O-47 - 33-cohort meta-analysis- no overall egg-CVD association (RR 0.98 per +1 egg-day), but I2=62.3% heterogeneity
The overall meta-analytic null across the world literature, resting on the 33-cohort aggregate corpus (D-26) rather than the Harvard cohorts alone. Substantial heterogeneity flags that the pooled point estimate is not a single homogeneous effect.
Link to original
Why this is evidence
The pooled point estimate (RR 0.98/egg-day) sits on the null, which H-21 (no causal effect) predicts and which is inconsistent with H-27 (monotonic harm) and H-6 (protection). But the large I2=62.3% is itself diagnostic for the residual H-49: a genuinely single population-wide sign should pool cleanly, whereas high between-study heterogeneity around a null is what a subgroup/region-dependent sign (H-49) predicts. H-44 (U-shape) is only weakly separated here since a per-egg linear meta-slope cannot detect curvature.