Copyright 1998 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.1998
I greatly appreciated Mossad's thoughtful comments. I agree that inoculation studies are a valid method of studying colds. However, since there is evidence that upper respiratory tract infections from nasal inocula may have a different time-course than those obtained de novo, they should probably not be combined in a meta-analysis evaluating cold symptoms at a specific point.
Meta-analysis is a powerful analytic tool but can be subject to abuse and misinterpretation if not applied carefully.1 Meta-analysis is most useful either in combining a number of homogeneous studies not individually powerful enough to show statistical significance, or in dissecting outcomes not central to the original investigation, but important and unlikely to merit a randomized controlled trial. With regard to our meta-analysis of the effects of zinc on the common cold, the simple fact is that there was a great deal of heterogeneity between the different trials. With such heterogeneity, results from meta-analysis should be viewed skeptically, and the focus should be on determining the underlying reasons for such disparate results.
Jackson JL. Another Look at a Meta-analysis of Zinc Salts Lozenges and the Common Cold. Arch Intern Med. 1998;158(9):1038-1040. doi: