Hormone Testing: When to Use Serum, Saliva, and Urine

Published in: Townsend Letter

By: Pushpa Larsen ND, Michael Kaplan ND, Leah Alvarado ND, and Mi-Jung Lee ND, LAc

Picture this:  A 52 year –old woman comes in to your office for her initial visit.  On the phone she told you that she wanted to come in “for hormones.”  In her intake, you find out that she is still having menstrual cycles, although they are somewhat irregular and sometimes quite heavy.  She has occasional hot flashes, but they are not bothersome.  She feels tired and burned out most of the time although she tends to be optimistic and tries to make the best of things.  She has noticed that her hair is thinning and she has put on a few pounds around her middle.  She had an aunt who died of breast cancer and her mother has osteoporosis, as did her maternal grandmother.  She has a very busy life, with two daughters, 13 and 15, who are active in sports and multiple other activities.  She has her own business as a graphic designer.  Her husband is an attorney.  She stays up late to get work done for clients after her daughters have gone to bed.  She tends to be a bit of a night owl anyway.  Her libido has been diminishing over the past couple of years, although she and her husband have a great partnership and a loving relationship.  A friend of hers said she got her life back after starting on hormones, and your patient would like to see if hormones could make that kind of difference for her, too.  You think she is probably a good candidate for hormones, but you like to know where you are starting, and it seems possible that your patient may have multiple endocrine issues. What is the best way to assess her current hormonal status and monitor any hormone replacement for safety and effectiveness?

Traditionally, all hormone testing was done via serum, but these days, it is possible to use serum, saliva, and urine to assess a wide array of hormones.  Each of these methodologies has advantages and limitations, and each has a best use.   Sometimes a combination of tests will be necessary provide you with the most complete information.


Serum testing of hormones has long been accepted by the conventional medical community as the standard for measuring hormones.  It has the advantage of being a relatively simple collection requiring little patient involvement and has very well-established reference ranges.  Serum is ideal for testing peptide hormones such as FSH, LH, prolactin, fasting insulin, and thyroid hormones including Reverse T3, as well as thyroid antibodies.  It is also used to measure Sex Hormone Binding Globulin (SHBG) and, less commonly, Cortisol Binding Globulin (CBG).  For sex hormones, serum testing has a more limited utility.  There are several reasons for this.  For most sex hormones, no distinction is made in serum between bound and free hormone.  Estradiol, estrone, estriol and progesterone are reported as total hormone and free hormones assays are not commonly available.  This may lead to misleading results in which hormone levels appear to be normal or even high normal because of an abundance of bound hormone.  However, if the free hormone level is low, the patient can be functionally deficient even with a normal total hormone level.  Serum testosterone is an exception in that it is commonly available as both total and free, and therefore can be useful in assessing hormone balance.

Another limitation of serum hormones testing is the “snapshot” nature of single point testing.   Because hormones are secreted in a pulsatile manner over the course of the day (and night), it is difficult to know whether the levels in serum represent a peak, a valley, or something in between.1-2  This also presents a difficulty in monitoring treatment, as it is not possible to know whether today’s test was drawn at a similar point of hormone secretion as a previous test.

Estradiol (E2) is the female hormone most commonly measured in serum, although estrone (E1) is also available from many labs.  Serum Estriol (E3) testing however, is not routinely performed.  E3 is an important estrogen, generally considered protective because it binds to Estrogen Receptor Beta (ERb)3, which is understood to increase differentiation and decrease differentiation of cells. 4-5

Monitoring progesterone supplementation in serum also poses a problem.  Transdermal progesterone does raise serum progesterone levels in a statistically significant manner, but the magnitude of change is quite small,6 which can lead to excessive dosing of progesterone as practitioners strive to achieve therapeutic levels.

Finally, serum hormone testing does not typically allow for the measurement of estrogen, androgen, and adrenal metabolites, which can provide a wealth of information to assist the practitioner in understanding their patient’s condition and help to guide and fine-tune treatment options.


In the past decade and a half, saliva testing has gained in popularity among practitioners of functional medicine.  Saliva has the advantage of being non-invasive as well as being accessible to practitioners such as naturopathic physicians, chiropractors, and acupuncturists who may be practicing in states where they are not licensed to order blood tests or draw blood.  Saliva collection also allows for multiple collections over a period of a day or month which can help elucidate abnormal hormonal patterns, such as a shortened luteal phase.  While this can theoretically be done with a serum test, it would be logistically cumbersome.

