Analysis of correlated bronchial responsiveness data

摘要: Correlated survival data analysis techniques were utilize d to determine the efficacy of such analyze correlated airway responsiveness data. When are correlated, standard maximum likelihood estimates regression coefficients obtained by using Cox's model ar e consistent, but errors may not be valid or consistent due within-subject dependencies and hence give rise wrong interpretations. We used jackknife, bootstrap method proposed Wei, Lin Weissfeld (WLW) obtain robust for errors. The analyzed in this report from a longitudinal study conducted investigate respiratory health effect s initial exposure grain dust among workers commencing employment industry province Saskatchewan, Canada. Bronchial was determined histamine inhalation test administered at four different time points. provocation concentration (PC20) inhaled that produced 20% fall lung function parameter forced expired volume one second (FEV1) as an indicator bronchial responsiveness. Failure defined dose which FEV1 occurred. In tests, censored occur failure reach end point. Censoring occurred when subject did have with 8 mg. Significant predicto rs FEV1, wheezing height. similar bootstrap, WLW, those method.