King's College London

Research portal

People’s understanding of verbal risk descriptors in patient information leaflets. A cross-sectional national survey of 18 to 65 year olds in England.

Research output: Contribution to journalArticlepeer-review

Original languageEnglish
Pages (from-to)743-754
JournalDrug Safety
Issue number8
Early online date11 May 2017
Accepted/In press20 Apr 2017
E-pub ahead of print11 May 2017
PublishedAug 2017


King's Authors



Evidence suggests the current verbal risk descriptors used to communicate side effect risk in patient information leaflets (PILs) are overestimated.


The aim was to establish how people understand the verbal risk descriptors recommended for use in PILs by the European Commission (EC), and alternative verbal risk descriptors, in the context of mild and severe side effects.


A cross-sectional online survey was carried out by a market research company recruiting participants aged between 18 and 65 years living in England. Data were collected between 18 March and 1 April 2016. Participants were given a hypothetical scenario regarding the risk of mild or severe medication side effects and asked to estimate how many out of 10,000 people would be affected for each of the verbal risk descriptors being tested.


A total of 1003 participants were included in the final sample. The risks conveyed by the EC recommended verbal risk descriptors were greatly overestimated by participants. Two distinct distributions were apparent for participant estimates of side effect risks: those for ‘high risk’ verbal descriptors (e.g. ‘common’, ‘likely’, ‘high chance’) and those for ‘low risk’ verbal descriptors (e.g. ‘uncommon’, ‘unlikely’, ‘low chance’). Within these two groups, the distributions were near to identical regardless of what adverb (e.g. very, high, fair) or adjective (e.g. common, likely, chance) was used. The EC recommended verbal risk descriptors were more likely to be understood in accordance with their intended meanings when describing severe side effects. Very few demographic or psychological factors were consistently associated with how well participants understood the EC recommended verbal risk descriptors.


The current verbal risk descriptors used in PILs are ineffective at best and misleading at worst. Discontinuing the use of verbal risk descriptors would limit the likelihood of people overestimating the risk of side effects.

Download statistics

No data available

View graph of relations

© 2020 King's College London | Strand | London WC2R 2LS | England | United Kingdom | Tel +44 (0)20 7836 5454