Number needed to treat (NNT)
number of patients need to be treated for one to benefit
Inverse of the absolute risk reduction
The difference between the rate of events in the treatment (Pa) and control (Pc)
Number needed to harm
Number of patients harmed rather than benefited.
Prevent heart attack or stroke - Risk of bleeding
May be close
Caries prevention >> Risk of rare bone cancer
•dental radiograph - same type of situation
Statistical power / Sample size determination:
1) Expected difference in the mean values between the groups.
2) Expected scatter.
3) Significance level (generally p< 0.05)
Applying evidence mid term