Final

Question 1

Procedure Selection Bias

Answer 1

When healthy patients are given one treatment over another which results in the inflation of good qualities towards the treatment

Fix: randomization

Question 2

Post Entry Exclusion

Answer 2

When the exclusion criteria is changed after the data has been seen

Fix: no post hoc changes

Question 3

Selective Loss of Data

Answer 3

Caused when there is data missing resulting in a difference in the population. Effects vary depending on what the mechanism of missing is.

Question 4

Assessment Bias

Answer 4

When the physician overestimates or underestimates a condition based upon the exposure or group of the individual

Fix: mask

Question 5

Ascertainment Bias

Answer 5

When one group is sampled more or in greater detail relating to their illness or treatment group

Fix: mask

Question 6

Single Mask

Answer 6

Patient doesn’t know treatment

Question 7

Double Masking

Answer 7

Patient and be physician do not know the status of the individual

Question 8

Triple mask

Answer 8

Patient doctor and DMC do not know the groups. This works best for objective testing of the stopping point so as not to favor one group over another.

Question 9

Concurrent Trials

Answer 9

Eliminates the time effect of the study that would be found in historical controls. Eliminates the differences in selection criteria and time treatment

Question 10

Active Follow Up

Answer 10

Follow up patient who drop out to make sure that there is no mechanism of why one group quits more.

Question 11

Drop In

Answer 11

Patients who should have been in the control are in the treatment (p)

Question 12

Drop Out

Answer 12

Patients who should have been in the treatment group are given the placebo

Question 13

SS calculation win drop in and drop out

Answer 13

N=N*/[(1-p-q)^2(1-r)]

Question 14

Cross Over Design

Answer 14

Repeated measure longitudinal study where the patient will receive both treatments. Subject to bias if there is a crossover effect.

Question 15

Binary Cross Over Null

Answer 15

p12-p21=0

This is the off diagonal of the patients who showed a preference to one treatment over another

Question 16

McNemars Test of Cross Over

Answer 16

Binomial(.5, n10+n01)

The null is that the preferential response is random with a frequency of 50%

Question 17

Factorial Design

Answer 17

Only test that can test interaction effects the interaction effect can be removed afterwards unless one reverses another.

Question 18

Incomplete Factorial

Answer 18

There is no placebo. Used to test is a combination of drugs is better than each of the mono therapies.

Null: one is better than the combination
Alt: the combination is better than both of the other two

Question 19

Adaptive Design

Answer 19

Allows for a change in the study design, statistics, sample size, treatment, hypothesis, power, and end points

Very flexible

Question 20

Sequential Analysis

Answer 20

SS is not fixed at start but while data is collected till a predefined stopping rule is reached. Lowers the over all cost

Can be done for futility, safety, superiority, or cost reasons

Question 21

Conditional Power

Answer 21

It is the probability of getting statistically significant results given the trend in data

Question 22

Drop the Loser Design

Answer 22

The inferior treatment group is dropped at a interim analysis and the population groups are changed to allow for new testing. Power is determined after second stage

Question 23

Interim Analysis

Answer 23

Data dependent stopping or continuation. Check to see if null can be accepted or rejected. Also allows stopping for new literature

Continuation alleys for increase in precision and power

Question 24

Type I Error

Answer 24

The probability of rejecting a null that is true. Overall type one error should be kept below .05

aT=1-(1-a1)(1-a2)…. Or aT=1-(1-a)^n

Question 25

Stopping Rule

Answer 25

If the test statistic at that point falls into the boundary of acceptance or failure then the trial is stopped and the parameters are estimated.

Question 26

Pocock p-value adjustment

Answer 26

Assumes a fixed p value at each location which means there is a high starting p-value and a lower ending one. Typically stops trials early and bad since the study cannot reject a significant hypothesis at the end

Question 27

OBrien Fleming P-value adjustment

Answer 27

Most used but with the most amount of rules. Required the amount of IA to be specified beforehand and that there is equal spacing in the IA. Ending p-value is .05

Question 28

Alpha spending function

Answer 28

The p-value is a function of the amount of information collected at each point a(t) where t=n/N or t=d/D a(0)=0 and a(t)=a

Question 29

Non-inferiority Trials

Answer 29

Null is that the new treatment is inferior (uC-uT>a) | One sided test where the entire CI must be (-a,0)

Question 30

Margin of Indifference

Answer 30

The allotted difference in the new treatment and the old treatment if the new treatment has some sort of advantage

Question 31

Longitudinal Study

Answer 31

Patients are blocks with high correlation with their measurements. Repeated measures with high correlation. Use GLMM.

Question 32

GLMM

Answer 32

Has both fixed and random coefficients for subject specific differences. Error is no longer independent for the subjects. The study parameter regression coefficients assumed same over the entire population and the coefficients of the person fixed for that person Estimated with the generalized least squares

Question 33

Case Symmetric Compound

Answer 33

Variances are the same but different from the covariances which are the same. Unrealistic

Question 34

Unstructured Covariance

Answer 34

Requires most amount of data to fit the most parameters. Most liberal

Question 35

Auto regressive

Answer 35

Makes the measurements closer together similar but drop off rapidly as the values get further apart.

Question 36

AIC

Answer 36

Want it to be lower and is a good check for model fit

Question 37

Goodness of Fit

Answer 37

Add the two -2loglikelihoods and test it with a chi square with two degrees of freedom

Question 38

Missing Data

Answer 38

Can use GLMM when there is missing data for MCAR or MAR

Question 39

GLM: Continuous Data

Answer 39

Use a normal distribution with g(u)=u

Question 40

GLM: Binomial Data

Answer 40

Use a proportion in a logistic regression g(u)=log(u/(1-u)) logic

Question 41

GLM: Count Data

Answer 41

Use a poison regression with g(u)=log(u)

Question 42

GLM: positive continuous data

Answer 42

Do a Gamma model with a gamma distribution and a link function g(u)=log(u)

Question 43

GEE

Answer 43

Generalized estimating equations when the Covariance is unknown

Question 44

Fixed Effect Meta Analysis

Answer 44

Mathematical assumption that the studies are looking at the same treatment effect

Question 45

Random Effects Meta Analysis

Answer 45

That there is a difference in the treatment effect over the studies. Good for when there is statistical heterogeneity in the studies (look at Forest plot)

Question 46

Selection Bias

Answer 46

It is when the sample is not representative of the target population resulting in lower external validity Fix: Random sampling