DUTCH Hormone Review

What's in the procedure

The full procedure the skill follows in Meelio, as published in the open source repo.

Decision support for licensed healthcare professionals. Analysis only — the practitioner reviews, verifies, and decides.

Gather context (before interpreting anything)

1. Call get_patient_documents to locate the DUTCH report (category lab_report) and its structured labFindings[], and search_patient_documents for the raw report text — the dial graphics render as text, and the lab's interpretive commentary and collection notes matter here. Identify which test it is (DUTCH Complete vs Plus — the Plus adds the salivary cortisol awakening response). If no DUTCH exists in the record, say so and stop.
2. Call get_patient_snapshot for demographics (age, sex), medicalHistory (conditions — especially PCOS, thyroid, hormone-sensitive cancers; medications — HRT/BHRT with route, hormonal contraception, 5α-reductase inhibitors, corticosteroids including inhaled/topical; supplements — melatonin, DHEA), dietAndLifestyle (sleep, stress), and recentEncounters for symptoms and cycle information.
3. Date by specimen/collection date (eventDate). For cycling patients, look for the cycle day at collection in the report text or encounter notes — luteal collection (~day 19–22) is required for progesterone conclusions.
4. Skip anything marked contradicted. If history was truncated, write "history shown may be incomplete" — never "no history".

Interpretation procedure

1. Validity gate — before any interpretation. Pregnancy: DUTCH is not validated — stop and say so. Combined hormonal contraceptives: endogenous sex-hormone output largely uninterpretable — restrict conclusions to cortisol/organic acids and say so. HRT route: oral progesterone inflates urinary pregnanediol out of proportion to tissue effect; vaginal/topical routes map poorly to urine; oral estradiol shifts metabolite proportions — interpret against route-specific expectations and flag uncertainty. Unknown or follicular cycle timing: progesterone data cannot assess luteal adequacy — say so. Melatonin supplementation invalidates 6-OHMS. Confirm the report was run against the correct reference population (premenopausal luteal vs follicular vs postmenopausal; male age bands) — check the report text via search_patient_documents; a normal-looking value against the wrong set invalidates the read, and if menopausal status is ambiguous in the chart (perimenopause, post-ablation, cycle-suppressing therapy), say so and interpret both ways where they diverge.
2. Cortisol — read three things separately, then reconcile: free-cortisol diurnal rhythm (waking/morning/afternoon/night — slope quality, flattening, elevated night vs sleep complaints, low throughout); metabolized cortisol (THF+THE = total production) and the informative discordances — high production + low free = fast clearance (hyperthyroid, obesity); low production + normal/high free = slow clearance (hypothyroid); CAR on the Plus (blunted → HPA hypo-responsiveness; exaggerated → anticipatory stress; check sampling compliance). Never read "adrenal output" off free cortisol alone. Add DHEA-S with metabolites, age-adjusted.
3. Androgens. Testosterone with 5α/5β preference: 5α shift (elevated 5α-DHT, androsterone > etiocholanolone) = androgenic tissue activity — correlate with documented PCOS-pattern symptoms, hair loss, acne even when total testosterone is normal. Low androgens: correlate with fatigue/libido and DHEA status.
4. Estrogens — quantity, then routing. Total output (E1/E2/E3) against age/menopausal status. Phase I proportions: 2-OH (preferred) vs 4-OH (DNA-reactive quinone potential — the one to watch) vs 16-OH (proliferative in excess). Phase II: 2-methoxy-E1/2-OH-E1 ratio as COMT/methylation activity — low methylation with high catechol estrogens is the classic support target. Frame 4-OH elevation as a modifiable-risk consideration, never a cancer prediction. Cross-reference β-glucuronidase on any GI-MAP in the documents (estrogen recirculation).
5. Progesterone. α- and β-pregnanediol (luteal collection only): adequacy vs estradiol (Pg/E2 balance) for cycle symptoms, sleep, perimenopausal picture.
6. Organic acids on the panel. 8-OHdG (oxidative DNA stress); melatonin 6-OHMS vs documented sleep complaints; vanilmandelate/homovanillate (catecholamine turnover vs stress picture); kynurenate/xanthurenate (B6), methylmalonate (B12), pyroglutamate (glutathione) as cofactor context.
7. Synthesis. Tie axes together and to the chart: flattened cortisol + low progesterone in perimenopause; high stress output + 5α preference in a PCOS picture; low methylation + high 4-OH + estrogen-dominance symptoms. Urine metabolites measure production-and-metabolism, not serum levels — where a decision needs serum (dosing, fertility), say so. Compare against any prior DUTCH for trend.

Red flags — escalate regardless of functional interpretation

Not exhaustive. Postmenopausal bleeding or markedly elevated estrogens in a postmenopausal patient; rapidly virilizing signs or extreme androgen elevations (rule out androgen-secreting tumor); cortisol extremes with corresponding clinical features (Cushingoid features, or suspected adrenal insufficiency — fatigue with hypotension, weight loss, hyperpigmentation) need conventional endocrine evaluation — DUTCH is not the diagnostic instrument for either; personal history of hormone-sensitive cancer — any hormone-related consideration should be coordinated with the treating oncology team. Put these first in the output.

Output format

1. Snapshot — one paragraph: dominant pattern and the single most important finding.
2. Validity notes — anything limiting interpretation (contraceptives, timing, HRT route), stated first.
3. Findings table — Axis | Marker | Result | Lab range | Flag | Note (cortisol rhythm; production vs clearance; androgens; estrogens + routing; progesterone; organic acids).
4. Pattern synthesis — the story across axes, tied to documented symptoms.
5. Red flags — or an explicit "none identified against the checked list".
6. Considerations for the practitioner — confirmatory serum testing where relevant, retest timing (typically 3–4 months after intervention), HPA/lifestyle and methylation-support directions as options, never directives.
7. Patient-friendly summary — only if asked.

If the practitioner wants it saved, save with create_note (a lab review is not a clinical note or care plan).

End with: "Decision support only — for review by the treating clinician."

Evidence & sources

Ranges are Precision Analytical's published, population- and cycle-phase-specific reference sets — the skill's first job is verifying results were read against the right set. The underlying biochemistry (2-OH/4-OH/16-OH estrogen pathways, COMT methylation, 4-OH quinone DNA reactivity, 5α/5β reduction, cortisol production vs clearance) is established literature; the cortisol awakening response has an independent salivary-research base. Honest limitations stated in the body: independent clinical validation of dried-urine hormone testing is limited and serum remains the endocrinology standard — the skill defers to serum for dosing and fertility decisions, restricts conclusions on hormonal contraceptives, excludes pregnancy (not validated), and forbids framing 4-OH elevation as cancer prediction. Escalation triggers (postmenopausal bleeding, virilization, Cushingoid/adrenal-insufficiency features) follow conventional endocrine practice.

Changelog

• 1.0.1Added reference-population verification to the validity gate and threshold provenance.
• 1.0.0Initial version.

Patient data on each platform

De-identification means removing all 18 HIPAA identifiers, not just the name. Read the full disclaimer.