Introduction The double blinded randomised control trial has been developed to provide gold standard estimation of causal effects. However, in many circumstances it is impossible to design studies that meet this standard of blinding and placebo effects potentially affect the estimates. One example of a study where it was impossible to blind the participants is the Heating Housing and Health Study (HHHS). Here the intervention was the installation of a modern efficient heater in the participants' homes.
Methods Using data from the HHHS, we explored three different approaches to estimate the placebo effect: (1) Dummy outcome variables (where we examined variables similar to the real outcomes, on which the intervention is known to have no effect); (2) identifying variables that may indicate a high susceptibility to placebo effects; and (3) modelling the effects of measured intermediate variables (in the heating example the direct effect of the intervention should be associated with a rise in temperatures).
Results Combining these approaches in a Bayesian framework we calculated estimates of the placebo effect and intervention effects across a range of outcome measures in the HHHS. The changes in the estimates of these intervention effects showed which results were likely to be affected by placebo effects. These findings agree well with our expectations.
Conclusion With carefully chosen assumptions, it is possible to use data already collected and a Bayesian modelling approach to obtain informative estimates of likely placebo effects and hence provide better estimates of the true effects of an intervention in unblinded RCT's.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.