Other issues

### ITT (interntion-to-treat) analysis
- includes all subjects.
- account for treatment effects, difficulties in administering the drug and compliance issues

### PP (per protocol) analysis
- excludes subjects who stop being compliant
- evaluate the maximum benefit possible from a treatment, given perfect compliance

### Subgroup Analysis
- examine whether the treatment effect or side-effects of treatments are the same or greater in patients with a specific feature or risk factor so that more specific treatment decisions can be made
- multiple comparison

### Regression Analysis
- adjust for differences in baseline characteristics
- identify the predictors of an outcome variable
- identify prognostic factors while controlling for potential confounders
- determine prognosis (prognositic models)
- determine diagnosis (diagnostic models)

### Confounding
- confounder is associated with the outcome and is unequally distributed between the treatment goups
- use stratification and regression mdoeling to control confounders

### Missing Data
1. Missing completely at random (MCAR)
2. missing at random (MAT)
3. missing not at random (MNAT)

### Interim Monitoring and Stopping Rules

# main reasons
- ensure that adverse event frequency and toxicity levels are acceptable
- ensure that patients are not recruited into a trial that is going to be unable to reach a definitive result
- ensure that randomization of patients is stopped as soon as there is sufficiently clear evidence either for or against the treatment being evaluated
- address unexpected problems with the study protocol such as exclusion criteria delaying recruitment

# most popular statistical methods - group sequencial methods
- Pocock
- O'Brien and Fleming
- DeMets