Modeling quality-adjusted life expectancy loss resulting from tobacco use in the United States

Purpose: To describe the development of a model for estimating the effects of tobacco use upon Quality Adjusted Life Years (QALYs) and to estimate the impact of tobacco use on health outcomes for the United States (US) population using the model. Method: We obtained estimates of tobacco consumption from 6 years of the National Health Interview Survey (NHIS). In addition, NHIS data were used to impute the Quality of Well-Being (QWB) Scale using a new methodology known as QWBX1. The QWB places health status on a continuum ranging from death (0.0) to full functioning without symptoms (1.0). The method allows the adjustment of life expectancy for reduced quality of life associated with health conditions. NHIS data were matched to the National Death Index for 14,464 deaths occurring by December 31, 1997. The analysis is limited to adults between the ages of 18 and 70 years. Results: Quality of Well-Being scores were broken down by age and for six smoking categories: (1) non-smokers, (2) those who smoke 1-10 cigarettes per day, (3) 11-20 cigarettes per day, (4) 21-30 cigarettes per day, and (5) 31-40 cigarettes per day, and (6) 40 or greater cigarettes per day. There was a systematic relationship between current tobacco use and health-related quality of life at each point along the age spectrum and there was a clear and systematic separation of quality-adjusted life expectancy by number of cigarettes smoked per day. Teenagers who continue to smoke loose 3.5 QALYs between ages 18 and 70 in comparison to non-smokers. A greater portion in the loss in QALE is attributable to quality of life than to shorten life expectancy. Conclusions: The overall goal of Healthy People 2010 is to increase Years of Healthy Life (or QALE) in the United States. Each year, tobacco use results in hundreds of thousands of quality-adjusted life years lost. Combined models of morbidity and mortality incorporating a range of tobacco consumption levels are required to best represent the impact of tobacco use.