Intention to Treat Flashcards

1
Q

What is the intention to treat principle?

A

Participants must be counted in the groups to which they were initially assigned, regardless of whether they were compliant with the intervention.
Includes those who withdraw.

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

Rationale behind intention to treat principle?

A

In an RCT one sets out to estimate broader effects of allocating intervention to a group and not the narrow effects of those who comply.
In reality, few people adhere to treatments strictly.

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

What does intention to treat principle give us?

A

Conservative estimate of treatment effect compared with what would be expected in ideal world of full compliance

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

Advantages of intention to treat principle

A

Makes results from RCT more generalisable and pragmatic

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

When might intention to treat principle not be suitable?

A

RCTs where investigation is of biological effects of drugs

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

Impact of ITT analysis in non inferiority trials?

A

Will favour equality due to dilution

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

When is ITT difficult to interpret?

A

If number of crossovers are high

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

Ways of dealing with missing data

A

Test the difference
Use available data
Imputation of data
Statistical methods

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

What is test the difference?

A

Test for difference between drop out group and available subject group.
Baseline characteristics if comparable can substantiate loss without further adjustment

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

Ways of using available data to deal with missing data

A

LOCF
Worst case scenario analysis

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

What does LOCF stand for?

A

Last observation carried forward

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

When is LOCF useful?

A

Only for longitudinal observations where continuous scales are used

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

Disadvantages of LOCF

A

Many psychiatric diseases have natural course of remission - LOCF does not allow for this’
‘Too good’ for controls and ‘too stringent’ for patient group

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

What is worst case scenario analysis?

A

Assumes that those lost to follow-up did not improve on trial intervention. Hence, data of treatment failure subjects used as proxy.

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

When is worst case scenario analysis used?

A

As part of sensitivity analysis

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

Types of imputations of data

A

Replacing missing values with mean of group
Hot deck imputation

17
Q

What are some statistical methods to make up for missing data?

A

Multiple regression
Growth curve analysis
Random effects modelling
Multiple imputation

18
Q

What is hot desk imputation?

A

Involves finding similar person among subjects and using that persons score

19
Q

What is multiple regression?

A

Used to predict missing values based on several other variables in data set

20
Q

What is growth curve analysis?

A

Line is fitted through existing data points for each subject. From 2 data points, we can derive parameters for straight line. For 3; quadratic curve. For 4; cubic curve.

21
Q

What is needed for growth curve analysis

A

At least 2 follow-up visits

22
Q

What is the definition of random effects modelling?

A

Derives from the fact that, when we usually fit a regression line to data, we assume that the parameters are fixed i.e. the same for all subjects
In random effects model, we allow them to differ for each person

23
Q

Who defined random effects modelling?

A

Streiner

24
Q

Implications of random effects modelling?

A

We do not expect curves for people who drop out to be same as for people who complete study

25
Q

What is multiple imputation?

A

Sophisticated and accurate methods which use maximum likelihood and Monte Carlo procedures.

26
Q

Recommendations if missing data are imputed?

A

Some sensitivity analysis is performed to ensure study conclusions are not misleading

27
Q

Alternatives to ITT

A

Per-protocol analysis
Treatment-received analysis

28
Q

What is per-protocol analysis?

A

Only patient who sufficiently complied with trials protocol are considered in final analysis.

29
Q

Another name for per-protocol analysis?

A

On treatment analysis

30
Q

Which types of RCTs use per-protocol analysis?

A

Biological explanatory RCTs
Secondary analysis in effectiveness trials

31
Q

What is treatment-received analysis?

A

Analyse subjects according to actual treatment received - not allocation made.

32
Q

Disadvantage of treatment-received analysis?

A

Effect of randomisation is compromised due to contamination of trial by one group of subjects moving to other group

33
Q

ITT is the norm except when?

A

Overwhelming justification of different analysis policy

34
Q

Advantages of ITT

A

Retains balance of prognostic factors arising from original random treatment allocation
Gives unbiased estimate of treatment effect
Admits non-compliance and protocol deviations, reflecting clinical life

35
Q

Limitations of ITT

A

Estimate of treatment effect is conservative
In equivalence trials, ITT will favour equality of treatment
Interpretation is difficult if large proportion of participants cross over to opposite treatment arms

36
Q

Why is estimate of treatment effect conservative in ITT?

A

Dilution effect of non-compliance

37
Q

Requirements for ideal ITT

A

Full compliance with randomised treatment
No missing responses
Follow-up of all participants