Avoiding immortal time bias in observational studies. Issue 3 (19th March 2020)

• Record Type: Journal Article
• Authors: Raymakers, Adam J.N.; Sadatsafavi, Mohsen; Sin, Don D.; FitzGerald, J. Mark; Marra, Carlo A.; Lynd, Larry D.
• Is Part Of: European respiratory journal. Volume 55:Issue 3(2020)
• Publication Date: 2020-03-19
• Subjects: Respiratory organs -- Diseases -- Periodicals; Respiration -- Periodicals; 616.2
• DOI: 10.1183/13993003.02494-2019
• Languages: English
• ISSNs: 0903-1936

Abstract

We read with interest the recent study by Suissa et al. [1] published in the European Respiratory Journal . The study used population-based administrative data from the province of Quebec, Canada, to evaluate the association between inhaled corticosteroid (ICS) exposure and lung cancer risk in patients with COPD. The results of the study indicated that ICS use was not associated with lung cancer risk in this patient population. Immortal time bias is common in observational studies; however, in the analysis conducted by Raymakers and colleagues appropriate measures were taken to avoid this bias http://bit.ly/36042OV