intention to treat preserves randomization and prevents selection bias
- related: Biostats
- tags: #permanent
In intention to treat analysis, all subjects initially allocated after randomization are included in their original group. This reduces effects of loss to follow up (attrition) and cross over effect from per-protocol and as-treated analysis.
In per-protocol analysis, only subjects who completed intervention are analyzed. Patients who are loss to follow up are not. For example, subject who was originally in intervention group dropping out may make the intervention look beneficial.
In as-treated analysis, subjects are treated based on the intervention they received than initially randomized to.
ITT analysis leads to more conservative estimate of effect of intervention
source
Links to this note
Intention-to-treat (ITT) analysis compares intervention groups in a randomized trial by including all subjects as initially allocated after randomization, regardless of what happens during the study period. The rationale is that if subjects are doing so poorly as to switch interventions or to drop out of the study, then their outcome should be attributed to that intervention. Therefore, ITT analysis is usually conducted to avoid the effects of crossover (eg, noncompliance to assigned intervention) and attrition (eg, loss to follow-up, drop-out), which may disrupt the benefit of randomization and introduce bias in the estimation of the effect of the intervention.
ITT analysis may lead to a more conservative estimate of the effect of the intervention. If attrition or crossover is significant, ITT analysis may be less likely to identify a statistically significant difference between interventions (shift toward the null hypothesis and reduced chances of false-positive conclusions). However, results will reflect the real effect of the intervention as intended in the population.
In this example, if subjects with more severe osteoarthritis drop out or cross over at a higher rate, even an ineffective intervention may appear beneficial if the analysis is conducted only on those who finished the protocol. Therefore, excluding subjects who were lost to follow-up or noncompliant to the assigned intervention introduces bias and may result in a statistically significant effect applicable only to subjects who completed the protocol. ITT analysis reduces bias by producing a more conservative but more valid estimate of the effect of using foot orthoses and motion control shoes as compared to motion control shoes alone.
In a per-protocol analysis, only data from subjects who completed the intervention originally allocated at randomization are analyzed. With as-treated analysis (a subtype of per-protocol analysis), subjects are evaluated based on the intervention they received rather than the intervention to which they were randomized. Therefore, the benefit of the randomization is lost. Usually, per-protocol analysis will overestimate the real effect of the intervention on the outcome.
Intention-to-treat is an important principle used in the analysis of randomized clinical trials. Intention-to-treat means that the patient's treatment status at the point of randomization is analyzed. If a patient who is assigned to the placebo group begins taking the medication assigned to the treatment group sometime after study initiation, or if a patient in the treatment group stops taking the prescribed medication, the data from these patients is still analyzed along with their original group. The value in the intention-to-treat approach is that it preserves the benefits of randomization and prevents bias due to selective non-compliance. Investigators may alternatively use the 'as treated' rule, which is the opposite of intention-to-treat (i.e. if a patient switches therapy they are counted as members of the new group during analysis).