gms | German Medical Science

Background: Spontaneous reports of risks related to the quality of a medicine detected by pharmacists can generate a drug safety signal. However, signal validation can be challenging without information about the frequency of prescribing of the individual medicine. In this case study, the number of spontaneous reports about mechanical defects of the novel inhaler generation Genuair^® over time was correlated with claims data and compared to its progenitor Novolizer^®.

Materials and Methods: Data from spontaneous reports concerning Genuair^® (aclidinium bromide) and Novolizer^® inhalers were drawn from the Drug Commission of German Pharmacists’ (AMK) database. The AMK as national institution of German pharmacists defined by the German Drug Law (AMG) receives case-based safety reports detected in German pharmacies and acts in concert with marketing authorization holders and competent authorities. Moreover, if a risk related to a specific medicine occurs prevalently, pharmacies are made aware of this signal, proactively. Claims data from prescriptions of the said inhalers were extracted from the DAPI database which contains anonymous claims data of prescribed medicines dispensed at community pharmacies at the expense of the German Statutory Health Insurance Funds (Gesetzliche Krankenversicherung, GKV). The database covers prescription data of more than 80 % of German community pharmacies. Monthly reclamation rates since marketing authorization were determined by linkage of reporting and prescription data over a 2-years-period. For further signal validation, monthly reclamation rates of the Genuair^® and its progenitor, Novolizer^®, were compared. Reclamation rates were calculated as percentage of inhalers with reclamation based on mechanical defects of the total number of inhalers dispensed at the expense of the GKV per year.

Results: When comparing data from 2012 to 2014 (two years after marketing authorization of Genuair^®), the Genuair^® inhaler shows an overall reclamation rate which is five times higher than the reclamation rate of the Novolizer^®. However, when comparing data of the respective period after marketing authorization, reclamation rates are comparable.

Conclusion: Claims data can generally be helpful when validating signals on spontaneously reported quality defects of medicines. By determination of reclamation rates, different medicines can be correlated. However, there are many questions to be addressed before extracting and interpreting data. For example, specific characteristics of a certain medicine have to be taken into account and general confounders such as the influence of information by the media on the frequency of reports should be considered. Hence and before drawing conclusions, data derived from spontaneous reports should be interpreted with caution.