Question

A randomized controlled study was done (Lewis et al, Annals of Internal Medicine, 1998) to see whether the cholesterol lowering drug pravastatin can lower the risk of coronary disease in older patients after myocardial infarction.

Results

Major coronary events occurred in 28.1% of placebo recipients and 19.7% of pravastatin recipients (difference, 9.0 percentage points [95% CI, 4 to 13 percentage points]; relative risk reduction, 32%; P < 0.001). Coronary death occurred in 10.3% of the placebo group and in 5.8% of the pravastatin group (difference, 4.6 percentage points [CI, 1.9 to 6.5 percentage points]; relative risk reduction, 45%; P = 0.004). Stroke incidence was 7.3% in the placebo group and 4.5% in the pravastatin group (absolute reduction, 2.9 percentage points [CI, 0.3 to 4.5 percentage points]; relative reduction, 40%; P = 0.03).

All of the following statements regarding interpreting the numbers are true EXCEPT one. Which statement is NOT accurate?:

a. If we repeated the study on a similar sample of subjects, we would observe a difference of 9 percentage points or greater between pravastatin and placebo less than 1 in 1000 times if there were actually no difference between the groups.

b. We are 95% confident that the true reduction in coronary events from pravastatin compared with placebo is is between 4 and 13 percentage points.

c. Subjects receiving pravastatin had 55% of the risk of coronary death as the placebo group.

d. We are more than 95% confident that pravastatin reduces the risk of coronary events, coronary death, and stroke.

e. We are 97% confident that pravastatin reduced the risk of stroke by 2.9 percentage points compared with placebo.

Answer 1

A randomized controlled study was done (Lewis et al, Annals of Internal Medicine, 1998) to see whether the cholesterol lowering drug pravastatin can lower the risk of coronary disease in older patients after myocardial infarction.

Results

Major coronary events occurred in 28.1% of placebo recipients and 19.7% of pravastatin recipients (difference, 9.0 percentage points [95% CI, 4 to 13 percentage points]; relative risk reduction, 32%; P < 0.001). Coronary death occurred in 10.3% of the placebo group and in 5.8% of the pravastatin group (difference, 4.6 percentage points [CI, 1.9 to 6.5 percentage points]; relative risk reduction, 45%; P = 0.004). Stroke incidence was 7.3% in the placebo group and 4.5% in the pravastatin group (absolute reduction, 2.9 percentage points [CI, 0.3 to 4.5 percentage points]; relative reduction, 40%; P = 0.03).

All of the following statements regarding interpreting the numbers are true EXCEPT one. Which statement is NOT accurate?:

1. If we repeated the study on a similar sample of subjects, we would observe a difference of 9 percentage points or greater between pravastatin and placebo less than 1 in 1000 times if there were actually no difference between the groups.

( true, relative risk reduction, 32%; P < 0.001)

2. We are 95% confident that the true reduction in coronary events from pravastatin compared with placebo is is between 4 and 13 percentage points.

(true, difference 9.0 percentage points , 95% CI, 4 to 13 percentage points)

c. Subjects receiving pravastatin had 55% of the risk of coronary death as the placebo group.

(true, relative risk reduction, 45%)

Correct option: d. We are more than 95% confident that pravastatin reduces the risk of coronary events, coronary death, and stroke.

( for this no data given)

e. We are 97% confident that pravastatin reduced the risk of stroke by 2.9 percentage points compared with placebo.

(TRUE, absolute reduction, 2.9 percentage points [CI, 0.3 to 4.5 percentage points]; relative reduction, 40%; P = 0.03)