Adequacy of patient information on adverse effects: an assessment of patient information leaflets in the UK

Abstract

Background: One of the most important categories of information that patients want to know about the drug they are taking is the likelihood or probability of adverse effects. All patients should receive such information in the patient information leaflet that is supplied with all drugs. Anecdotal evidence suggests that most leaflets give little indication of the likelihood of adverse effects. The UK Medicines and Healthcare products Regulatory Agency (MHRA) suggests using a combination of words and numbers to convey this information. However, an EU guideline suggests using five verbal descriptors on a scale from common to rare, the use of which has been shown to lead to gross overestimation of the risk of adverse effects.

Methods: We assessed the leaflets supplied with the 50 most frequently prescribed drugs in England, to determine the extent to which the likelihood of adverse effects was described, and whether it met the requirements of the EU guidance and/or best practice. We examined both the method used to describe the likelihood of adverse effects, and the format of this information in the leaflet.

Results: Twenty of the 50 leaflets (40%) gave no indication of the likelihood of adverse effects occurring. Six (12%) used the recommended EU terms and a further 20 (40%) used a wide range of other verbal descriptors. Only four leaflets (8%) provided any form of numerical indication of risk. Over half (52%) presented long lists of adverse effects in paragraphs of continuous text.

Conclusions: Patient need is not being met in terms of the provision of usable information about the likelihood of adverse effects. Most patients receive no information, whereas some are given verbal descriptors, both of which lead to overestimation of the risk. Very few of the patient information leaflets assessed used currently described best practice, i.e. to present verbal descriptions alongside numerical information in the form of natural frequencies, e.g. 'rare (affects less than 1 in 1000 people)'.