Cross-signal correlations

Insights

Advanced Labs looks for patterns where a lab biomarker moves alongside a wearable trend. Every card states the strength, the direction and the window so you can judge how much weight the pattern deserves.

Informational, not diagnostic

These insights describe statistical co-movement in your own data. They are not a medical opinion, do not diagnose or treat any condition, and should not be used to start, stop or change medication. Bring anything that concerns you to a qualified clinician.

5 associations surfaced

Ranked by association strength across your available history

Open lab detail

Cortisol, AMRolling 7-day strain
Your highest strain block preceded the elevated morning cortisol draw.
Out of range
Association
Moderate–strong
Direction
positive
Window
Last 45 days
Rolling 7-day strain, last 7 daysr ≈ 0.72
FerritinRecovery score
Lower stored iron lines up with your weaker recovery weeks.
Borderline
Association
Moderate–strong
Direction
positive
Window
Last 90 days
Recovery score, last 7 daysr ≈ 0.68
Across the 12 weeks since your May draw, weeks where recovery averaged under 55% clustered in the same period ferritin fell from 41 to 28 ng/mL.
Worth exploring: Worth reviewing dietary iron and absorption context with a clinician before changing supplementation.
This is an observed association in a short synthetic series, not a causal finding. It cannot confirm, rule out or explain any medical condition.
Creatine kinaseWorkout strain spikes
Creatine kinase peaked two days after your heaviest logged session.
Out of range
Association
Moderate
Direction
positive
Window
Last 30 days
Workout strain spikes, last 7 daysr ≈ 0.61
hs-CRPSleep performance
Higher inflammation coincides with your shortest sleep stretches.
Borderline
Association
Moderate
Direction
inverse
Window
Last 60 days
Sleep performance, last 7 daysr ≈ 0.54
Magnesium, RBCSleep efficiency
RBC magnesium improved alongside a modest gain in sleep efficiency.
Optimal
Association
Weak
Direction
positive
Window
Last 90 days
Sleep efficiency, last 7 daysr ≈ 0.41

How these are built

Method, in plain terms

Step 1
Align the timelines
Each biomarker result is anchored to its collection date, then compared with the wearable window leading up to and following that date.
Step 2
Score the co-movement
A rank correlation is computed across the available points. Short series produce weak, unstable coefficients and are labelled as such.
Step 3
Explain, do not prescribe
Cards describe what moved together and suggest what to observe next. They never assert cause, and never recommend treatment.

Association quality improves with more wearable history. Connecting a wearable widens the window from a 7-day sample series to your full authorized record, imported server-side after OAuth consent.