To the Editor Dr Lee and colleagues1 addressed several important questions in their review of the management of community-acquired pneumonia. We have concerns about the analysis of time to first antibiotic dose and the conclusions inferred from the use of observational studies.
For the time to first antibiotic dose, the authors found only observational studies and included only those adjusting the mortality analysis for other risk factors. Exclusion of studies that did not report an adjusted association estimate disregards studies in which there was no association between time to antibiotic initiation and mortality in the bivariable analysis and therefore did not enter the variable into the adjusted analysis. Another analysis including such studies showed no association between time to first antibiotic dose and mortality.2
Paul M, Yahav D, Leibovici L. Management of Community-Acquired Pneumonia. JAMA. 2016;316(2):220-221. doi:10.1001/jama.2016.5013