Figure 2.

Adjusted HRs for all-cause and cardiopulmonary mortality according to years since cessation of solid fuel use for cooking among female never-smokers and male ever-smokers

HRs were derived from Cox models for all-cause mortality (A, D), cardiovascular mortality (B, E), and respiratory mortality (C, F), stratified by study area and adjusted for age at baseline, education level, household income, alcohol consumption, passive smoking, physical activity, body-mass index, diet (consumption of fresh fruit, preserved vegetables, and meat), cookstove ventilation, and solid fuel use for heating. Line at 1·0 represents the reference category persistent clean fuel users and previous solid fuel users with >10 years of cessation, indicating participants who had reported always using clean fuels or switching from solid to clean fuels for >10 years. We only present results for female never-smokers (n=116 105; A–C) and male ever-smokers (n=41 506; D–F) because of the small sample size of female ever-smokers (n=5261) and male never-smokers (n=8805). Solid red lines show HR estimates and shaded areas show 95% CIs. HR=hazard ratio.