Recommendations encourage CEA to report intention to treat (ITT) estimates, but this may be insufficient for policy making as new decision problems may arise subsequent to the trial design. Clinical decision-makers also require cost-effectiveness estimates for patients who meet particular treatment thresholds, or for lower levels of compliance, which better reflect routine practice. Per protocol (PP) analyses are common but will provide biased estimates.
Our team aims to improve methods for estimating causal treatment effects after deviation from protocol:
- Assessment of an Instrumental Variable (IV) approach for dealing with non-compliance in cost-effectiveness of alternative decision problems.
- Developing multiple imputation approaches for dealing with non-compliance in settings with time-varying deviations.
These novel methods are illustrated in the evaluation of within-study alternative research questions:
- IMPROVE trial. Approximately 40% of those randomised to eEVAR will be judged ineligible according to CT scan immediately post randomisation, and will switch to open surgery (binary treatment switching).
Original question: Is eEVAR for aortic aneurism cost-effective for all those potentially eligible?
New question: Is eEVAR for aortic aneurism cost-effective, for the subgroup of patients who meet the CT scan
eligibility criteria, i.e. ‘the compliers’?
- PROMISE trial. Compliance levels with the new protocol are recorded hourly, and anticipated to be higher than in routine practice. This is an example of continuous levels of adherence.
Original question: Is the new protocol cost-effective at the average levels of adherence in the trial?
New question: Is the new protocol cost-effective at different levels of adherence?
KEY PEOPLE: Richard Grieve