Pediatric Drug Safety Signal Detection: A New Drug–Event Reference Set for Performance Testing of Data-Mining Methods and Systems

Abstract

BackgroundBetter evidence regarding drug safety in the pediatric population might be generated from existing data sources such as spontaneous reporting systems and electronic healthcare records. The Global Research in Paediatrics (GRiP)–Network of Excellence aims to develop pediatric-specific methods that can be applied to these data sources. A reference set of positive and negative drug–event associations is required.ObjectiveThe aim of this study was to develop a pediatric-specific reference set of positive and negative drug–event associations.MethodsConsidering user patterns and expert opinion, 16 drugs that are used in individuals aged 0–18 years were selected and evaluated against 16 events, regarded as important safety outcomes. A cross-table of unique drug–event pairs was created. Each pair was classified as potential positive or negative control based on information from the drug’s Summary of Product Characteristics and Micromedex. If both information sources consistently listed the event as an adverse event, the combination was reviewed as potential positive control. If both did not, the combination was evaluated as potential negative control. Further evaluation was based on published literature.ResultsSelected drugs include ibuprofen, flucloxacillin, domperidone, methylphenidate, montelukast, quinine, and cyproterone/ethinylestradiol. Selected events include bullous eruption, aplastic anemia, ventricular arrhythmia, sudden death, acute kidney injury, psychosis, and seizure. Altogether, 256 unique combinations were reviewed, yielding 37 positive (17 with evidence from the pediatric population and 20 with evidence from adults only) and 90 negative control pairs, with the remainder being unclassifiable.ConclusionWe propose a drug–event reference set that can be used to compare different signal detection methods in the pediatric population.Electronic supplementary materialThe online version of this article (doi:10.1007/s40264-015-0265-0) contains supplementary material, which is available to authorized users.

Masking by vaccines in pediatric drug safety signal detection in the EudraVigilance database

“…For this approach, we used the GRiP pediatric-specific drug-event reference set for nonvaccines, which composes 256 unique DECs, 37 of which are classified as positive control pairs, 90 as negative control pairs, and the remainder unclassifiable. 20 We calculated standard performance metrics: area under the curve (AUC) (calculated by varying the threshold for the ROR value), sensitivity, specificity, positive predictive value, and negative predictive value at the predefined threshold of ROR > 1 and number of reports ≥3 in the full and restricted settings. Since the AUC assesses performance based upon the entire reference set and may potentially hide different patterns across DECs, we also looked at the change per DEC for the positive and negative controls.…”

“…Briefly, the algorithm attempts to predict a target variable by splitting the data recursively according to categories of the predictor variables and repeating until no further gain in group “purity” can be achieved or until a user‐specified stopping rule is reached. We used algorithms that minimized entropy (or maximized node homogeneity) while growing the classification tree and balanced cost (misclassification) with complexity (tree size) while “pruning” or reducing the tree . In this case, the classification tree algorithm was used to split all DECs into the classes “increased ROR,” “decreased ROR,” or “no change.” An increase in ROR was defined as >10% increase in ROR value and a decrease was defined as >10% decrease .…”

Reference set for performance testing of pediatric vaccine safety signal detection methods and systems

“…Previous approaches to develop such standards for drugs included consulting reference books such as the Physicians Drug Reference or Martindale [8], considering label changes [9], combining information from summary of product characteristics (SPC) and the literature as in two recent initiatives, the 'Observational Medical Outcomes Partnership (OMOP)' and the 'EU-ADR project' [10,11]. Work on validating approaches for the pediatric population is in progress [12].…”

How are we assessing the safety and quality use of medicines used by young people in Australia?

“…Children's physiological development may affect the bioavailability of medicines, making them potentially more vulnerable to adverse drug events . The impact of adverse events during growth and development may also be different, more serious and longer term compared with adults …”

“…3,4 The impact of adverse events during growth and development may also be different, more serious and longer term compared with adults. 5 Safety monitoring of medicines for the entire population, and particularly for children, relies heavily on the post-regulatory approval period. Spontaneous reporting systems, such as adverse event reports to the Therapeutic Goods Administration, are useful for signalling safety issues, but are subject to selective underreporting or stimulated reporting 2 and cannot quantify the incidence or rates of adverse events in the community or in population subgroups.…”