Supplementary Material for: Weight change and ischemic stroke risk among participants in the Atherosclerosis Risk in Communities (ARIC) Study

Introduction Conflicting findings exist between weight change and ischemic stroke risk. Studying the association between body weight change and ischemic stroke risk helps clarify the true nature of the association and supports future health promotion and stroke prevention strategies. Methods We longitudinally assessed data from 10,985 ARIC visit 4 participants (1996-1998). After excluding individuals with missing data on previous stroke or coronary heart disease at baseline (visit 4), we classified 9-year weight change (visit 4 minus visit 1 weight) into quintiles of weight change and weight loss (> -2.7 kg), no change (-2.7 to +2.7 kg), and weight gain (> +2.7 kg) categories. We used crude and adjusted Cox regression models to assess ischemic stroke hazard. We also performed an analysis stratified by BMI status to see if the weight change–stroke risk relationship differed by baseline BMI. Results Among 9,574 participants, 676 developed ischemic stroke during the 20-year follow-up. Most participants at baseline were female (58.25%) and drinkers (50.52%), with mean age of 62 and mean BMI of 28.78 kg/m2. Compared to participants with no change, those who gained weight had 23% lower hazards of ischemic stroke (hazard ratio (HR) = 0.77 (95% confidence interval (CI) = 0.60, 0.99)), while those who lost weight had 30% higher hazards (HR=1.30 (95% CI =1.05,1.62)). Conclusion Weight change showed minimal association with stroke risk overall, with moderate weight gain potentially lowering the risk, while weight loss increased it. These results emphasize the intricate relationship between weight dynamics and cerebrovascular health and the potential complex implications of the degree and direction of weight change for stroke prevention.