© 2007 American Thoracic Society
Complications at Birth and Subsequent Wheeze: Risk of Attrition BiasTo the Editor:Rusconi and colleagues in their recent article present interesting data on associations between events during pregnancy and birth with childhood wheezing (1). However, in our opinion, the authors have not adequately addressed the issue of attrition, with possible consequent selection bias. Summarizing the study phases where attrition took place, there was an initial response rate of almost 90%, implying that there were about 2,000 nonresponders. It is not unlikely that among these nonresponders there were relatively few children with wheezing. Second, almost 1,700 questionnaires were left out of the analysis because they were not completed by the biological mother, leaving us with about 83% of the original sample. This second group is possibly also not a random selection: one could, for example, think of deceased or divorced mothers. Third, we lose another 2,700 children whose parents, have relatively little education and have reported incomplete data on wheezing and respiratory disease. Again, these are possibly children with relatively few symptoms. At this stage, about 71% of the original sample is left. Finally, more than 3,400 children were excluded because of missing answers on maternal variables or confounders. Again, we have no clue how this group differed from the remaining (roughly) 12,000 children (only 55% of the original sample!). The prevalence of wheezing in the past 12 months found in this study is about 11%. Comparing this result with the prevalence found in Italy in the third phase of the worldwide ISAAC study (2), which was almost 8%, it appears that Rusconi and coworkers studied a sample where subjects with wheezing were overrepresented. Whether subjects with complications during pregnancy and birth were overrepresented and whether this resulted in spurious positive associations is unclear.
United Arab Emirates University, Al Ain, United Arab Emirates
Erasmus MC Rotterdam, Rotterdam, The Netherlands FOOTNOTES Conflict of Interest Statement: Neither author has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||