© 2007 American Thoracic Society
Complications at Birth and Subsequent Wheeze: Risk of Attrition BiasFrom the Authors:We thank Drs. Bernsen and van der Wouden for their comments on our recent article (1). First, we would like to note that the overall response rate for the SIDRIA study (89.2%) was very high (1), and was much better than in most previous studies that have addressed these issues. It is correct that a number of children were excluded from the analyses for various reasons. However, if we compare subjects included (n = 12,187) and excluded (n = 7,829), we do not observe any significant difference in the prevalence of current wheezing as defined by the ISAAC question (8.0 and 7.7%, respectively), or as defined in our study (10.9 and 11.2%, respectively), where several wheezing questions were used to define "current wheezers." Second, it should be emphasized that exclusion of study participants with incomplete information or potentially invalid information will not in itself result in bias unless these exclusions are systematically related both to exposure and to disease status (e.g., if children who had no symptoms and whose mothers had had maternal complications were more likely to be excluded than children who had no symptoms and whose mothers had not had maternal complications) (2). Therefore, exclusion could be related to disease status, but this will not in itself result in a bias unless it is also systematically related to exposure status (i.e., maternal complications). In fact, any bias resulting from the exclusion of these children is likely to have been relatively small. For example, for those included in the analyses (n = 12,187), 5.4% of mothers had had hypertension or pre-eclampsia; among the 7,829 children excluded from the analysis, information on hypertension or pre-eclampsia was available for 6,455 (82%), and 5.8% of their mothers had had hypertension or pre-eclampsia. If these children had been included in the analysis, then the unadjusted odds ratio for late-onset wheezing would have changed from 1.36 to 1.41. Thus, it appears very unlikely that the exclusion of children with missing or incomplete information has resulted in any significant bias.
Anna Meyer Children's Hospital, Florence, Italy
San Giovanni Battista Hospital, Turin, Italy
Rome E Local Health Authority, Rome, Italy
Massey University, Wellington, New Zealand FOOTNOTES Conflict of Interest Statement: None of the authors has a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||