DUTCH Test: Normal Reference Ranges vs. Functional Optimal Levels

What the DUTCH Test Measures and Why It Exists

The DUTCH test collects four or five dried urine samples over a single 24-hour cycle and quantifies over 35 hormone metabolites using liquid chromatography-tandem mass spectrometry (LC-MS/MS). It captures cortisol and cortisone free fractions, their downstream metabolites (tetrahydrocortisol, tetrahydrocortisone, and others), estrogen metabolites across three hydroxylation pathways, androgen metabolites including 5α- and 5β-reduced species, progesterone metabolites, and a small panel of organic acids tied to nutrient cofactors.

Serum hormone testing captures a single-point snapshot. Urine metabolite profiling adds a different dimension: it reflects total daily production, downstream enzyme activity, and clearance patterns that serum alone cannot show. The Endocrine Society's 2014 clinical practice guideline on testosterone deficiency acknowledged that no single test format captures all axes of androgen status, and recommended considering total testosterone alongside free testosterone and clinical context [1]. Urinary metabolite panels extend this principle further by mapping where hormones go after synthesis. A 2019 review in the Journal of Clinical Endocrinology & Metabolism found that urinary cortisol metabolites correlated more tightly with adiposity-related outcomes than morning serum cortisol in a cohort of 1,600 adults from the ECLIPSE study [2]. The clinical question is not whether the DUTCH test is "better" than serum. It answers different questions.

Reference Ranges: How "Normal" Gets Defined

Standard reference ranges on a DUTCH report reflect population-based distributions. A lab samples a reference population (typically healthy adults stratified by sex and, for some analytes, menopausal status or age bracket), measures metabolite concentrations, and sets the "normal" interval as the central 95% of results. Anything between the 2.5th and 97.5th percentile is reported as within range.

This method has a structural limitation. A premenopausal woman with a cortisol metabolite sum at the 5th percentile and another at the 90th percentile both receive a "normal" flag. Their clinical pictures may look completely different. The American Association of Clinical Endocrinologists (AACE) has noted this tension across multiple guidelines, writing in their 2020 position statement on thyroid testing that "reference ranges are statistical constructs, not biological thresholds for health" [3]. The same principle applies to every analyte on a DUTCH panel.

Reference populations also carry selection bias. Volunteers for reference-range studies skew healthier, younger, and more compliant than the general patient population. And because urinary metabolite concentrations vary with hydration status, collection timing, and even ambient temperature during sample drying, interlaboratory variation is wider than for serum assays. This is not a flaw unique to the DUTCH test. A 2017 Endocrine Society scientific statement on steroid hormone measurement emphasized that "standardization of urinary steroid profiling remains incomplete" across all platforms [4].

Functional Optimal: Where the Narrower Targets Come From

Functional optimal ranges are not arbitrary. They emerge from three data streams: symptom-resolution thresholds observed in clinical practice, epidemiologic cutpoints tied to disease risk, and intervention studies that document at which metabolite levels patients respond to therapy.

For example, total DHEA-S measured in serum has well-established age-stratified references. The DUTCH test reports DHEA metabolites (etiocholanolone and androsterone from the DHEA pathway) that parallel these levels. A 2006 study published in the New England Journal of Medicine (N=295) found that DHEA supplementation at 50 mg/day for two years did not improve quality of life or body composition in elderly adults whose levels fell within the lower quartile of the age-matched reference range [5]. Functional practitioners interpret this differently: the "reference-range normal" lower quartile may still warrant investigation if paired with fatigue, low libido, or poor stress resilience.

Dr. Carrie Jones, a naturopathic physician and former medical director at Precision Analytical (the lab that developed the DUTCH test), has explained: "A result can be statistically normal and still not represent where that individual functions best. The reference range tells you where 95% of people land. It does not tell you where the patient in front of you should be."

The distinction matters most for metabolites with continuous risk relationships, where outcomes worsen linearly across the "normal" range rather than flipping at a binary threshold.

Cortisol Metabolites: Pattern Over Total

The DUTCH test reports free cortisol (the bioavailable fraction), free cortisone, and their metabolized downstream products: tetrahydrocortisol (THF), allo-THF (5α-THF), and tetrahydrocortisone (THE). It also maps the diurnal free cortisol pattern across the collection timepoints.

