are statins statistically beneficial in the prevention of a first cardiac event or first stroke?

For a person without prior heart disease or stroke, the evidence for statins preventing a first event is statistically significant but the absolute benefit is small, and the trials that show it have important limits.

The best evidence for primary prevention comes from a 2013 Cochrane review that pooled 14 trials with about 48,000 people. It found statins reduced all-cause mortality by about 14% (odds ratio 0.86) and total stroke events by about 22% (risk ratio 0.78) [7]. A separate 1999 meta-analysis of 14 randomized trials reported a similar all-cause mortality reduction (odds ratio 0.76) and a non-significant reduction in fatal stroke (odds ratio 0.77, confidence interval crossing 1.0) [1]. These are real statistical signals.

But the absolute numbers matter. The Cochrane review's numbers translate to roughly 1 fewer death per 100 people treated for about 4 years. The JUPITER trial, which tested rosuvastatin in people with normal LDL but elevated CRP, found a hazard ratio of 0.55 for the primary endpoint in white participants, but the number needed to treat was about 84 under strict JUPITER criteria and 92 under extended criteria [6, 8]. That means you treat about 85 people for nearly 2 years to prevent one first event.

The evidence also shows that the benefit is not uniform. A 2020 meta-analysis of 29 randomized trials found that the cardiovascular mortality benefit of intensive LDL-lowering was strongest in people with a body mass index under 25 (hazard ratio 0.57) and disappeared entirely in those with BMI over 30 (hazard ratio 0.96, confidence interval crossing 1.0) [5]. A 2021 study of new statin users over age 70 found a reduced risk of major cardiovascular events (hazard ratio 0.71) but no significant increase in new diabetes, while younger adults showed a diabetes risk with higher adherence [4].

The harms are real and undercounted. A 2016 retrospective study of healthy adults who took statins for a short period found increased odds of developing diabetes (odds ratio about 2.0) compared to non-users [9]. In dialysis patients, a 2015 meta-analysis showed statins improved cholesterol numbers but had no effect on cardiovascular death or all-cause mortality [3]. The SPARCL trial in stroke patients found atorvastatin 80 mg reduced recurrent stroke by 16% but increased hemorrhagic stroke by 66% (hazard ratio 1.66) [2].

The funding picture matters. Most of these trials were funded by the manufacturers. The Cochrane review is independent, but the trials it pooled were industry-sponsored. The JUPITER trial was funded by AstraZeneca, the maker of rosuvastatin [8]. No large primary prevention trial has compared statin users to a truly untreated group over a decade with complete adverse event tracking.

My call: statins provide a statistically significant but modest reduction in first cardiac events and first stroke for some people, with the benefit concentrated in younger, leaner individuals and vanishing in the obese and in dialysis patients. The absolute benefit is small enough that about 85 people must take the drug for 2 years to prevent one event, while the diabetes risk and other harms are real and underreported. Confidence: moderate for the statistical significance of the benefit in selected populations; low for the net benefit in the average person considering it, because the absolute gain is small and the harm data is incomplete.