Intention to treat analysis

Question 1

Intention to treat principle

Answer 1

• holds that participants must be counted in the groups to which they were initially randomised regardless of whether they were compliant with the allocated intervention
• this includes people that were withdrawn

Question 2

Why do a ITT?

Answer 2

-gives a conservative estimate of the treatment effect compared with what would be expected in an ideal world with full compliance

Question 3

Benefits of ITT

Answer 3

• tends to make RCT results more generalisable and pragmatic

- helps to deal with missing data

Question 4

How to deal with missing data- using available data

Answer 4

1. test the differences between the lost group and the available subject group-baseline characteristics
2. the last observation can be carried forward- only for longitudinal observations
3. worst case scenario analysis-assume all those who were lost to follow up did not improve

Question 5

How to deal with missing data- using imputation of data

Answer 5

• replace missing values with the mean of the group

- hot deck imputation- find a similar person among subjects and use their score

Question 6

Statistical methods of dealing with missing data

Answer 6

• use multiple regression to predict missing values
• growth curve analysis- fit a curve to the existing data points and extrapolate
• random effects modeling
• multiple imputation-use maximum likelihood and Monte Carlo procedures

Question 7

Alternatives to Intention to treat

Answer 7

• per protocol analysis-only patients who sufficiently complied with the protocol are analysed
• treatment-received analysis: analyse subjects according to the actual treatment received, not actual allocation made

Question 8

ITT must be the norm

Answer 8

-unless there is an overwhelming justification for a different analysis policy

Question 9

Advantages of ITT

Answer 9

• retains balance in prognostic factors arising from the original random treatment allocation
• gives unbiased estimate of treatment effect
• admits non-compliance and protocol deviations, reflecting real clinical situation

Question 10

Limitations of ITT

Answer 10

• estimate of treatment effect is generally conservation due to dilution due to non-compliance
• in equivalence trials this analysis will favour equality of treatments

Question 11

Requirements for an ideal ITT analysis

Answer 11

• full compliance with randomised treatment
• no missing responses
• follow-up on all participants