Report on International ADE databases

Introduction: Within the GRiP network (Global Research in Paediatrics-Network of Excellence) we aim to implement an infrastructure to stimulate and facilitate the development and safe use of medicine in children. Promoting the best possible information on the safety of drugs used in children is a crucial part of this endeavour. However, existing and readily available datasources from spontaneous reporting databases including AERS (FDA), VAERS (FDA/CDC) and VigiBase (WHO-UMC) are underused. Recently, WHO-UMC published a seminal paper on the characteristics of reports within VigiBase. Little is known on the available information with respect to paediatric reports in AERS and VAERS and whether there are any similarities and differences between the different spontaneous reporting databases. A characterization of available databases and their reports is useful to guide optimal (collaborative) signal detection methods and interpretation of results. Aim: To study the characteristics of paediatric ICSRs (individual case safety reports) within AERS, VAERS and EudraVigilance. Methods: Characteristics of paediatric ICSRs within VigiBase were derived from the recently published paper by Star et al. [period 1968-Feb 2010]. AERS-data was downloaded from the FDA-website for the period Jan 2004-Dec 2011. From the ICSRs, the non-vaccine related reports on children (0-18 yrs) were selected. VAERS data was downloaded from the CDC website for the period 1990-2011, all vaccine related reports on children (0-18 years) were selected. To control for duplicate reporting, each unique ICSR was only included once. EudraVigilance data were not yet available due to the duration of required public access procedures. Characteristics of the reports, including the reported drugs and events, were described and stratified by age-groups. Results: From AERS we included a total of 106,122 paediatric ICSRs (55% boys) (58% from the USA) with a median of 1 drug [range 0-157] and 3 events [1-94] per ICSR. Mean age was 9.1 years. 90% was submitted through expedited (15-days)(65%) or periodic reporting (25%) and 10% by non-manufacturers. From VAERS we included a total of 174,949 paediatric ICSRs (48% boys). Mean age was 4.7 years. The details on report type were not available in the public VAERS data set. In comparison with VigiBase, which included 3,472,183 reports (53% boys) (39% from US), reporters in AERS were more often consumers (25 vs. 4%) and less often physicians (32 vs. 55%). This information was not available from the public VAERS dataset. Most commonly reported drug classes in AERS by decreasing frequency were ‘Neurological’ (58%), ‘Antineoplastic’ (32%) and ‘Anti-infectives’ (25%) versus ‘Anti-infectives’ (33%), ‘Neurological’ (29%) and ‘Dermatological’ (12%) in VigiBase. The most frequently reported vaccines within VAERS were MMR vaccine (12%), DTaP vaccine (12%) and varicella vaccine (11%). Most commonly reported SOCs in AERS were ‘General’ (13%), ‘Nervous system’ (12%) and ‘Psychiatric’ (11%). In Vigibase the most frequent reported SOCs were ‘Skin’ (35%), ‘General’ (20%), and ‘Nervous system’ (19%). Within VAERS they were ‘General’ (34%), ‘Skin’ (14%), and ‘Nervous system’ (11%). Conclusions: In comparison with VigiBase, in AERS there are more ICSRs reported for neurological and antineoplastic drugs. This might be explained by the large proportion of study-reports in AERS or be biased due to different study-periods. The reports in VAERS had the lowest mean age, concerned routine childhood immunizations and were mainly classified as SOC “general disorders”