Trace the construct.
Read the evidence.

BioMIR reads the Health records you authorize. Set up each compatible device in its manufacturer or companion app, allow that app to write the relevant measurements to Apple Health, then authorize BioMIR to read those Health categories. Pairing a device to the iPhone is not enough; the measurement must actually appear in Health before BioMIR can use it.

To verify a stream, open the measurement in the Health app and review Data Sources & Access. Apple shows which apps and devices contribute to that data type and lets you review source priority.

Connectivity does not establish measurement validity. Device class, acquisition protocol, calibration, algorithm, wear conditions, and user technique can materially affect the record. For blood pressure, use an appropriately authorized, independently validated upper-arm device with correct cuff size and standardized technique when measurements may inform clinical care.

Data Confidence (shown as Completeness in ABA Trends) summarizes how much of the expected daily input information is available to support the estimate. It is not a probability that the estimate is correct, a measure of device accuracy, or diagnostic certainty.

CMA Freshness summarizes how recently BMI, systolic blood pressure, and glucose were observed. Carry-forward preserves the latest eligible value and its original observation date; it does not create a new measurement.

High overall Completeness can therefore coexist with older cardiometabolic anchors. Neither measure captures every source of uncertainty, including device error, reference-population mismatch, irregular sampling, or model misspecification.

Chronological age is elapsed time. A biological-age model summarizes selected measurements using a reference population and a statistical method. Different panels can represent different dimensions of physiology, so disagreement between clocks is not inherently contradictory.

The Klemera–Doubal Method combines age-related information across biomarkers while accounting for statistical variation. It provides a methodological framework for age-equivalent physiological modeling rather than a direct measurement of remaining lifespan.

Clinical PhenoAge combines chronological age with nine laboratory biomarkers in an outcome-linked model. DNAm PhenoAge is the related methylation model derived from the clinical phenotype; the two should not be conflated.

BioMIR’s clinical view uses the laboratory-based form. Collection date and underlying laboratory values remain essential when comparing this episodic clock with daily wearable or behavioral data.

The BioAge methods literature shows that biological-age estimates depend on biomarker selection, reference data, parameterization, and statistical objective. Implementing an established framework does not transfer every published performance claim to a new panel, app, population, or composite.

Evaluation of BioMIR’s exact implementation therefore requires independent calibration, reliability, subgroup performance, sensitivity analysis, and outcome validation beyond component-level evidence.

Allostasis describes coordinated physiological adjustment to changing demands. Allostatic-load research operationalizes cumulative multisystem dysregulation using heterogeneous biomarker panels; there is no single standardized clinical allostatic-load test.

BioMIR’s Recovery domain uses HRV, resting heart rate, sleep, and aerobic-fitness signals as partial, non-equivalent views of physiological response and reserve. It should not be interpreted as a direct measurement of global allostatic load.

Prospective and meta-analytic evidence relates sleep duration and regularity to health outcomes at the population level. These associations do not establish how changing one individual’s sleep will change lifespan.

Consumer sleep stages are algorithmic estimates rather than polysomnographic measurements. For BioMIR, repeated within-device trends are generally more interpretable than treating a single estimated stage duration as an absolute physiological quantity.

Human metabolic health depends on the capacity to adapt fuel use to nutrient availability and energetic demand. Dietary patterns can alter glycemic exposure, lipid metabolism, energy balance, and substrate use, but mechanistic pathway language should not be mistaken for demonstrated human longevity effects.

Randomized-trial meta-analysis indicates that intermittent-fasting strategies generally produce cardiometabolic benefits similar to continuous energy restriction, with only modest differences between strategies. Umbrella evidence for ketogenic diets supports selected short-term effects on body weight, triglycerides, and HbA1c in some populations while also identifying clinically meaningful LDL-cholesterol increases in some settings. High-quality carbohydrate evidence favors fiber-rich, minimally processed sources over simplistic high-versus-low carbohydrate framing.

Accordingly, BioMIR treats dietary energy and carbohydrate as exposure signals whose interpretation depends on food quality, activity, adiposity, glucose regulation, and longitudinal context. AMPK, mTOR, ketosis, or the label “fasting-mimetic” are mechanistic descriptors—not sufficient evidence for a favorable model weight or an anti-aging claim.

Common cardiometabolic and laboratory traits are polygenic. Large biobank studies demonstrate substantial genetic architecture across clinical biomarkers, but genotype does not replace measured phenotype, longitudinal trajectory, or clinical interpretation.

BioMIR’s future genomic roadmap is therefore framed around carefully curated high-value variants and, where stronger evidence exists, validated polygenic scores that could refine selected population baselines or model modifiers. Baseline shifts and weight modifiers are distinct operations: a SNP association or disease odds ratio should not be copied directly into a biomarker coefficient.

Any genomic implementation should be versioned, provenance-preserving, ancestry-aware, calibrated to the target population, and validated out of sample. Portability limitations and uncertainty must remain explicit. The feature is not intended to infer genetic diagnoses, carrier status, or treatment recommendations.