Do case-only designs yield consistent results across design and different databases? A case study of hip fractures and benzodiazepines.

Gema RequenaJohn LogieElisa MartinNada BoudiafRocío González GonzálezConsuelo HuertaArturo AlvarezDavid WebbAndrew BateLuis A García RodríguezRobert ReynoldsRaymond SchliengerHelga GardarsdottirMark C H de GrootOlaf H KlungelFancisco de AbajoIan J Douglas

Published in: Pharmacoepidemiology and drug safety (2015)

CXO designs yielded consistent results across DBs, while initial SCCS analyses did not. Accounting for event-exposure dependence, estimates derived from SCCS were more consistent across DBs and designs.

Keyphrases

deep brain stimulation
finite element analysis
big data
machine learning
artificial intelligence