Background: Analyses of secular trends in tobacco consumption can provide evidence of potential tobacco-disease relationships and have utility in terms of public health projections and policy. The purpose of this project was to provide a unique set of continuous apparent tobacco product consumption estimates for the United States over the period spanning 1900 through 1990.
Methods: Two U.S. Department of Agriculture data sources provide information on long-term apparent tobacco consumption in the United States; however, differences exist between these data sets. The consumption estimates in these reports were adjusted to a common population base. A 9-year overlap of the data sets was then used to calibrate one data series to the other using inverse regression. Predicted tobacco consumption estimates for the years 1900 through 1944 were then combined with the adjusted 1945-1990 data.
Results: Inverse regression showed a strong linear relationship between the two U.S. Department of Agriculture summaries for each tobacco product during the 9-year overlap period. A continuous set of annual per capita tobacco consumption estimates is reported by product for the United States.
Conclusions: The two U.S. Department of Agriculture reports can be combined to provide a history of tobacco product consumption in the United States over the period 1900-1990.
Copyright 2000 American Health Foundation and Academic Press.