Illusion of filtration: Evidence from tobacco industry documents

We compared the design features of popular filtered and non-filtered cigarettes sold in the United States between 1960 and 1990, to assess the relationship between cigarette filter and tobacco weight.

We analyzed data on the design features of six popular filtered and three non-filtered cigarette brands sold in the US including the weight of tobacco used provided in the Cigarette Information Reports produced by Philip Morris Tobacco Company between 1960 and 1990. We also collected information on other design features such as stick length and circumference, the percentage of reconstituted tobacco in the blend, among other product parameters. We used joinpoint regression to test for trends in outcome variables for each brand assessed between 1960 and 1990.

In all years, filtered cigarettes had less tobacco by weight compared to non-filtered cigarettes. The lower average weight of tobacco found in filtered cigarettes appears to be due to a combination of factors including stick and filter length, and the amount of reconstituted tobacco in the blend. The average percentages of total alkaloids and expanded tobacco increased over time but were similar between filtered and non-filtered brands.

While various design features of popular filtered and non-filtered brands changed between 1960 and 1990, the observed reduction in tobacco weight among filtered brands was perhaps the most salient in terms of disease risk. Less tobacco in a filtered cigarette calls into question the presumed exclusive role of cigarette filter tips in the reduced health risks of filtered versus non-filtered cigarette smoking.