Saliva is best used to evaluate the balance and flow of the estrogens and progesterone in women who are still having menstrual cycles.  It can also be used to evaluate cortisol secretion patterns by taking multiple samples over the course of a day and evening.   A saliva test measures free hormone and its multi-point versatility makes it a better measure than serum for evaluating un-supplemented hormone status.  Another novel  use of saliva hormone testing is in pregnancy: salivary estriol spikes about two weeks before the onset of labor and can be used to identify women who are at risk for pre-term labor.7-9

Saliva production is difficult for some patients, and there are multiple restrictions regarding eating, drinking, gum-chewing, make-up use, topical application, and tooth-brushing that must be observed to get a useable specimen.  Micro-damage from tooth brushing can result in elevated salivary testosterone levels for up to an hour after brushing, even in the absence of visible signs of bleeding, such as “pink toobrush”.10  Saliva can only be used to evaluate steroid hormones.  Peptide hormones, such as growth hormone and thyroid are not available.  While estradiol, estrone, and estriol, progesterone, testosterone, DHEA and cortisol are all available, depending on the lab, steroid hormone metabolites are not measured in saliva, limiting its utility in assessing metabolism of hormones.

Saliva measurements are greatly affected by the use of exogenous hormones.11-12  Transdermal progesterone and testosterone, in particular, can result in supra-physiological levels in saliva testing, but all exogenous hormone use seems to distort results to some extent.  Because of this, patients are instructed to discontinue hormone use for between 12-36 hours prior to collection, depending on the hormone preparation.    This can pose problems for practitioners who want to monitor hormone therapy.  Estrogen, for example, washes out of the system almost entirely in 20 hours and drops significantly within 12 hours. 13

Saliva collection, like blood, is a single point collection.   Although cortisol can be collected at multiple points, sex hormones are measured from a single morning collection.  Just as with serum tests, a single point of collection does not account for individual variation and may catch a peak or a valley in hormonal secretion – or perhaps a peak for one hormone and a valley for another.


Measuring hormones in urine is less common in clinical practice than either serum or saliva, yet it is quite common in research.   A 24-hour urine collection is the preferred method for testing hormones that are secreted at night and during deep sleep, such as Growth Hormone14 and Melatonin15, and is the most economical and reliable way to evaluate steroid hormone metabolites.  This method has been gaining in popularity among integrative practitioners since it was brought into clinical practice by Dr. Jonathan Wright in the early 1980s.

Not all urine hormone testing methods are equal.  Serial single-point urine collection does not account for individual differences in hormone secretion, especially for patients with non-traditional schedules, such as shift workers.  Twenty-four hour urine collection accounts for the full day and night of hormonal secretion.  This eliminates the possibility of falsely elevated or depressed levels that may be obtained when a single point collection occurs at a peak or valley of an individual’s secretory cycle.

Just as single-point and 24-hour collection are different, not all technical methods of urine assay are equally accurate. Highly sophisticated Gas Chromatography(GC) run in tandem with Mass Spectrometry(MS) is emerging as the method against which all other methods are measured.16

Urine assays measure unbound hormone, reflecting that which is bioavailable.  The use of 24-hour urine hormone profiles in clinical practice has found these profiles to correlate well with symptoms reported by patients on hormone symptom questionnaires. Similarly, supplementation with exogenous hormones or other treatments and improvement in symptoms is reflected in values seen on follow-up tests.

An important advantage of a 24-hour urine hormone collection is the ability to measure hormone metabolites.  This is most important when evaluating the adequacy and safety of exogenous estrogen supplementation and in assessing adrenal function.   It is well-established that certain estrogen metabolites are “good” estrogens, having a protective effect on estrogen-sensitive tissues.  Other metabolites are known to have more carcinogenic effects.17-18

The utility of measuring estrogen metabolites has been the focus of much recent attention.  Studies from 2012 and 201319-20 (a prospective case-control study and a retrospective case-control study) uncover new data that finds a statistically significant association between 2-hydroxylation pathway estrogen metabolites and lower breast cancer risk. These studies focused on postmenopausal women not taking hormone replacement.  While there are no studies to date that examine estrogen metabolite ratios in postmenopausal women while concurrently taking hormone replacement, it stands to reason that supplemented levels of parent estrogens will make estrogen metabolite analyses an important test in this population group.

