Relative effects of air pollution on lungs and heart.

IIt has been assumed that damage from exposure to tobacco smoke and other particulate air pollutants is imposed primarily on the lungs and is associated with increased morbidity and mortality rates in patients with preexisting lung disease. This is supported by a considerable amount of previous data, such as the mortality data from the December 1952 London smog disaster, which may have caused as many as 12 000 deaths, almost all in patients with preexisting lung disease.1 Total suspended particulate matter (PM) was as high as 3000 g/m. Similar patterns of elevated morbidity and mortality rates, primarily in patients with preexisting lung disease, have been documented in other acute episodes of air pollution in the past.2 However, evidence from the past 10 years shows that sudden increases in ambient air pollution can also rapidly raise morbidity and mortality rates in patients with existing cardiovascular disease, as much or more than the rise associated with lung disease. In the present issue of Circulation, Pope and associates3 report interesting new data on the effects on mortality rate of long-term differences, as opposed to sudden transient increases, in levels of air pollution. Data were derived from a large, comprehensive study initiated by the American Cancer Society and linked to cancer prevention. The study involved a large population of subjects enrolled in 1982 from metropolitan centers in all 50 states and Puerto Rico. Vital status of participants was collected every 2 years for the subsequent 16 years, and a cause of death was identified for 98% of the known fatalities. Metropolitan area of residence was known for each participant, and particle counts of fine particulates were averaged over quarterly intervals throughout the year for each included metropolitan area. A questionnaire provided additional data, allowing for differences in mortality rate in response to pollution to be controlled for potentially confounding differences in age, sex, race, smoking, education, marital status, body mass, alcohol use, occupational exposures, and diet.