomes studies (which should be heavily considered!) all conflict with that conclusion.

Finally, it’s very important to consider your own clinical experience as you make decisions.  They should be a consideration as you decide what to do in your own practice. Studies have shown, however, that VMS are significantly impacted by the placebo effect (meaning that placebo is effective also for many women), so balancing your experience with larger, controlled placebo-controlled studies on therapeutic response and optimal dosing is very important.<span style="vertical-align: super; font-size: 10px;">1</span>

We encourage you to read the original manuscript highlighting the estrogen exposure with patches, gels, and creams here. Also, you can find the letter to the editor, and subsequent response, here.


Become a DUTCH Provider to learn more about how comprehensive hormone testing can help you create informed patient treatment plans.