Estrogen metabolism is also malleable, being easily altered with supplements such as Diindolmethane (DIM), Indol-3-Carbinol (I3C), methylated B-vitamins and magnesium. 21-23  Being able to evaluate the relative balance of protective estrogens to potentially harmful estrogens and see the effects of treatment interventions is invaluable to the clinician working to maximize quality of life while also protecting against harmful effects.  Because 24-hour urine hormone profiles measure both Phase 1 and Phase 2 liver metabolites of estrogen, these profiles also offer a peek into liver function and may suggest further avenues for treatment.

Twenty-four hour urine hormone panels are excellent for evaluating adrenal health and function.  In addition to measuring DHEA and cortisol, a complete panel also measures cortisone (the storage form of cortisol) and several important cortisol and cortisone metabolites, as well as aldosterone and other mineralocorticoids.  The importance of measuring these metabolites can be seen in the stressed patient with normal or high-normal cortisol.  Cortisol and cortisone metabolites can point to decreasing daily cortisol production that signals adrenal deficiency.   The ratio of cortisol and cortisone is another important indicator of adrenal function, as are the mineralocorticoids:  low mineralocorticoids are a clear indicator of chronic adrenal fatigue and are an excellent marker to monitor adrenal recovery with treatment. 24-25

Twenty-four hour urine hormone results may be altered in patients with significant liver or kidney disease.  Dehydration or excessive fluid intake can also affect the results.  As with all other methods of testing, non-bio-identical hormone substances are not measured.  The 24-hour urine collection does not allow elucidation of the diurnal cortisol pattern.  However, it has been observed clinically that an elevated cortisol and cortisone in the 24-hour urine panel may be related to night-time cortisol spikes.  Some practitioners have expressed a concern about the convenience of a 24-hour urine collection, however all methods have their challenges.  24-hour urine hormone panels contain a wealth of information and may appear overwhelming. A good laboratory will have experts with clinical experience available to help you get the most from your results.

So back our theoretical patient.  What is the best way to work her up?  Let’s start with a review of her symptoms and a DDx.  She is 52, with irregular, sometimes heavy menstrual periods, mild hot flashes, thinning hair, and some central weight gain.  She feels tired and burned out, has a busy and stressful life, probably doesn’t get enough sleep and is losing her sex drive.  She has a family history which suggests some increased risk for breast cancer and a stronger risk for osteoporosis.  She hasn’t seen a doctor in several years, as she has felt well and been busy raising her daughters.  Because of the central weight gain, it would be worth determining if she has some early insulin resistance.  Table 1 outlines a rationale for choosing appropriate laboratory investigations.


For the most complete assessment of hormone function the ideal combination of tests would be an in-depth 24-hour urine hormone profile, a 4-point salivary cortisol, a serum thyroid panel, and serum FSH and LH. FSH and LH are not absolutely necessary. Additional work-up appropriate to DDX would be CBC, serum Fe, TIBC, ferritin, comprehensive metabolic panel, Hgb1C, fasting insulin and a baseline DEXA.

Serum, saliva and urine testing can offer important insight into a patient’s hormonal status. Each has a set of clinical strengths and limitations, such that a combination of testing methodologies may occasionally be appropriate. Multiple testing methods may not always be practical given financial or insurance constraints. However, a well-chosen series of laboratory evaluations can play a critical role in prioritizing and choosing among treatment options.


1. Brambilla DJ et al. Intraindividual variation in levels of serum testosterone and other reproductive and adrenal hormones in men. Clin Endocrinol (Oxf). 2007;67:853-862

2. Plymate SR et al. Circadian variation in testosterone, sex hormone binding globulin, and calculated non-sex hormone binding globulin bound testosterone in healthy young and elderly men. J Androl. 1989. 10:366-371

3. Mishra RG et al. Metabolite ligands of estrogen receptor-beta reduce primate hyperreactivity. Am J physiol. 2006;290(1):H295-303.

4. Maehle BO, et al. Estrogen receptor beta – an independent prognostic marker in estrogen receptor alpha and progesterone receptor-positive breast cancer? APMIS. 2009. Sep;117(9):644-50.

5. Imanov O. et al. Estrogen receptor in health and disease. Biol Reproduct. 2005. 73(5):866-71.

6. Elshafie M. et al. Transdermal natural progesterone cream for postmenopausal women: inconsistent data and complex pharmacokinetics. J Obstet Gynecol. 2007. 27(7)655-9.

7. McGregor JA. et al. Salivary estriol as risk assessment for preterm labor: A prospective trial. Am J Obstet Gynecol. 1995. Oct;173(4):1337-42.

