Ross McKitrick takes on smog:

Prof. McKitrick, also a fellow of the conservative Fraser Institute think tank, said he became interested in the air-pollution and health link because of the dramatic estimates of pollution impacts released by government and medical officials, extrapolations of research on the effects of bad air.
A 2008 report by the Canadian Medical Association, for instance, estimated there would be 11,000 hospital admissions and 90,000 emergency-department visits that year – as well as 21,000 deaths – because of the chronic and acute effects of air pollution.
The Guelph economist said he took the same model used to produce Toronto public health’s estimate of hospitalizations and deaths annually and applied it to a period in the 1960s when the city’s pollution levels were much higher. The result was a huge number of air pollution-related deaths, almost as many as the total mortality from all causes at the time, said Prof. McKitrick.

Which raises the question, “Where are the bodies?“

For his study, he looked at breathing-related hospital admissions and pollution levels in 11 Canadian cities between 1974 and 1994. The period covered a time when pollution was worse and there were more dramatic changes in air quality, which he said would produce more robust results.
He also factored in smoking rates, with their obvious effects on respiratory health, and income levels, which he said can also alter hospitalization statistics.
Prof. McKitrick and two colleagues at the University of Strathclyde in Glasgow used the Bayesian method of averaging, which he argued would, in essence, provide a more all-encompassing picture of the situation.
The results were clear: lower smoking rates meant fewer respiratory-health hospital admissions, higher income levels were sometimes linked to slightly more admissions and air-pollution had no effect, he said.