Résumé
The aim of the present study was to analyse the short-term relationship between pollen exposure and allergic rhinitis and conjunctivitis (ARC) using drug deliveries in two urban areas in the south of France, from 2004 to 2007. An ARC case was defined as the association of an oral antihistamine drug and a local (nose or eye) antiallergenic drug on the same medical prescription. Generalized additive models and generalized additive distributed lag models, taking into account confounding factors such as meteorological data or days of the week, were computed to estimate the relationship between pollen counts and ARC. This study showed a statistically significant association between pollen concentration in the air and the increase in the number of deliveries in pharmacy for ARC in both urban areas. This relation was statistically significant in at least one of the two studied cities for Artemisia, Plantaginanceae, Poaceae, Cupressaceae, Quercus, Fraxinus, Olea, Pinus and Populus. In this southern area of France where Cupressaceae planting is very developed, relative risks estimated for an interquartile increase in Cupressaceae pollen concentration were 1.09 (1.06–1.13) in Montpellier and 1.03 (1.01–1.04) in Nîmes. These results indicate that Cupressaceae planting has a population health impact and that alternative needs to be developed.