The reference range for total cortisol metabolite output spans roughly 3,000 to 12,000 μg per 24 hours in adult men (slightly lower in women). That is a four-fold spread. A patient producing 3,200 μg/day and one producing 11,500 μg/day both sit inside the reference interval. Functional interpretation narrows this: practitioners often target the 40th to 60th percentile of the sex- and age-matched range for total metabolite output, especially when cortisol-related symptoms (disrupted sleep, afternoon crashes, morning sluggishness) are present.

The diurnal pattern is arguably more informative than total output. The Cortisol Awakening Response (CAR), the spike in free cortisol 30 to 45 minutes after waking, should rise at least 50% above the waking baseline. A flattened or absent CAR has been associated with burnout, PTSD, and chronic fatigue in multiple studies. A 2015 meta-analysis of 80 studies (combined N > 10,000) published in Psychoneuroendocrinology found that a blunted CAR predicted depressive symptoms with moderate effect size (d = 0.41) [6].

The free cortisol-to-free cortisone ratio also matters. The enzyme 11β-HSD1 converts cortisone back to active cortisol in tissues, while 11β-HSD2 inactivates cortisol to cortisone. A skewed ratio toward cortisone dominance (high cortisone, low free cortisol) may suggest excessive 11β-HSD2 activity or impaired 11β-HSD1 regeneration. Reference ranges flag both metabolites independently. Functional interpretation reads the ratio.

Estrogen Metabolites: Three Pathways, Different Implications

Estrogens follow three Phase I hydroxylation routes after synthesis: the 2-hydroxy pathway (2-OHE1), the 4-hydroxy pathway (4-OHE1), and the 16α-hydroxy pathway (16α-OHE1). The DUTCH test quantifies all three, plus their downstream methylated products (2-MeOE1, 4-MeOE1).

The 2-hydroxy pathway produces metabolites generally considered less proliferative. The 4-hydroxy pathway generates catechol estrogens capable of forming DNA-damaging quinones if not properly methylated by COMT (catechol-O-methyltransferase). The 16α-hydroxy pathway produces estriol precursors that bind estrogen receptors with moderate affinity.

A 2012 analysis in Cancer Epidemiology, Biomarkers & Prevention found that premenopausal women in the highest quartile of the urinary 2-OHE1:16α-OHE1 ratio had a 30% lower risk of invasive breast cancer compared to the lowest quartile (OR 0.70 to 95% CI 0.50 to 0.99) [7]. This ratio is one of the most cited functional targets on the DUTCH test. Reference ranges report absolute concentrations. Functional interpretation targets a 2:16 ratio above 2.0, and in some clinical frameworks, above 4.0 for patients with strong family histories of estrogen-receptor-positive breast cancer.

The methylation ratio (2-MeOE1 / 2-OHE1) reflects COMT enzyme activity. Low methylation of 2-OH and 4-OH catechol estrogens can increase oxidative stress. The Endocrine Society has not issued specific guidance on urinary catechol estrogen methylation ratios, but the biochemistry is well-established in the published literature. Patients with homozygous COMT Val158Met polymorphisms (the "slow COMT" variant) tend to show lower methylation activity, which may appear on the DUTCH test as elevated unmethylated catechol estrogens relative to their methylated counterparts [8].

Androgen Metabolites: Testosterone Is Only the Starting Point

The DUTCH test breaks testosterone metabolism into 5α-reduced metabolites (5α-DHT, 5α-androstanediol, androsterone) and 5β-reduced metabolites (etiocholanolone). This split is clinically useful. A patient could have a "normal" total testosterone on serum testing while showing excessive 5α-reductase activity on the DUTCH, meaning more testosterone is being converted to DHT. This pattern is associated with androgenic alopecia, acne, and in women, hirsutism.

The Endocrine Society's 2018 guideline on polycystic ovary syndrome (PCOS) recommended measuring total and free testosterone, DHEA-S, and androstenedione as the first-line androgen workup [9]. Urinary androgen metabolite profiling goes a step further by showing the enzymatic fate of those androgens.

DHEA-S metabolites on the DUTCH test include etiocholanolone and androsterone from the adrenal pathway. Age-related decline in DHEA-S is well-documented. Serum DHEA-S peaks in the mid-20s at approximately 300 to 500 μg/dL in men and 200 to 350 μg/dL in women, then drops roughly 2% per year [10]. Functional practitioners often target the upper half of the age-matched range for DHEA-S metabolites, especially in patients reporting fatigue, poor stress tolerance, or decreased libido.

