Patients' perception of risk presentation of medication side effects: implications for medication counselling, safety and adherence

Background Effective communication of medication side effects is crucial for informed patient decision-making. Medication package inserts (MPIs) provide essential safety information, but patients often misinterpret risk descriptors. The study assessed patients' understanding and perception of medication side effect risk descriptors vis-a-vis information satisfaction, perceived severity of side effects, likelihood of experiencing side effects, perceived overall health risk, effect on medication adherence, and risk estimation. Methods A cross-sectional survey was conducted among 750 patients in five community pharmacies in Ibadan, Nigeria. Participants were assigned to receive one of three questionnaire formats, each presenting six hypothetical medication side effects from three drugs (pizotifen, diclofenac and amoxicillin) using the five bands of three risk descriptors (verbal, percentage, and natural frequency). Responses to perceived satisfaction, severity, health risk, likelihood of taking medications, general risk to health, medication adherence, and risk estimates were collected using a six-point Likert scale. Data were analyzed using ANOVA to assess mean differences in perception across the three side effect descriptors with SPSS for Windows version 27. Results Patients were not very satisfied with the side effects described as very common, common and uncommon, with their corresponding percentage and natural frequency descriptors. They considered the side effect slightly severe and perceived it would affect medication adherence in the order of percentage > natural frequency > verbal descriptor. While with rare and very rare side effects or corresponding percentage and natural frequency descriptors, patients were more satisfied, considered the side effects less severe and were less likely to discontinue medication in the order verbal = percentage > natural frequency. All risk descriptions were overestimated (verbal > percentage > natural frequency). Likelihood of experiencing side effect: percentage > natural frequency > verbal descriptor. Patients perceived slightly elevated general risk to health with no difference in descriptors. Conclusions Patients prefer verbal descriptors over percentages and natural frequencies for understanding medication risks, especially for rare side effects, viewing them as less severe. While patients generally overestimate risks, this tendency is stronger with verbal descriptors. A combined approach using both numerical and verbal descriptors could improve patient comprehension, medication counselling and adherence.