Trust Center
What CurieSense can and cannot tell you — methods, limits, reviews, and data handling in one place.
Published clinical measures
These outputs follow published methods and guideline thresholds exactly.
- Phenotypic Age — Levine et al. 2018 chemistry clock, with the MCV/RDW-normalized counterfactual shown alongside.
- FIB-4 — AASLD-guideline liver fibrosis estimate with its low/indeterminate/high bands.
- eGFR — CKD-EPI 2021 race-free equation; KDIGO staging applies only with persistence over time plus urine testing.
- ASCVD proxy — ACC/AHA lipid thresholds adapted to the panel. It is not a 10-year cardiovascular risk estimate: blood pressure, smoking, family history, and treatment context are missing.
CurieSense estimates
These are product constructs built on top of the measures above — useful for tracking, not clinical standards.
- Panel score (0–100) — weighted by marker importance, graded by zone, docked for computed risks.
- Organ scores (0–100) — per-system averages over the markers you submitted; thin coverage means wide uncertainty.
- Recommendation grades — strength of the underlying evidence, not a priority order.
- PhenoAge drivers — model attribution of years per marker, not a causal aging ranking.
- Change-vs-previous-report — marker-by-marker comparison with regression-to-the-mean shrinkage, only within your own account and only against your previous snapshot.
Limitations
Every output below inherits these limits; the result page repeats the ones that matter per report.
- One draw cannot diagnose — hydration, illness, and time of day move markers.
- Adults 18+ with routine, non-acute results only — no pediatric, pregnancy-specific, emergency, or time-sensitive use.
- Partial panels limit outputs — estimates show 'not enough data' instead of guessing.
- Values outside urgent-care bounds route to prompt-care guidance with routine advice withheld. Responsible owner for urgent values: your clinician — CurieSense flags, it does not triage.
Clinical review status
CurieSense is built by a technology company with no in-house clinicians, so no recommendation here carries our own clinical sign-off. Interim safety wording is in place (clinician-discussion language, categories, no medication-stopping instructions, no claims about unmeasured markers) — and you must review these recommendations with your own physician, registered dietitian, and pharmacist before acting on them.
Last reviewed: no in-house clinical review completed. Rules are reworded the moment evidence or safety guidance changes; the review matrix (docs/RECOMMENDATION_REVIEW.md) tracks every rule.
Analysis method v1. Unchanged panels always reproduce identical results; the version is stamped on every snapshot and export, and a bump is disclosed rather than silently recomputed.
Conflicts of interest
Conflicts of interest: none to disclose. CurieSense sells no lab tests, takes no referral fees, and Plus is not for sale — no paid placement exists in recommendations.
Regulatory posture: an educational product, not medical advice or diagnosis. Jurisdiction-appropriate regulatory review is pending before broad launch.
Privacy and deletion
Only extracted biomarker values are stored — uploaded PDFs are parsed in memory and never kept. First-party usage events (reports, saves, action progress, summary and pricing views) measure whether the product works and are never sold. Snapshots stay until you delete them; deleting your account deletes your health data. We keep no separate product-level backup.
Support
Questions about your report or your account — chat with us in the livechat on this site, or email [email protected]. Anything urgent or medical belongs with your physician, not email.
Recommendation sources
Every recommendation cites its sources. Read them before acting — and review them with your clinicians.
- Cut LDL + ApoB with soluble fiber + plant sterols + low sat-fat swap
- Discuss high triglycerides with your clinician: refined carbs, sugar, alcohol, omega-3, cardio
- Discuss liver markers and possible liver fat with your clinician: 7-10% weight-loss program + Mediterranean pattern
- Work up isolated indirect hyperbilirubinemia (likely Gilbert vs hemolysis)
- Protect kidneys: hydration, BP/salt check, NSAID + protein audit
- Lower uric acid: fructose/alcohol purge + weight + hydration
- Macrocytosis workup (MCV high): folate, alcohol, thyroid, liver, meds
- Eosinophilia: allergy/parasite/drug screen
- High-normal TSH: recheck + anti-TPO, iodine/selenium audit
- Lock in glycemic win: keep fiber-first meal order + steps
- hs-CRP good: preserve with sleep + oral health + Zone-2
- Hold vitamin D: maintain, do not megadose
- ApoB + non-HDL high: discuss statin threshold with doctor
CurieSense is an educational research tool, not a medical diagnosis. Always review results with your physician.