Final Flashcards

1
Q

Procedure Selection Bias

A

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

Fix: randomization

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Post Entry Exclusion

A

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

Fix: no post hoc changes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Selective Loss of Data

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Assessment Bias

A

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

Fix: mask

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Ascertainment Bias

A

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

Fix: mask

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Single Mask

A

Patient doesn’t know treatment

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Double Masking

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Triple mask

A

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.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Concurrent Trials

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Active Follow Up

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Drop In

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Drop Out

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

SS calculation win drop in and drop out

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Cross Over Design

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Binary Cross Over Null

A

p12-p21=0

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

McNemars Test of Cross Over

A

Binomial(.5, n10+n01)

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Factorial Design

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Incomplete Factorial

A

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

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Adaptive Design

A

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

Very flexible

20
Q

Sequential Analysis

A

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

21
Q

Conditional Power

A

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

22
Q

Drop the Loser Design

A

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

23
Q

Interim Analysis

A

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

24
Q

Type I Error

A

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

25
Q

Stopping Rule

A

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.

26
Q

Pocock p-value adjustment

A

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

27
Q

OBrien Fleming P-value adjustment

A

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

28
Q

Alpha spending function

A

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

29
Q

Non-inferiority Trials

A

Null is that the new treatment is inferior (uC-uT>a)

One sided test where the entire CI must be (-a,0)

30
Q

Margin of Indifference

A

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

31
Q

Longitudinal Study

A

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

32
Q

GLMM

A

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

33
Q

Case Symmetric Compound

A

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

34
Q

Unstructured Covariance

A

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

35
Q

Auto regressive

A

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

36
Q

AIC

A

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

37
Q

Goodness of Fit

A

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

38
Q

Missing Data

A

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

39
Q

GLM: Continuous Data

A

Use a normal distribution with g(u)=u

40
Q

GLM: Binomial Data

A

Use a proportion in a logistic regression

g(u)=log(u/(1-u)) logic

41
Q

GLM: Count Data

A

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

42
Q

GLM: positive continuous data

A

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

43
Q

GEE

A

Generalized estimating equations when the Covariance is unknown

44
Q

Fixed Effect Meta Analysis

A

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

45
Q

Random Effects Meta Analysis

A

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)

46
Q

Selection Bias

A

It is when the sample is not representative of the target population resulting in lower external validity

Fix: Random sampling