You are here

Predicting expectations of side-effects for those which are warned versus not warned about in patient information leaflets: A secondary analysis of a prospective cohort study

Publication date: 

1 Nov 2021


Annals of Behavioural Medicine 2021; Apr3: kaab015. DoI: 10.1093/abm/kaab015


Webster RK, Rubin GJ

Publication type: 



Abstract Background: Research investigating predictors of side-effect expectations is disparate and largely based on hypothetical vignettes. Purpose: To carry out a secondary analysis of a randomized controlled trial and investigate the predictors of side-effect expectations for side-effects that were, or were not, warned about. Methods: Two hundred and three healthy adults completed measures concerning demographics, psychological factors, baseline symptoms, and medication-related beliefs before reading one of two types of patient information leaflet (PIL) (standard or positively framed PIL) for a sham medication and asking them about their side-effect expectations. Associations between these measures and side-effect expectations whilst controlling for the PIL received were assessed using regression analyses. Results: 82.8% of participants expected side-effects that were warned about in the PIL, and 29.1% expected side-effects that were not warned about. Participants who were younger, from White backgrounds, less optimistic, experienced increased anxiety and received the standard PIL were more likely to expect side-effects that were warned about. Those with higher beliefs about medicine overuse and lower trust in medicine development were more likely to expect side-effects that were not warned about. Higher somatization, baseline symptoms, modern health worries scores, and lower trust in pharmaceutical companies were associated with increased expectations for all side-effects. The results suggest we can not only rely on altering side-effect risk communication to reduce side-effect expectations and therefore nocebo effects. We must also consider patients' beliefs about trust in medicines. More work is needed to investigate this in a patientsample in which the medication is known to them. Keywords: Expectations; Nocebo effect; Predictors; Side-effects.