Biostats 7

Question 1

What is the Gold Standard of Study Design?

Answer 1

Randomized, Double-Blind, Placebo- Controlled Trial

Question 2

What is Intention to Treat Analysis?

Answer 2

• In intention to treat analysis, groups are analyzed according to original, randomized group regardless of whether they received the treatment or not.
• Otherwise, your randomization is lost, and a selection bias is introduced
• Remember randomization usually ensures that known and unknown factors are evenly distributed in the various treatment groups

Question 3

What is a Type II error?

Answer 3

If null is not rejected when it should be

Concluding incorrectly that exposure and disease
are not associated

Concluding incorrectly that two treatment groups are not different

Question 4

What is power?

Answer 4

Power is ability to detect an association or difference if it really exists

Usually power for a study should be 80% or more

Type II error is committed when there is inadequate power for the study

Power is determined to a great extent by sample size. If sample is small, then likely to commit Type II error.

Question 5

Eqn for power?

Answer 5

1-beta error

Higher the power, the lower the beta error

● Higher the power, the better our chances of finding a treatment benefit if there is one

Question 6

Power can be increased by:

Answer 6

● Increasing Sample Size
● Increasing the treatment effect
● Increasing the alpha
Power of 80-90% is respectable

Question 7

What is the eqn for Risk difference?

Answer 7

incidence (exposed)- incidence (unexposed)

Question 8

What is the eqn for number needed to treat?

Answer 8

1/ risk difference

Question 9

What is a Type I error?

Answer 9

Type I error occurs when the null hypothesis is true, but it is incorrectly rejected.

Often a significance level of 0.05 is chosen; then 5% of the time the null will be rejected when it should not be.

The level of significance (alpha level) is pre-determined by the investigator; it is the probability of making a Type I error that the investigator is willing to accept.

Question 10

P value – the probability that the findings are due
to chance
● IF the P value = .24, there is a 24% probability these
findings are due to chance
● Generally, the alpha value is set at .05 or less. If
the p value is .05 or less, the null hypothesis is rejected
● The alpha level of .05 to reject the null hypothesis was selected arbitrarily; 1/20 probability that results are due to chance seems reasonable

Answer 10

P value – the probability that the findings are due
to chance
● IF the P value = .24, there is a 24% probability these
findings are due to chance
● Generally, the alpha value is set at .05 or less. If
the p value is .05 or less, the null hypothesis is rejected
● The alpha level of .05 to reject the null hypothesis was selected arbitrarily; 1/20 probability that results are due to chance seems reasonable