Question

Epidemiology

4 a) Pre and post studies with placebo in a variety of studies indicated that the Standard Deviation of Blood Pressure change was about 6 mmHg while mean reduction in placebo group was typically almost 5 mmHg. To make a preliminary estimate of the value of a new intervention designed to lower blood pressure it was planned to enroll subjects and test the null hypothesis that mean reduction was 5 mmHg.The new intervention would be of interest if the mean reduction was 10 or greater. How large a sample would be necessary to test,at the 5 % level of significance with a power of 90%,whether the average blood pressure reduction is 5 mmHg, assuming a standard deviation of 6 mmHg?

Answer 1

4.

a.

given data,
population mean =5 mmHg
sample mean =10
power =90%
type 2 error = 1- power
type 2 error =1-0.9 =0.10
standard deviation =6 mmHg
level of significance =0.05
sample size = n
n= ((σ(Zalpha +Zbeta))/(U-Uo))^2
n =((6*(Z0.05+Z0.1)/(10-5))^2
n= ((6*(1.645+1.28))/(10-5))^2
n =12.321
n= 13
large sample is 13