Hypothesis

We propose that increased short-term variability in wearable-derived heart rate variability (HRV) and intermittent spikes in high-sensitivity C-reactive protein (hs-CRP) precede measurable acceleration in DunedinPACE epigenetic aging rate, and that reducing this variability through targeted stress‑reduction interventions will normalize DunedinPACE trajectories.

Mechanistic Rationale

Chronic autonomic imbalance elevates sympathetic tone, lowering HRV and triggering transient inflammatory bursts [2]. These bursts activate NF‑κB signaling, which can disturb DNA methylation maintenance at CpG sites linked to metabolic and immune function, thereby increasing the pace of epigenetic aging measured by DunedinPACE [4]. Thus, the noise traditionally dismissed in biomarker time series may actually be a causal driver of accelerated aging rather than mere measurement error.

Testable Predictions

1. Within an individual, weeks where the coefficient of variation (CV) of nightly HRV exceeds the personal baseline by >20% will be followed, after a 1‑2 week lag, by a rise in weekly averaged hs-CRP >1.5 mg/L.
2. Such combined HRV‑CRP excursions will predict a >0.05 increase in DunedinPACE change over the subsequent month, relative to periods of stable HRV and CRP.
3. A 4‑week mindfulness‑based stress reduction (MBSR) protocol that lowers HRV CV by ≥15% will attenuate hs-CRP spikes and halt the DunedinPACE acceleration observed in prediction 2.

Experimental Design

• Recruit 20 healthy adults aged 30‑50 with baseline DunedinPACE measured from blood [3].
• Equip each with a chest‑strap HRV logger sampling at 5 min intervals for 24 h/day.
• Collect finger‑stick blood weekly for hs-CRP and monthly for DunedinPACE (using the DunedinPACE algorithm) over 6 months.
• Compute nightly HRV CV (SD/mean) per week; flag weeks exceeding personal mean + 20%.
• Use time‑lagged cross‑correlation to test prediction 1 and linear mixed‑effects models for prediction 2 (fixed effect = HRV‑CRP excursion, random intercept = participant).
• After month 3, randomize participants to MBSR (n=10) or wait‑list control (n=10) for 4 weeks; resume tracking for another 8 weeks to assess prediction 3.

Potential Confounds

• Acute infections can spike hs‑CRP independently of HRV; we will exclude weeks with self‑reported fever or antibiotic use.
• Medication changes affecting heart rate (e.g., beta‑blockers) will be recorded and used as covariates.
• Seasonal variation in activity may influence HRV; we will include ambient temperature as a covariate.

By linking high‑frequency physiological noise to epigenetic aging pace, this hypothesis turns a major limitation of quantified‑self research into a actionable biomarker. If falsified, it would reinforce the view that short‑term variability is largely measurement noise, sharpening focus on longer‑term trends for health‑span optimization.

References [1] https://journals.plos.org/digitalhealth/article/file?id=10.1371%2Fjournal.pdig.0001271&type=printable [2] https://formative.jmir.org/2024/1/e48783 [3] https://optimalhealth.co/resources/blood-testing/biomarkers-longevity [4] https://holisticare.io/blog/longevity-medicine-2026-strategy/ [5] https://www.tandfonline.com/doi/full/10.1080/02673843.2025.2590907 [6] https://nursing.jmir.org/2023/1/e50991/ [7] https://pmc.ncbi.nlm.nih.gov/articles/PMC12505433/