Healthy subjects were selected by including children who had no history of physician-diagnosed asthma, allergies, or chronic disease reported in the previously administered ATS DLD-78-C questionnaire, and whose caregivers did not report the use of asthma medications in the current brief respiratory questionnaire. Feno distribution was evaluated in this population using the arithmetic mean and SD. The influence of age, gender, race, body mass index (BMI), body surface area (BSA),12 pulmonary function, prematurity (ie, birth before 36 weeks gestational age), exercise performed on the day of testing (ie, organized physical class or recreational activities), and ARI in the previous 2 weeks was analyzed using a Student t test for dichotomous variables and linear regression for continuous variables. There were no active smokers in the data set. A detailed table was developed to classify subjects’ race (data available on request). Children who identified their ancestry as the “Indian subcontinent,” were included with whites, based on preliminary analyses suggesting similar Feno concentrations. Children with one white and one nonwhite parent were initially classified according to the latter, and later were analyzed separately. No child had two nonwhite parents of different ancestries. The influence of race was analyzed with one-way analysis of variance, and differences between groups were evaluated using the Scheffé test. Multivariate models were developed using the general linear models procedure, with forward entry of variables and removal at a p value of < 0.10. Analyses were repeated with Feno log-transformed, but as the results were essentially unchanged, the results presented used untransformed values except where otherwise noted. Analyses were performed using a statistical software package (SPSS for Windows, version 14.0; SPSS; Chicago, IL). A p value < 0.05 was considered to be statistically significant.