Nocebo effects and participant information leaflets: evaluating information provided on adverse effects in UK clinical trials

BACKGROUND: Nocebo effects ('negative placebo' effects) experienced by clinical trial participants can arise from an underlying condition or through communication about side effects in the participant information leaflets (or elsewhere). Misattributing nocebo effects to the medicinal intervention can lead to participants experiencing harmful nocebo effects and may result in distortion of adverse effect reporting. However, little is known about how information on potential side effects is provided to trial participants. There is increasing concern that the way in which potential side effects in clinical trials are described to patients in participant information leaflets (PIL) can in itself cause harm by either increased anxiety, poor adherence or inducing the side effect itself. In this study, we aimed to explore these concerns and identify the way in which potential side effects from investigational medicinal products used in trials are presented in written information to potential participants. METHODS: Trials were identified from the International Standard Randomised Controlled Trials Number (ISRCTN) clinical trial registry (a primary registry of the WHO International Clinical Trials Registry Platform (ICTRP)). Eligible studies were placebo-controlled clinical trials of investigational medicinal products (IMP) in adults conducted in the UK. We assessed readability using the Flesch Reading Ease scale, Gunning-Fog Index and Flesch-Kincaid Grade. Data extracted from the PILs were divided into 8 predefined qualitative themes for analysis in NVivo11. RESULTS: Most of the patient information leaflets were ranked as 'fairly difficult to read' or 'difficult to read' according to the Flesch Reading Ease scale. All studies presented information about adverse events, whereas only a third presented information about intervention benefits. Where intervention or study benefits were presented, they were usually after adverse events (21/33, 64%). DISCUSSION: Participant information leaflets scored poorly on ease of readability and had more content relating to adverse effects than any potential beneficial effects. The way in which adverse events were presented was heterogeneous in terms of their likelihood and severity and the amount and level of detail provided. By comparison, potential benefits from the intervention and/or study were described less often, by shorter text, and only after information about harms.