Fig. 1From: Revisiting sequential attributable fractionsHypothesized causal Bayesian network describing direct and indirect effects pertaining to causal risk factors and associated confounders for stroke. Abbreviations for variables in the causal graph are as follows. Sex: gender of participant; Region: Geographic area of participant either Western Europe, North America, Africa, South Asia, China, South America and South East Asia; Educ: years of education (None, less than 8, 9-12, more than 12); Stress: Summary variable for psychological stress (yes or no); Smoke: smoking status (current, ex-smoker, never smoker); Diet: AHEI diet score (in tertiles); Exer: physically active (yes or no); Alcoh: alcohol consumption (none; moderate; binge drinker); lipids: Apolipoproterion B/Apolipoprotein A1 ratio (in tertiles); WHR: waist hip ratio (in tertiles); HBP: clinically diagnosed high blood pressure (yes or no); HD (history of risk factors for heart disease - yes or no); DM (clinically diagnosed diabetes mellitus or measured Hba1c level at least 4.5 - yes or no)Back to article page