Sir,
With reference to your report, "Infection milder but more widespread, study finds" (January 10), the new study from Vietnam and the Karolinska Institute in Stockholm, which concludes that there are far more human cases of mild H5N1 avian influenza than World Health Organisation tallies suggest, does not warrant rethinking the current bird flu paradigm or viewing H5N1 as a lesser threat.
Aside from the preliminary nature of these findings, two important methodological limitations in the study should be noted.
First, reports by those affected are often insufficient and unreliable on their own. The recall of symptoms by the study population may have been influenced by their awareness of local outbreaks. This phenomenon, known as "recall bias," is a common occurrence in epidemiological studies.
Of greater import, the absence of serological testing to confirm the presence of H5N1 antibodies prevents us from ruling out the possibility that another pathogen might be responsible for the reported symptoms. Clearly, further studies are needed to substantiate these conclusions. Until then, this report presents supposition.
Aubrey Noelle Stimola
Assistant Director of Public Health,
American Council on Science and Health,
New York, NY 10023, U.S.