Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
What is the absolute risk reduction and NNT for statins in secondary prevention?
Executive summary
Randomized trials and pooled analyses show a substantially larger absolute benefit from statins in secondary prevention (people with known cardiovascular disease) than in primary prevention: pooled secondary-prevention ARR about 4.8% over 5 years (NNT ≈ 21) and trial-specific examples such as pravastatin showing ARR 4.3% (NNT 23) over ~3.2 years; total mortality ARR in secondary-prevention pools ≈2.1% (NNT ≈48) for 3–5 years [1] [2] [3]. Available sources do not mention every possible subgroup or every modern high‑intensity regimen; estimates vary by baseline risk, endpoint chosen, follow‑up time, and trial mix [1] [4] [5].
1. Secondary prevention: the headline ARRs and NNTs
Major secondary‑prevention trial pools reported absolute risk reductions (ARRs) large enough to produce low NNTs: a pooled ARR of 4.8% for "total myocardial infarction or total stroke" over 5 years corresponds to an NNT ≈ 21 to prevent one such event [1]. Individual trials within the secondary‑prevention literature give similar magnitudes — for example, pravastatin 40 mg in elderly secondary‑prevention patients reduced the combined MI/stroke outcome with ARR 4.3%, NNT 23 over about 3.2 years [2] [3]. Those figures are the clearest, repeatedly cited metrics in the provided sources [1] [2].
2. Mortality benefit in secondary prevention: measurable but smaller
Pooled data from secondary‑prevention trials show a reduction in total mortality with an ARR of roughly 2.1% over 3–5 years, giving an NNT around 48 to prevent one death [2] [3]. That mortality benefit is smaller than the ARR for nonfatal MI/stroke but is still statistically and clinically meaningful in these higher‑risk populations, according to the Therapeutics Letter summaries [2] [3].
3. Why absolute numbers vary: baseline risk, endpoints and follow‑up
ARR and NNT depend on the baseline event rate (higher baseline risk → larger ARR for a given relative risk reduction), which explains why secondary‑prevention NNTs are much lower than many primary‑prevention estimates. Different trials use different endpoints (nonfatal MI, stroke, composite cardiovascular events, total mortality) and different follow‑up durations; pooled secondary data quoted here use 3–5 years or 5‑year windows and are specific to the trials included [1] [2] [3]. Method guides and reviews caution that NNTs must be interpreted in light of these parameters [5].
4. Comparison with primary prevention — the contrast matters
By contrast, pooled primary‑prevention ARRs are much smaller. The Therapeutics Letter and other syntheses cited here report primary‑prevention ARRs often in the 1–2% range over similar timeframes, leading to NNTs typically much higher (examples: primary prevention NNT ≈50 for some coronary outcomes over 3–5 years; other reviews show NNTs in the tens to hundreds depending on subgroup and endpoint) [1] [6] [7]. The practical implication is that patients with established CVD derive a far larger absolute benefit from statins than average-risk people without prior events [1] [7].
5. Heterogeneity, dosing and “real-world” considerations
Sources note heterogeneity across trials and call for large RCTs to test dosing strategies in secondary prevention; the Heart Protection Study and other large trials contribute to pooled estimates but mixes of doses and patient types matter [3]. Reviews also emphasize that industry funding and trial mix can shape event rates and reported effects, and that NNT depends on the comparator, adherence, and which events are counted [4] [5].
6. How clinicians and patients should read these numbers
NNTs are practical but context‑dependent: an NNT of ~21 over 5 years for preventing MI/stroke in secondary prevention means treating ~21 similar patients for that time will prevent one such event; an NNT ≈48 for all‑cause death over 3–5 years communicates a smaller but tangible mortality benefit [1] [2]. Shared decision‑making should use the specific patient’s baseline risk, desired endpoints (symptoms vs death vs nonfatal events), and potential harms — sources on primary prevention emphasize much higher NNTs and the need to weigh small absolute benefits against side‑effect risks [7] [6].
7. Limits of the available reporting and open questions
The provided sources do not cover every contemporary high‑intensity statin regimen or fully stratify modern subgroups (e.g., extreme LDL targets, very old patients, or combined lipid therapies), and they note missing reporting of total serious adverse events in some trials [3] [2]. For up-to-date, regimen‑specific ARRs and NNTs one should consult recent meta‑analyses and guideline appendices that present stratified absolute risks; those are not included in the supplied excerpted sources (not found in current reporting).
If you want, I can extract specific trial ARRs/NNTs from the large trials cited (4S, CARE, LIPID, HPS, PROSPER) and tabulate them with follow‑up times and endpoints using the sources you prefer.