摘要：Objectives. We used 2 approaches based on published information to compare the impacts on leukemia incidence and benzene exposure of the 1990 US Clean Air Act (CAA) amendments and smoking prevention and cessation efforts. Methods. We extrapolated leukemia mortality related to community air pollution levels and to cigarette smoking from data from the US Environmental Protection Agency and the US Surgeon General. We also estimated relative decline in total exposures to benzene (a known human leukemogen) owing to the CAA amendments and to smoking prevention and cessation efforts. Results. We estimated that because of the CAA, there will be approximately 300 fewer leukemia deaths in the United States during the period 2000 through 2020. During the closest comparable period (1987–2007), we estimated that decline in cigarette smoking led to 7120 fewer leukemia deaths, of which 1282 to 3702 were attributable to benzene. Similarly, the decline in smoking led to about a tenfold greater decrease in total-population benzene exposure than did the 1990 CAA amendments. Conclusions. Both the CAA and smoking cessation activities contribute to a decrease in leukemia incidence. Smoking cessation activities have had a greater effect in the past. A systems approach to public health, including its incorporation as a cross-cutting competency in public health education and in the new core public health credentialing examination, has been advocated. 1 – 3 One of the many challenges to a systems approach is to provide information relevant to understanding the relative efficacy of disparate attempts to deal with the cause of a disease, particularly if these preventive approaches involve multiple governmental and nongovernmental agencies. We provide an example of these complexities by using available literature to provide 2 separate approaches to extrapolate the effects of the 1990 US Clean Air Act (CAA) amendments and of smoking cessation and prevention activities on leukemia incidence. We recognize that regulatory activities under the CAA differ substantially from smoking cessation efforts, and that both have objectives well beyond the reduction of leukemia incidence. However, addressing the complexities of virtually all public health causation issues is confounded by overlapping goals and benefits. The need for comparative information often is particularly important to local health departments—which, in this case, must prioritize between tobacco control efforts and more stringent control of local sources of air toxics than is required by federal and state mandates. Benzene is a ubiquitous component of our petrochemical era, and the only known human leukemogen present in gasoline. 4 It has been regulated under the CAA as a hazardous air pollutant since 1984, and its use is also governed by the newer 1990 CAA amendment regulations. Many common CAA regulatory approaches, ranging from the collar on gasoline pumps to emission controls required of the petrochemical industry, are based to a significant extent on protecting the public from hematological cancer caused by benzene. Regulation has resulted in a decline in stationary and mobile emissions of benzene and of outdoor benzene concentrations. 5 , 6 Benzene also is causally related to other myeloid hematological neoplasms, such as myelodysplasia and myeloproliferative disorders, and to lymphoproliferative disorders including multiple myeloma and non-Hodgkins lymphoma. 7 , 8 However, in contrast to data on acute myeloid leukemia (AML), the data for these other hematological conditions have not been sufficient for the Environmental Protection Agency (EPA) to assign a cancer potency factor linking benzene dose to cancer risk. The centrality of benzene to the control of hazardous air pollutants, as well as extensive knowledge of its health effects, led to this substance being the first chosen for a comprehensive review to meet the EPA's required cost–benefit analysis of its Hazardous Air Pollutant program under Section 812 of the CAA amendments. 9 Benzene is produced not only by evaporative emissions from petrochemicals but also through the combustion of hydrocarbons. For cigarette smokers, the major source of benzene is through the inhalation of cigarette smoke. 10 Secondhand smoke is usually the major source of benzene exposure for those who live or work with cigarette smokers. 11 , 12 In 1990, the US Surgeon General reported that although smoking had been implicated in causing leukemia, the evidence was inconsistent and inconclusive. However, with the advent of further studies, the next Surgeon General's report on the health consequence of smoking, in 2004, concluded that the evidence was sufficient to infer a causal relationship between smoking and AML, and that the risk increased with the number of cigarettes smoked. 13 , 14 We used 2 different approaches to compare the relative impacts of CAA activities and of smoking cessation and control efforts on risk of AML. Our first approach was to estimate the decrease in leukemia risk on the basis of (1) Centers for Disease Control and Prevention (CDC) estimates of the extent of leukemia caused by cigarette smoking and (2) EPA analysis of the effectiveness of the CAA amendments in reducing leukemia risk in the Houston–Galveston metropolitan statistical area (MSA). 9 Our second approach was to estimate the reduction in total-population benzene exposure achieved through these 2 preventive modalities. We also estimated the further impact that additional control of these 2 major sources of human benzene exposure could have on the incidence of leukemia.
