Time-series analysis of respiratory and cardiovascular morbidity in Nicosia, Cyprus, 1995-2004: the effect of air pollution and dust storms

Abstract

Background - A substantial body of research has shown adverse health effects of short-term changes in levels of air pollution. Associations have not been investigated in smaller size Mediterranean cities where re-suspended dust from desert regions can raise particle concentrations considerably above European guidelines. It is unclear whether such natural phenomena pose the same risks. Methods – The effect of daily changes in particulate matter (PM) and ozone on hospital admissions was investigated in generalized additive Poisson models controlling for the effect of weather. Meteorological records and wind trajectories were reviewed to identify dust-storm days and analyses were repeated to quantify their effect. Results – An increasing risk of hospitalisation was observed across quartiles of days with increasing levels of PM. For a 10μg/m3 increase, there was a 0.9% 95%CI (0.6, 1.2) increase in all and 1.2% (-0.0, 2.4) increase in cardiovascular admissions. With respect to respiratory causes, the effect was stronger but restricted to the warm season and may indicate a synergy with temperature. Admissions were 4.8% (0.7, 9.0), 10.4% (-4.7, 27.9) and 3.1% (-10.2, 18.3) higher on dust storm days for all, cardiovascular and respiratory causes respectively. Conclusions – Estimates of the effect of daily changes in air pollution were generally consistent with those seen elsewhere. There was also some evidence of increased admissions on dust-storm days. While inference is limited, the magnitude was at least comparable to that seen on days with the highest levels of PM from traffic sources and may, thus, merit special warnings to vulnerable groups.