Among adults in the All of Us cohort who already met standard cardiovascular prevention targets — systolic BP below 140 mmHg, LDL below 100 mg/dL, or A1c below 8% — a composite social context score (combining household income, education, food security, physical activity, and neighbourhood deprivation) strongly predicted subsequent heart attack, ischemic stroke, or heart failure. Each one-standard-deviation improvement in the social score was associated with hazard reductions of 39%, 37%, and 46% across the three cohorts (HRs 0.611, 0.631, and 0.544). Event rates among those with less favourable social context were roughly double to quadruple those with more favourable context: for LDL, 7.16 versus 2.05 per 1,000 person-years; for A1c, 8.67 versus 2.34.
This finding matters because it directly challenges the clinical assumption that a biomarker within range signals uniform safety. The magnitude of the social gradient rivals or exceeds many pharmacological risk reductions seen in landmark statin and antihypertensive trials. For adults, this suggests that lifestyle, economic stability, and neighbourhood environment carry independent cardiovascular weight that a lab value cannot capture — worth raising with clinicians. Limitations are real: this is observational and retrospective, event counts are modest in primary cohorts (129–191 events), and residual confounding is possible. Crucially, this is a preprint posted on medRxiv and has not yet been peer-reviewed — effect sizes and conclusions may shift. Still, the consistency across three independent biomarker cohorts makes this more than incremental.