516 chapter 13 hoped to avoid surgery by participating in the EfFECTS trial (also see paragraph below on diagnostic misconception and misestimation). Second, during counselling of eligible patients, we have experienced that the complexity of the diagnostic RCT design kept some patients from participating, whereas they may have participated if an easier-to-understand cohort design had been used. Laymen are often more familiar with the design of therapeutic RCTs; the design of a diagnostic RCT can be much more difficult to grasp. Despite extensive counselling by treating physicians and researchers, many of these patients could not appreciate the need for blinding and/or randomization (“If you believe that this scan works – which you do, if you let people cancel their surgery because of it – then I don’t want to be in this lottery” is a phrase that I have heard many times), felt overwhelmed by the volume and complexity of the information, and/or could not understand the trial design altogether, as a result of which proper informed consent could not be obtained. This may at least have resulted in a skewed representation of intellectual capacities in our study population, which is also suggested by the observed highest completed educational level (Table 1, Chapter 7) showing that 61 of 118 (51%) trial participants completed higher professional education or university as compared approximately 33% of the general population [718, 767]. It is unclear whether this also may have influenced HRQoL estimates. Blinding of patients Blinding is an essential element of RCTs, but extremely challenging to maintain in a diagnostic RCT, as treatment consequences are typically based on a combination of the index test result and/ or randomization, depending on the trial design [34, 765]. In an ideal blinding situation, patients, physicians and researchers should be blinded to allocation, test result, and treatment, which could be possible if treatment options are drug or placebo [34, 765]. In case of a de-escalation study such as the EfFECTS trial, with surgery or active surveillance as treatment options, blinding can never be complete unless ethically disputable sham surgeries would be performed. As such, patients who were advised active surveillance in the EfFECTS trial, were inevitably unblinded to their allocation and [18F]FDG-PET/CT result. Blinding is of greater importance for some outcomes than others. For the EfFECTS trial, the primary outcome measure was the accurate reduction in the number of unbeneficial diagnostic surgeries for benign thyroid nodules. In case of surgery, the benign or malignant histopathological diagnosis is not influenced by patient blinding to their allocation and [18F]FDG-PET/CT result. Blinding may, however, have affected other outcomes, including HRQoL. In general, blinding and the suspense of not knowing that generally accompanies clinical trial participation may influence patient (perceptions of) physical and mental health, emotions, coping mechanisms, and other behaviour [765, 768]. It may also influence patient compliance to study procedures, change a patients’
RkJQdWJsaXNoZXIy MTk4NDMw