Ehvidence
Favorable Efficacy (RCT)

Statins for primary prevention

Statin therapy vs. Placebo or usual care · for Cardiovascular disease (primary prevention) · real-time analysis of 7 studies · updated 2026-05-29

In people without established cardiovascular disease, statins reduce major cardiovascular events, myocardial infarction, stroke, and modestly reduce all-cause mortality. Because baseline risk is lower than in secondary prevention, the relative reductions translate into large numbers-needed-to-treat over a few years.

Efficacy (RCT): Efficacy evidence on clinical outcomes. Effects are risk ratios with number-needed-to-treat where a baseline risk is available. Glossary →

Interpretation & tips

These are RCTs, so NNTs are meaningful—but large, because baseline risk is low; read them with their ~2–6 year horizon. Relative reductions are bigger for events (MI, stroke) than for all-cause mortality. The trials shown are a selection of landmark studies and run somewhat stronger than the comprehensive USPSTF meta-analysis (mortality ~17% here vs ~8%); treat USPSTF as the more reliable summary.

34% lower risk of major cardiovascular events
Pooled RR 0.66 (95% CI 0.60–0.73) across 6 studies, 61,844 patients (random-effects, I² 22%).
0.66
Pooled RR
0.60–0.73
95% CI
6
Studies
61,844
Patients
65
NNT

Forest plot—Major cardiovascular events

USPSTF 2022 (pooled) 2022 0.72 [0.64–0.81] HOPE-3 (Yusuf) 2016 0.77 [0.65–0.91] JUPITER (Ridker) 2008 0.57 [0.46–0.69] MEGA (Nakamura) 2006 0.67 [0.49–0.91] ASCOT-LLA (Sever) 2003 0.65 [0.50–0.83] AFCAPS/TexCAPS (Downs) 1998 0.62 [0.53–0.73] WOSCOPS (Shepherd) 1995 0.70 [0.58–0.84] Pooled (RE) 0.66 [0.60–0.73] 0.1 0.25 0.5 1 2 4 ← favors treatment favors control →

Study results—Major cardiovascular events

Study Design Dose / regimen Treatment Control RR [95% CI] Improvement NNT Weight
USPSTF 2022 (pooled) 2022 excl note review Statins (various) — — 0.72 [0.64–0.81] 28% 78 —
HOPE-3 (Yusuf) 2016 note DB-RCT Rosuvastatin 10 mg 235/6361 304/6344 0.77 [0.65–0.91] 23% 92 22%
JUPITER (Ridker) 2008 note DB-RCT Rosuvastatin 20 mg 142/8901 251/8901 0.57 [0.46–0.69] 43% 82 16%
MEGA (Nakamura) 2006 note RCT Pravastatin 10-20 mg + diet 66/3866 101/3966 0.67 [0.49–0.91] 33% 120 8%
ASCOT-LLA (Sever) 2003 note DB-RCT Atorvastatin 10 mg 100/5168 154/5137 0.65 [0.50–0.83] 35% 95 12%
AFCAPS/TexCAPS (Downs) 1998 note DB-RCT Lovastatin 20-40 mg 224/3304 359/3301 0.62 [0.53–0.73] 38% 25 23%
WOSCOPS (Shepherd) 1995 note DB-RCT Pravastatin 40 mg 174/3302 248/3293 0.70 [0.58–0.84] 30% 45 18%

RR < 1 favors treatment for outcomes where lower is better. Rows in gray have a confidence interval crossing 1 (individually inconclusive). “~” marks effects reported as OR/HR and treated as RR-approximations. “excl” = excluded from pooling (e.g. reviews).

Notes & interpretation

  • USPSTF 2022 (pooled) 2022—Pooled composite CV outcome; shown for reference and excluded from pooling to avoid double-counting. USPSTF NNT ≈ 78.
  • HOPE-3 (Yusuf) 2016—First co-primary composite (CV death, nonfatal MI, nonfatal stroke). Published HR 0.76 (0.64-0.91).
  • JUPITER (Ridker) 2008—Primary composite (nonfatal MI/stroke, revascularization, unstable-angina hospitalization, CV death). Stopped early. Published HR 0.56 (0.46-0.69).
  • MEGA (Nakamura) 2006—CHD primary endpoint. Open-label (PROBE) design. Published HR 0.67 (0.49-0.91).
  • ASCOT-LLA (Sever) 2003—Primary endpoint: nonfatal MI + fatal CHD. Stopped early. Published HR 0.64 (0.50-0.83).
  • AFCAPS/TexCAPS (Downs) 1998—First acute major coronary event. Counts from a secondary summary—verify against the original JAMA table.
  • WOSCOPS (Shepherd) 1995—Primary endpoint: definite nonfatal MI or CHD death.

Background

Primary prevention treats people without established cardiovascular disease. Placebo-controlled trials (WOSCOPS, AFCAPS, ASCOT-LLA, JUPITER, MEGA, HOPE-3) and pooled analyses (USPSTF, Cochrane) evaluated statins here. Effects are computed from each trial's event counts; published hazard ratios are very close.

Topic methodology & caveats

Risk ratios are computed from each trial's event counts (2×2); published hazard ratios are very close. The USPSTF 2022 pooled estimate is shown per outcome for reference and excluded from pooling. AFCAPS counts are from a secondary summary; MEGA was open-label.

Studies