8. McGregor JA. et al. Diurnal variation in saliva estriol level during pregnancy: A pilot study. Am J Obstet Gynecol. 1999. Jan;180(1 pt3):S223-5.

9. Heine RP. et al. Serial salivary estriol to detect an increased risk of preterm birth. Obstet Gynecol. 2000. Oct;96(4):490-7.

10. Granger DA. et al. The “trouble” with salivary testosterone. Psychoneuroendocrinology. 2004 Nov;29(10):1229-40.

11. O’Leary P. et al. Salivary, but not serum or urinary levels of progesterone are elevated after topical application of progesterone cream to pre- and postmenopausal women. Clin Endocrinol. 2000. 53(5):615-20.

12. Lewis JG. et al. Caution on the use of saliva measurements to monitor absorption of progesterone from transdermal creams in postmenopausal women. Maturitas. 2002. 41(1):1-6.

13. Ginsberg ES. et al. Half-life of estradiol in postmenopausal women. Gynecol Obstet Invest. 1998;45(1):45-8.

14. Hirohata T. et al. Urinary growth hormone level and insulin-like growth factor-1 standard deviation score (IGF-SDS) can discriminate adult patients with severe growth hormone deficiency. Endocrine J. 2013. 60(3):369-73.

15. Kovacs J et al. Measurement of urinary melatonin: a useful tool for monitoring serum melatonin after its oral administration. J Clin Endocrinol Metab. 2000. 85(2):666-70.

16. Stanczyk FZ et al. Standardization of steroid hormone assays: Why, how and when? Cancer Epidemiol Biomarkers Prev. 2007. Sep;16(9):1713-9.

17. Bradlow HL. et al. 2-hydroxyestrogen: the ‘good’ estrogen. J Endocrinol. 1996 Sep;150. Suppl: S259-65.

18. Saeed M, et al. Formation of depurinating N3Adenine and N7Guanine adducts by MCF-10F cells cultured in the presence of 4-hydroxyestradiol. Int. J. Cancer. 2007 Apr 15;120(8):1821–4.

19. Fuhrman BJ. et al. Estrogen metabolism and the risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 2012. Feb 22;104(4):326-39.

20. Falk RT. et al. Relationship of serum estrogens and estrogen metabolites to postmenopausal breast cancer: a nested case-control study. Breast Cancer Res. 2013. Apr 22;15(2):R34.

21. Dalessandri KM. et al. Pilot study: Effect of 3,3’-diindolymethane supplements on urinary hormone metabolites in postmenopausal women with a history of early-stage breast cancer. Nutrition and Cancer. 2004. 50(2):161-7.

22. Anderson et al. Nutrition and epigenetics: an interplay of dietary methyl donors, one-carbon metabolism and DNA methylation. J Nutr Biochem. 2012 Aug;23(8):853-9.

23. http://www.genecards.org/cgi-bin/carddisp.pl?gene=COMT&search=COMT

24. Tomlinson J. et al. 11β-hydroxysteroid dehydrogenase type 1: A tissue specific regulator of glucocorticoid response. Endocrine Reviews. 2004. 25(5);831-866.

25. Jerjes WK, Taylor NF, Peters TJ, Wessely S, Cleare AJ. Urinary cortisol and cortisol metabolite excretion in chronic fatigue syndrome. Psychosom. Med. 2006;68(4):578–82.

26. Lotinun S. et al. Tissue selective effects of continuous release of 2-hydroxyestrone and 16alpha-hydroxyestrone on bone, uterus and mammary gland in ovariectomized growing rats. J Endocrinol. 2001. 170(1):165-74.

27. Vassiliadi D. et al. Increased 5 alpha-reductase activity and adrenocortical drive in women with polycystic ovary syndrome. J Clin Endocrinol Metab. 2009. 94(9):3558-66.

28. Baisier WV. et al. Thyroid insufficiency. Is TSH measurement the only diagnostic tool? J Nutr Environ Med. 2000. 10:105-113.

29. Hoshiro M. et al. Comprehensive study of urinary cortisol metabolites in hyperthyroid and hypothyroid patients. Clin Endocrinol. 2006. 64(1):37-45.

30. Taniyama M. et al. Urinary cortisol metabolites in the assessment of peripheral thyroid hormone action: application for diagnosis of resistance to thyroid hormone. Thyroid. 1993. 3(3): 229-33.