Intention-to-treat analyses for randomised controlled trials in hospice/palliative care: the case for analyses to be of people exposed to the intervention.

INTRODUCTION: Minimising bias in randomised controlled trials (RCTs) includes intention-to-treat (ITT) analyses. Hospice/palliative care RCTs are constrained by high attrition unpredictable when consenting, including withdrawals between randomisation and first exposure to the intervention. Such withdrawals may systematically bias findings away from the new intervention being evaluated if they are considered non-responders. This study aimed to quantify this impact within ITT principles.

METHODS: A theoretical model was developed to assess the impact of withdrawals between randomisation and first exposure on i) study power and ii) effect sizes. Ten reported hospice/palliative care studies had power recalculated accounting for such withdrawal.

RESULTS: In the theoretical model, when 5% of withdrawals occurred between randomisation and first exposure to the intervention, change in power was demonstrated in binary outcomes (2.0-2.2%), continuous outcomes (0.8-2.0%) and time-to-event outcomes (1.6-2.0%), and odds ratios were changed by 0.06-0.17. Greater power loss was observed with larger effect sizes. Withdrawal rates were 0.9%-10% in the ten reported RCTs, corresponding to power losses of 0.1%-2.2%. For studies with binary outcomes, withdrawal rates were 0.3-1.2%, changing odds ratios by 0.01-0.22.

DISCUSSION: If blinding is maintained and all interventions are available simultaneously, our model suggests that excluding data from withdrawals between randomisation and first exposure to the intervention minimises one bias. This is the safety population as defined by the International Committee on Harmonisation. When planning for future trials, minimising the time between randomisation and first exposure to the intervention will minimise the problem. Power should be calculated on people who receive the intervention.