Sample size estimation for field studies on the effects of aircraft noise on sleep

Abstract

Environmental noise disturbs sleep, may impair daytime performance and lead to long-term health consequences. Unbiased and precise exposure–response relationships inform the political and legislative decision making process, e.g. in the context of setting exposure limits, but up-to-date exposure–response relationships for noise-induced sleep disturbances are missing for many noise sources and countries. Sample size calculations are needed to power field studies on the effects of noise on sleep adequately, but are complicated by the fact that each investigated subject is exposed to multiple noise events (i.e., data are clustered within subjects). Using existing data of a field study on the effects of aircraft noise on polysomnographically determined awakenings and a bootstrap approach, we performed Monte Carlo simulations for the power calculations. We estimated the power of the study to find a significant effect of maximum sound pressure level and the precision of a 95% confidence interval surrounding exposure–response relationships depending on the number of investigated subjects and the number of noise events each subject will be exposed to. Functions were fit to the simulation results that can be used by researchers to conveniently power future studies on the effects of aircraft noise on sleep.