Table 1Number needed to treat with statins to prevent one cardiovascular event in 5 years

TrialsNumber of SubjectsBaseline Risk of CHD Mortality per 100 Person-YearsRate RatiosNumber Needed To Treat (5 years)
Total MortalityCHD MortalityAll CV EventsTotal MortalityCHD MortalityAll CV Events
Primary Prevention
Secondary Prevention
Scandinavian simvastatin survival study trial4,4441.60.710.590.6433318
Long-term intervention with pravastatin in ischemic disease9,0141.40.780.770.8416417
Combined Effects (95% CI)0.80 (0.74 to 0.87)0.73 (0.66 to 0.81)0.74 (0.71 to 0.77)113 (77 to 285)500 (222 to -)**20 (17 to 25)

Adapted by permission from BMJ Publishing Group Limited. BMJ. Smeeth L, Haines A, Ebrahim S, vol. 318, pp. 1548–51, 1999.

CHD = coronary heart disease; CV = cardiovascular


AFCAPS/TexCAPS study reported a nonsignificant increased total and CHD mortality in the intervention group. Numbers needed to treat are derived from the lower limit of the 95% CIs of the risk differences in event rates to illustrate the lower limit within which the numbers might lie.


No upper number needed to treat can be calculated as the upper 95% CI of pooled absolute risk difference is greater than zero. In these circumstances, the number needed to treat is a number needed to harm.

From: Expanded Guidance on Selected Quantitative Synthesis Topics

Cover of Methods Guide for Effectiveness and Comparative Effectiveness Reviews
Methods Guide for Effectiveness and Comparative Effectiveness Reviews [Internet].

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.