Road traffic noise: self-reported noise annoyance versus GIS modelled road traffic noise exposure

Abstract

Background and objectives: self-reported road traffic noise annoyance is commonly used in epidemiological studies for assessment of potential health effects. Alternatively, some studies have used geographic information system (GIS) modelled exposure to road traffic noise as an objective parameter. The aim of this study was to analyse the association between noise exposure due to neighbouring road traffic and the noise annoyance of adults, taking other determinants into consideration. Methods: parents of 951 Munich children from the two German birth cohorts GINIplus and LISAplus reported their annoyance due to road traffic noise at home. GIS modelled road traffic noise exposure (L_den, maximum within a 50 m buffer) from the noise map of the city of Munich was available for all families. GIS-based calculated distance to the closest major road (≥10 000 vehicles per day) and questionnaire based-information about family income, parental education and the type of the street of residence were explored for their potential influence. An ordered logit regression model was applied. The noise levels (L_den) and the reported noise annoyance were compared with an established exposure–response function. Results: the correlation between noise annoyance and noise exposure (L_den) was fair (Spearman correlation r_s = 0.37). The distance to a major road and the type of street were strong predictors for the noise annoyance. The annoyance modelled by the established exposure–response function and that estimated by the ordered logit model were moderately associated (Pearson's correlation r_p = 0.50). Conclusions: road traffic noise annoyance was associated with GIS modelled neighbouring road traffic noise exposure (L_den). The distance to a major road and the type of street were additional explanatory factors of the noise annoyance appraisal.