Skip to main content
. 2022 May 17;45(5):459–476. doi: 10.1007/s40264-022-01155-6

Fig. 1.

Fig. 1

Relationships between pharmacovigilance data sources, analytical approaches, pharmacovigilance tasks, and causal inference paradigms. Each data source is commonly analyzed by specific analytical approaches depending on the characteristics of data in those data sources. Each pharmacovigilance task is also associated with specific analytical approaches. Causal inference paradigms are integrated with different analytical approaches and applied to pharmacovigilance tasks. ADE adverse drug event, LSTM long short-term memory, NLP natural language processing, RNN recurrent neural network, RWD real-world data, SVM support vector machine