gms | German Medical Science

Verbal probability expressions such as ‘good chance’ or ‘likely’ are commonly used in physician-patient-communication to describe uncertainty and risk as for example in the context of adverse drug causality appraisal. The WHO for instance suggests four verbal expressions ‘certain’, ‘probable’, ’possible’, and ‘unlikely’ for classifying the probability of adverse drug reactions (ADRs) [1]. According to [1] the numerical range for these verbal probabilities is given by 0-5% for ‘unlikely’, 5-50% for ‘possible’, 50-95% for ‘probable’ and 95-100% for ‘certain’. However, there is insufficient evidence of this numerical interpretation. Moreover this interpretation might also differ from the numerical values assigned to those verbal expressions in [1]. We therefore aimed at answering these questions by asking a sample of N=476 individuals above the age of 18 (mean age 29.45 ± 14.47; median: 23 years) to numerically rate the set of four verbal probability expressions on a visual analogue scale from 0-100. The format of the question for the numerical rating task was given in the form “An event is ...” in which “...” was replaced by the verbal probabilities. As a result, numerical probability ratings of the sample differed notably from the proposed numerical values in [1] with values of ‘unlikely’ of 24.99 ± 17.28 (Mean ± SD), ’possible’ of 62.03 ± 24.72, ‘probable’ of 70.75 ± 18.69 and ‘certain’ of 84.21 ± 26.8 indicating a lack of agreement between the originally intended and perceived probability assessment. However, in order to substantiate the need to adapt the WHO classification system, further research based on larger sample sizes should be taken into consideration with a stronger focus on ADRs.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.