Wouter Leclercq

Personalised perioperative care by e-health after intermediate-grade abdominal surgery: A multicentre, single-blind, randomised, placebo-controlled trial 175 10 obvious explanation would be the additional burden of the research setting. Second, we defined our outcome measures—return to normal activities and return to work—as time elapsing between surgery and reaching the event. Preferably, we would define these outcomes as time elapsing between randomisation and reaching the event. However, from a clinical perspective, this definition would be irrelevant because the event could not be reached before surgery. Doing the randomisation procedure at the time of surgery was also not possible for logistical reasons. To address this issue, we did an extra sensitivity analysis in which return to normal activities was measured from the date of randomisation. This analysis showed comparable results to the main analysis. Moreover, median lead times between randomisation and surgery were the same (8 days) in both groups. This finding, together with the comparable results between both analyses, reduce the likelihood of bias. Finally, we were not able to mask healthcare providers to the random allocation, but because these workers had no function in the data collection or analysis process, we are confident that the amount of bias is scant. We postulated that patients who had surgery and used a personalised e-health intervention would return to normal activities sooner after the surgical procedure than patients who received usual care. Our hypothesis was based on the idea that informing patients about the surgical procedure and the recovery process before surgery and during the recovery process (management of expectations) would improve the recovery process. 4,5 This hypothesis was confirmed by our study findings and, moreover, we showed that the e-health intervention improved postoperative recovery with little effort for health-care providers. However, the effect on return to normal activities was not reflected in the difference in QALYs between the study groups, which was small and not clinically relevant. Total health-care costs were higher in the intervention group than in the control group, but this difference was not significant. Primary health-care costs were lower and home care costs were higher in the intervention group compared with the control group. A possible explanation for this discrepancy between primary and home care costs is that the extra information and guidance that patients received by using the intervention reduced the demand for primary care. However, extra information about the expected recovery period might have resulted in a larger anticipated need for extra help after discharge, which might have induced the increase in home care costs. Future research should focus on the needs and motivation of different stakeholders to invest in these types of interventions and the logistical facilitators and barriers for implementation. This knowledge, together with results of the cost-effectiveness analysis of this intervention (appendix), will provide important information with a view to the implementation process of this e-health intervention. Moreover, future studies of the e-health intervention should be done in patients undergoing major abdominal surgical