CDC's April 2002 Report On Smoking:
Estimates of Selected Health Consequences of Cigarette Smoking Were Reasonable
GAO-03-942R: Published: Jul 17, 2003. Publicly Released: Aug 18, 2003.
- Accessible Text:
Despite a recent decline in the population that smokes, smoking is considered the leading cause of preventable death in this country. According to the Centers for Disease Control and Prevention (CDC), over 2 million deaths in the 5-year period from 1995 through 1999 were attributable to cigarette smoking. CDC, part of the Department of Health and Human Services (HHS), is a primary source of information on the health consequences of smoking tobacco. CDC reported its most recent estimates of selected health consequences of cigarette smoking in an April 2002 issue of its publication Morbidity and Mortality Weekly Report. CDC reported that, on average, over 440,000 deaths, 5.6 million years of potential life lost, $82 billion in mortality-related productivity losses, and $76 billion in medical expenditures were attributable to cigarette smoking each year from 1995 through 1999. CDC and others tasked with making such estimates face challenges. They build estimates on a set of assumptions and make choices about the data sources and methods used, each of which may have limitations that must be weighed against its advantages. Policymakers at both the state and federal levels have relied on estimates like these in considering bans on smoking in public places, taxes on cigarettes, litigation to recoup medical expenditures, and other matters concerning tobacco. Thus it is essential that the estimates CDC provides are sound and that their limitations are clear. In recognition of this, Congress asked us to review CDC's April 2002 report and determine whether its estimates of selected health consequences of cigarette smoking were reasonable. Specifically, we examined CDC's estimates of (1) deaths and years of potential life lost and (2) mortality-related productivity losses and medical expenditures attributable to cigarette smoking.
In summary, CDC's estimates of the average number of deaths and years of potential life lost each year due to cigarette smoking were reasonable. The estimates were based on the increases in deaths from 23 causes that were linked to cigarette smoking. The linkages of cigarette smoking to increased mortality due to the included causes, such as lung cancer or cardiovascular disease, had been well established by the Surgeon General. CDC used the method generally accepted among epidemiologists for estimating the increased deaths attributable to cigarette smoking. The data sources CDC used were the best available and included: the largest study of smoking behavior and health status available for data on the risk of death in smokers relative to nonsmokers; the National Health Interview Survey of over 97,000 persons for data on the prevalence of smoking; and death certificates compiled from all states for mortality data. CDC recognized and handled appropriately the limitations in the data from these sources. CDC's estimates of the annual mortality-related productivity losses and medical expenditures due to cigarette smoking also were reasonable. CDC estimated productivity losses associated with the years of potential life lost using assumptions about employment and earnings that are generally accepted among economists, well-established methods for extrapolating from present earnings to earnings that would be made in the future, and large federal data sources on earnings. The assumptions that CDC made and the methods it used to estimate medical expenditures were also generally accepted among health care economists. CDC relied on the most comprehensive data available on medical expenditures, the federally sponsored National Medical Expenditure Survey of over 38,000 persons. For both productivity losses and medical expenditures, CDC recognized and handled appropriately the limitations in the data.