This Article  • Full text  • PDF  • Send to a friend  • Save in My Folder  • Save to citation manager  • Permissions  Citing Articles  • Contact me when this article is cited  Related Content  • Related letter  • Similar articles in this journal  Topic Collections  • Immunization  • Tobacco  • Pneumonia  • Statistics and Research Methods  • Bacterial Infections  • Viral Infections  • Alert me on articles by topic

Michael L. Jackson, MPH; Noel S. Weiss, MD, DrPH; Jennifer C. Nelson, PhD; Lisa A. Jackson, MD, MPH

	Since this article does not have an abstract, we have provided the first 150 words of the full text and any section headings.

In their study of adults (primarily older adults) hospitalized for pneumonia during influenza season, Spaude and colleagues¹ found that prior receipt of influenza vaccine was associated with a 39% reduction in risk of in-hospital mortality. The authors interpreted this association as evidence of a protective effect of vaccine. An alternate explanation is that persons at lower risk of death were more likely to have been vaccinated—in other words, that the observed association was due to confounding.

Confounding by underlying health status is a major problem in observational studies of influenza vaccine.^2-3 Strong residual confounding in the study by Spaude et al¹ is suggested by their Table 2, in which smoking was associated with an even greater reduced risk of in-hospital mortality than was influenza vaccine in multivariate analyses.

Fortunately, because influenza is . . . [Full Text of this Article]

AUTHOR INFORMATION

RELATED LETTER

To Rule Out Confounding, Observational Studies of Influenza Vaccine Need to Include Analyses During the "Preinfluenza Period"—Reply
David N. Fisman
Arch Intern Med. 2007;167(14):1554-1555.
EXTRACT | FULL TEXT