How effective are statins at preventing first heart attacks and strokes?

The short version: for a person without prior heart disease or stroke, the evidence shows a small absolute benefit from statins, not a dramatic one, and the benefit is clearest in people with elevated inflammation or higher calculated risk.

The best evidence for primary prevention comes from the JUPITER trial, which tested rosuvastatin in people with normal LDL but elevated C-reactive protein (a marker of inflammation). That trial found a 20% relative reduction in all-cause mortality, but the absolute reduction was small: about 0.8% fewer deaths over roughly two years [3]. The HOPE-3 trial, which tested rosuvastatin in intermediate-risk people without elevated CRP, found no difference in all-cause mortality at all [3]. A large meta-analysis of 21 trials put the absolute mortality reduction at 0.8% and the relative reduction at 9% [3]. That means about 125 people need to take a statin for five years to prevent one death.

The benefit for preventing first heart attacks is somewhat larger. The same meta-analysis found statins reduced nonfatal heart attacks by about 34% relative, which translates to about 2 fewer heart attacks per 100 people treated [1]. For stroke, the reduction was smaller and less consistent: a 1999 meta-analysis found a 23% relative reduction in fatal stroke that did not reach statistical significance, and a 31% relative reduction in nonfatal stroke that did [1]. The number needed to treat to prevent one nonfatal stroke was about 143 over the trial periods [1].

The evidence is dominated by industry-funded trials that compared statins to placebo, not to diet or lifestyle changes. One study found that people who followed four or more healthy lifestyle factors had an ASCVD incidence rate of 8.7 per 1000 person-years, compared to 13.4 for those with none [2]. That lifestyle effect is in the same range as the statin effect, and the two are additive, not substitutes. The JUPITER trial was stopped early once benefit appeared, which tends to overestimate effect size [3]. The meta-analysis also found little association between how much LDL dropped and how much mortality fell, which undermines the simple "lower LDL = lower risk" story [3].

My call: statins reduce first heart attacks and strokes by a modest amount in people with elevated risk or inflammation, but the absolute benefit is small and the evidence does not support treating everyone regardless of risk. Confidence: moderate for the size of the benefit, low for whether it applies to people without elevated CRP or calculated risk above 7.5%.