[Show abstract][Hide abstract] ABSTRACT: The multicenter International Study of Wheezing in Infants (EISL) was developed to study the prevalence of recurrent wheezing and related risk factors in infants during the first year of life using a written questionnaire (EISL-WQ).
To constructively validate a modified, shortened version of the EISL-WQ in children up to 36 months of age in São Paulo, Brazil, and to verify its usefulness in diagnosing probable asthma in these children.
The parents of 170 infants aged 12 to 36 months answered the shortened EISL-WQ in an emergency room and were asked if their child was currently wheezing before a diagnosis was made by a physician. The consistency between parent perception and the physician's diagnosis was then evaluated. A second group (n = 55) participated in the validation of the short-term repeatability of the shortened questionnaire by completing it twice (mean interval, 23 days).
There was good agreement between parent perception of wheezing and the physician's diagnosis following auscultation (Kappa statistic = 0.7; odds ratio = 38.33; 95% confidence interval, 15.8 to 92.8; P < .001); sensitivity (82.8%), specificity (85.0%), positive predictive value (81.5%), and negative predictive value (86.0%) were all high.The short-term repeatability of the shortened version of the EISL-WQ was also high (kappa > 0.75). Questions added to the shortened EISL-WQ improved the internal consistency of the original questionnaire (Cronbach alpha = 0.823, P < .001) and a high Youden index was found for patients defined as probable asthmatics.
The shortened version of the EISL-WQ translated into Portuguese has high internal consistency, and is a valid, reliable, and reproducible instrument for obtaining data on wheezing in children below 36 months of age and for identifying those with probable asthma.
Journal of investigational allergology & clinical immunology: official organ of the International Association of Asthmology (INTERASMA) and Sociedad Latinoamericana de Alergia e Inmunología 01/2009; 19(1):35-42. · 1.89 Impact Factor