Does HIV Infection Independently Increase the Incidence of Lung Cancer?

Abstract

SIR—A number of studies have shown that persons with HIV infection are at higher risk for lung cancer than is the general population [1–6]. Most studies have used administrative data to assemble cohort studies of persons with HIV infection and have compared the observed incidence of lung cancer to the expected incidence, according to the age and sex of cohort subjects, using population-based cancer surveillance data to calculate a standardized incidence ratio. These ratios range from 1 to 7.4. Unfortunately, such studies have not been able to adjust for cigarette smoking. Persons with HIV infection have a higher prevalence of smoking than does the general population. For example, among a nationally representative sample of persons receiving care for HIV infection in the United States in the late 1990s, 73% had ever smoked and 51% currently smoked (personal communication, S. Bozzette), which is much higher than the 20%–30% of US adults who smoke [7]. Although it is recognized that a bias is created by the inability to adjust for cigarette smoking, such a bias has not been quantified. We estimated the magnitude of the error introduced to the standardized incidence ratio by calculating the incidence rate ratio of lung cancer one would find when comparing a heavy-smoking HIV-infected population to the general population.