This study attempts to establish a concentration response (CR) function, which describes the relation between air pollution (PM10) and premature deaths, and to quantify the monetary benefit attributed to PM10 reduction. This is the first effort to estimate the CR function in Dhaka city. This study conducted a daily time series analysis of premature mortality with PM10, temperature, humidity and seasons. Both the Poisson and ordinary least square (OLS) regressions were used and these results were compared. Estimated health impacts by PM10 were converted to monetary value, using the willingness to pay (WTP) approach, while considering the estimated per capita income of Dhaka city and income elasticity of health benefits. It was estimated that a 0.65% (95% confidence interval of 0.62 to 0.68%) change in all causes mortality is associated with 10 mg/m3 change in PM10. If annual concentration of PM10 reduced to national standard from its 2004 level, approximately 1,210 premature deaths (95% CI of 1,160-1,270) could be avoided. This benefit was converted to monetary term as US $97 million, which is equivalent to 4.1% of GNI of Dhaka city.