gms | German Medical Science

Background: The overall completeness of structured information submitted in spontaneous adverse drug reaction (ADR) reports from physicians, pharmacists and consumers (e.g. patients and their relatives) from Germany was described as largely comparable in a previous study [1]. So far, a comprehensive analysis of other variables that might be associated with the completeness of ADR reports has not been published.

The aim of this study was to analyse which variables are associated with the completeness of information provided in the structured format of spontaneous ADR reports. Additionally, we investigated in a random sample of spontaneous ADR reports whether those with an at least possible causal relationship between the intake of the suspected drug and the occurrence of the ADR carry a higher grade of completeness.

Materials and Methods: We analysed 69,976 spontaneous ADR reports from physicians, 42,396 from pharmacists and 121,144 from consumers received between 2018 and 2021 from Germany contained in EudraVigilance, the European ADR database. The completeness of the information provided in the structured format of each spontaneous ADR report was evaluated by applying an adjusted vigiGrade completeness score originally developed by the Uppsala Monitoring Centre of the World Health Organization [2]. Subsequently, the mean values of the vigiGrade completeness scores were calculated for various variables. For categorical variables, unpaired t-tests with Holm’s correction for multiple testing were performed to identify differences between the analysed variables.

Results: Spontaneous ADR reports referring to older (> 60 years) and younger patients (< 18 years) received a lower mean value of completeness score than those referring to middle-aged patients. Further on, spontaneous ADR reports relating to females achieved a higher mean value of completeness score than those relating to males. Reports with unknown sex and age of the patient were generally associated with poorly documented spontaneous ADR reports. Likewise, the mean values of completeness scores decreased, the more drugs were reported as suspected/interacting per spontaneous ADR report. Spontaneous ADR reports without any free-text description (narrative) or with descriptions of less than 125 words and more than 300 words were associated with a higher mean value of completeness score.

In our random sample, spontaneous ADR reports with an at least possible causal relationship achieved a higher mean value of completeness score than those in which the causal relationship was not assessable. All of the aforementioned results were consistent for all three reporter types. Differences between the reporter types were only observed for specific seriousness criteria (e.g. hospitalisation) and over the years of the analysed period.

Conclusion: Our results underline that as much relevant information as possible should be provided in each spontaneous ADR report. This particularly applies to spontaneous ADR reports of children and older adults, which represent vulnerable patient populations. Our analyses may provide dedicated approaches to improve the completeness of ADR reporting.