A 5α-reductase preference ratio above the 75th percentile for the patient's sex and age bracket is a flag that functional clinicians use to consider interventions such as saw palmetto, zinc optimization, or in medically appropriate cases, low-dose finasteride. Standard reference ranges would not flag this pattern because both 5α and 5β metabolites independently fall within "normal."

Progesterone Metabolites and Luteal Phase Assessment

The DUTCH test reports α-pregnanediol and β-pregnanediol as markers of total progesterone production. In cycling women, the test is ideally collected on days 19 to 22 of the cycle (roughly 5 to 7 days post-ovulation) to capture peak luteal output.

Reference ranges for α-pregnanediol in the luteal phase span approximately 300 to 3 to 000 ng/mg creatinine. That ten-fold range encompasses both barely-adequate and robustly-sufficient progesterone production. Functional targets typically aim for the 50th percentile or above. A serum progesterone of 10 ng/mL mid-luteal is often cited as a minimum for adequate luteal function. The American Society for Reproductive Medicine (ASRM) has stated that a single mid-luteal serum progesterone > 3 ng/mL confirms ovulation, but did not define "optimal" for symptom management or cycle quality beyond conception [11]. The DUTCH test fills a gap here by integrating total progesterone output across the collection period rather than capturing a single-point serum level.

The β-pregnanediol-to-α-pregnanediol ratio can indicate 5β-reductase preference in progesterone metabolism. High 5β-reductase activity may shunt more progesterone toward neuroinactive metabolites, potentially reducing the anxiolytic and sleep-promoting effects of progesterone's neuroactive metabolite, allopregnanolone.

Organic Acids: Nutrient Cofactor Markers

The DUTCH test includes four organic acid markers: methylmalonate (MMA, a marker of functional B12 status), xanthurenate (a marker of functional B6 status), pyroglutamate (reflecting glutathione demand), and 6-OH melatonin sulfate (the primary urinary metabolite of melatonin).

The National Institutes of Health Office of Dietary Supplements defines serum B12 < 200 pg/mL as deficient and 200 to 300 pg/mL as borderline [12]. Urinary MMA provides a functional readout: it rises when intracellular B12 is insufficient to run the methylmalonyl-CoA mutase reaction, even if serum B12 appears adequate. A 2014 systematic review in Clinical Chemistry and Laboratory Medicine found that elevated urinary MMA (above ~3.6 μg/mg creatinine) had 86% sensitivity for detecting tissue-level B12 deficiency in patients with serum B12 in the 200 to 400 pg/mL "gray zone" [13].

6-OH melatonin sulfate reference ranges span roughly 15 to 75 ng/mg creatinine. Low melatonin metabolite output can correlate with poor sleep onset, reduced antioxidant reserve, and in emerging research, altered estrogen metabolism (melatonin modulates aromatase activity). Functional practitioners typically want this marker above the 50th percentile, particularly in perimenopausal women dealing with insomnia alongside hormone fluctuations.

How to Act on Results: Retesting and Intervention Timelines

A DUTCH test result is most useful when paired with symptoms, medical history, and concurrent serum labs. No urine metabolite panel should be interpreted in isolation. The AACE's general guidance on laboratory interpretation recommends confirming abnormal results with repeat testing before initiating therapy, particularly for analytes with high intra-individual variation [3].

For cortisol-focused interventions (adaptogen protocols, sleep hygiene restructuring, stress management changes), a 30-day retest interval may capture early shifts in the diurnal pattern. For sex hormone interventions (HRT initiation or dose adjustment, DHEA supplementation, progesterone therapy, aromatase modulation), the typical recommendation is 90 days before repeating the DUTCH panel. This allows time for downstream metabolite pools to equilibrate.

Dr. Mark Newman, the developer of the DUTCH test and founder of Precision Analytical, has stated: "Retesting too early is one of the most common mistakes. Metabolite pools don't shift overnight. You need at least two to three full menstrual cycles for sex hormone metabolites to reflect a true new steady state."

Patients starting bioidentical progesterone at 100 to 200 mg oral micronized per night should expect α-pregnanediol to rise on the follow-up DUTCH, while the ratio of 5β to 5α metabolites may also shift depending on oral versus transdermal delivery route (oral progesterone undergoes extensive first-pass hepatic metabolism, favoring 5α-reduced products including allopregnanolone). Transdermal progesterone shows a different metabolite signature on the DUTCH, typically with lower total α-pregnanediol but a more physiologic metabolite distribution [14].