Estimating the Expected Health Benefits to Non-Smokers of Policies to Reduce Exposure to Environmental Tobacco Smoke

Objectives: To estimate the effects of (a) exposure to environmental tobacco smoke (ETS) on cotinine intake and (b) the effect of cotinine intake on health and health care utilisation, and thereby simulate the effect of policy intervention. Data: Over 16,000 non-smoking respondents to the 1995, 1998 and 2003 Scottish Health Surveys who participated in a follow-up nurse visit at which biomedical tests were undertaken. Over 90% gave consent to having their data linked to subsequent hospital records. Methodology: We first regressed cotinine values on self-reported exposure to ETS in the home, at work, pubs and other public places. We then regressed (a) respiratory function, (b) self-assessed health and (c) the cost of subsequent hospital admissions on cotinine intake, age, height, educational attainment and occupational class. Observations are weighted using inverse probability weights to improve the representation of the general population. Exposure to ETS may be endogenous, so we instrument for exposure using household composition, availability of private and public transport and access to shared social facilities. We simulate the effects of reducing exposure to ETS by combining the estimated parameters for ETS on cotinine intake and cotinine intake on health and healthcare. Results: All sources of exposure to ETS have a significant effect on the level of cotinine. Unsurprisingly, exposure in the respondent's own home has the largest effect. Cotinine intake has a significant negative effect on self-assessed health. Levels of self-reported exposure have decreased over time while both the effects of exposure on cotinine and cotinine on ill-health have increased in magnitude over time. Using EQ-5D utility values to scale the changes in self-assessed health suggests that a total eradication of exposure to ETS in public places would increased the average quality of life of non-smokers by 0.001 QALYs per year. Conclusion: Exposure to ETS has a statistically significant effect on a range of health outcomes. The health gains from a total eradication of ETS exposure in public places are small on a per-capita basis but substantial when applied to entire non-smoking populations.