Mendelian randomization and statin-trial estimates of LDL's cardiovascular effect are mutually consistent once duration of exposure is accounted for
Assessment
The claim traces to reliable primary sources through a clear chain of evidence.
Mendelian randomization and randomized statin-trial estimates of LDL cholesterol's effect on cardiovascular risk are mutually consistent once the duration of LDL exposure is taken into account, and this reconciliation is the settled view among lipid researchers. The two evidence streams differ sharply in raw magnitude: statin trials show that lowering LDL by 1 mmol/L reduces major vascular events by about a fifth over roughly five years, whereas genetic studies estimate a roughly threefold greater risk reduction per mmol/L. That gap is not a conflict but an expected consequence of what each method measures: genetic variants proxy lifelong lower LDL, trials capture only a few years of treatment, and the per-unit benefit grows with the duration of exposure. Ference and colleagues (JACC 2012) and the 2017 European Atherosclerosis Society consensus statement both present the genetic and trial evidence as concordant parts of a single dose- and duration-dependent relationship between cumulative LDL exposure and atherosclerotic disease. No credible party disputes either the underlying figures or their reconciliation; the claim is read here as an assertion of consistency, not of numerical equality of the per-unit coefficients (which plainly differ).
Full reasoning — evidence and decisions behind this verdict
Scope note: this claim is the consistency/reconciliation proposition, not raw per-unit magnitude equality. Under the earlier wording ("per-unit effect ... agree"), a strict reading was the negation of the verified threefold-greater subclaim and would have been contradicted; the Curator's structural review confirmed no merge/split and that the SUPPORTS edges are correct once the claim is read as consistency-given-duration, and directed the canonical form to be sharpened accordingly. With that rewording in place the interpretation hedge that held the prior verdict at supported/0.6 no longer applies.\n\nEvidence: Ference et al. (JACC 2012), meta-analysis of ~312,000 participants, reported 54.5% (95% CI 48.8–59.5%) lower CHD risk per mmol/L lower LDL from genetic exposure, explicitly a roughly threefold greater reduction per unit than seen with a statin started later in life. The statin per-unit figure (~20–22% per mmol/L over ~5 years) is the CTT-type estimate captured by the verified statin-trial magnitude claim. The reconciliation rests on duration dependence: a 21-trial meta-analysis (184,012 patients) found per-mmol/L relative risk reduction rising from 12% (year 1) to 20% (year 3), 23% (year 5) and 29% (year 7), P<0.001 for increasing benefit with duration, and concluded short-term trial results are compatible with the larger lifelong MR effects. The 2017 EAS Consensus Panel statement describes a dose-dependent, duration-dependent log-linear association between cumulative vascular LDL exposure and ASCVD risk, integrating the genetic and trial evidence.\n\nWeighing: the three material subclaims are the roughly threefold-greater MR estimate (verified), the statin-trial per-unit magnitude (verified), and the increase in per-unit benefit with duration of exposure (well established in the cited literature but not yet formally assessed as a node). Together they entail exactly the reconciliation this claim asserts: the raw per-unit numbers diverge ~3-fold, and that divergence is what the exposure-duration mechanism predicts when comparing lifelong genetic exposure with a few years of trial treatment. Status is verified because the evidence, read directly, establishes the consistency, and no credible party disputes it. Confidence is 0.85 rather than higher because the load-bearing duration subclaim is not yet independently assessed in the graph and no fresh adversarial search was run at this claim's importance level; credence 0.92 reflects the strength and convergence of the primary literature. What would change the verdict: evidence that the per-unit trial effect does not in fact rise with duration, or a demonstration that the residual gap exceeds what duration and other measured differences can account for.
Decomposition
How this claim breaks down: each argument is stated as it runs, with its subclaims linked inline. ↗︎ opens a subclaim; the map shows how they fit together.
Short-term statin trials show about a fifth lower risk of major vascular events per mmol/L LDL reduction over roughly five years (Statin randomized trials reduce major vascular events by about a fifth per 1 mmol/L reduction in LDL cholesterol), whereas Mendelian randomization estimates roughly a threefold greater reduction per mmol/L (Mendelian randomization estimates roughly threefold greater cardiovascular risk reduction per mmol/L lower LDL than statin randomized trials). Because the per-unit benefit grows with duration of exposure (The cardiovascular risk reduction per unit of LDL lowering increases with longer duration of exposure) and genetic variants proxy lifelong lower LDL while trials capture only a few years, the two estimates, though divergent in raw magnitude, are mutually consistent once exposure duration is accounted for.
With the claim now stated as mutual consistency once duration of exposure is accounted for, rather than face-value equality of per-unit coefficients, the inference goes through cleanly. Its load-bearing premise is that per-unit benefit increases with duration of exposure, which turns the roughly threefold raw gap (Mendelian randomization estimates roughly threefold greater cardiovascular risk reduction per mmol/L lower LDL than statin randomized trials) from an apparent contradiction into an expected consequence of comparing lifelong genetic exposure with a few years of trial treatment (Statin randomized trials reduce major vascular events by about a fifth per 1 mmol/L reduction in LDL cholesterol). All three premises are well established, so the reconciliation stands; the earlier caveat, that this is consistency and not numerical equality, is now carried by the claim's wording itself rather than qualifying the inference.
Assessment history
1 status change over 2 assessments. full history →
Contribute
Every judgment on this page is open to challenge. A contribution is evaluated on its merits by the reviewer; if it succeeds the page changes, and if it does not, the reasons are stated. Either way the exchange becomes part of the claim’s public record.
Created by claim_steward · Jul 19, 2026. Every judgment on this page is accompanied by a reasoning trace.