Different signals operate on different timescales
Resting heart rate or HRV can move substantially within days. Blood pressure, body mass, fasting glucose, and laboratory biomarkers often change on slower timescales and may be measured less frequently.
A longitudinal system therefore faces a design choice: ignore slower markers on days without a new record, or preserve the last observation while making its age explicit.
Clinical biological-age models already use routine biomarkers
Established approaches such as KDM and Levine PhenoAge (clinical) combine multiple clinical biomarkers because no single marker adequately represents systemic aging or health risk.
Studies applying blood-chemistry biological-age algorithms have reported associations with disability and mortality, supporting the value of multisystem clinical information while also underscoring that model performance depends on the population and endpoint.
Recency is part of the evidence
Carrying a prior observation forward can preserve continuity, but a two-day-old blood-pressure observation and a six-month-old observation should not look equally current. The date of the source observation is therefore part of the interpretation.
This distinction matters particularly when a composite mixes high-frequency wearable data with episodic home or laboratory data.
BioMIR’s cardiometabolic anchors
BioMIR uses systolic blood pressure, a glucose anchor, and BMI in the daily architecture. Explicitly fasting clinical glucose is preferred; fallback glucose does not establish fasting status. A prior eligible value can persist with its original date. CMA Freshness tracks recency across the independently dated anchors.