Possible biases in pharmacovigilance reporting may impact epidermal necrolysis (EN) and drugs associations.
To assess biases associated with EN-reporting.
Using VigiBase, the World Health Organization-pharmacovigilance database, among drugs associated with EN between 2016 and 2020, we used an unsupervised clustering including reports characteristics, that is, reporter quality, time from drug intake to EN onset, and only one suspected drug in the report.
Among 152 drugs, three clusters were identified. Cluster 1 (n = 41) exhibited drugs frequently reported within a time from intake to onset longer than 4 days, in 57 ± 13% of reports. It corresponded to well-reported drugs and was composed mainly of antivirals and antiepileptics. Cluster 2 (n = 42) exhibited drugs frequently reported within a time from drug intake to onset shorter than 4 days, in 31 ± 12% of reports. It corresponded to drugs with a high risk of protopathic bias and was composed of nonsteroidal anti-inflammatory drugs (NSAIDs), analgesics, and antibiotics. Cluster 3 (n = 69) exhibited drugs frequently reported with an unavailable time from drug intake to reaction, in 66 ± 11% of reports, and reported by a high frequency of consumers (9 ± 9%). It corresponded to drugs reported with a high risk of classification bias, and was composed of anticancer therapies and cardiovascular drugs.
Protopathic and classification biases impact EN-reporting and should be considered regarding associations with antibiotics, NSAIDs, analgesics, anticancer therapies, and cardiovascular drugs.

暂无摘要。

暂无摘要。

BACKGROUND: Reliable estimates of accuracy are important for any diagnostic test. Diagnostic accuracy studies are subject to unique sources of bias. Verification bias and classification bias are 2 sources of bias that commonly occur in diagnostic accuracy studies. Statistical methods are available to estimate the impact of these sources of bias when they occur alone. The impact of interactions when these types of bias occur together has not been investigated.
METHODS: We developed mathematical relationships to show the combined effect of verification bias and classification bias. A wide range of case scenarios were generated to assess the impact of bias components and interactions on total bias.
RESULTS: Interactions between verification bias and classification bias caused overestimation of sensitivity and underestimation of specificity. Interactions had more effect on sensitivity than specificity. Sensitivity was overestimated by at least 7% in approximately 6% of the tested scenarios. Specificity was underestimated by at least 7% in less than 0.1% of the scenarios.
CONCLUSIONS: Interactions between verification bias and classification bias create distortions in accuracy estimates that are greater than would be predicted from each source of bias acting independently.