[Relationship between waist-to-height ratio and overall and type specific incidence of stroke in adults in China: a prospective study].

To analyze the association between waist-to-height ratio and the overall and type specific incidence of stroke in adults in China. A total of 36 632 people were selected from 60 surveillance sites (25 in urban area and 35 in rural area) in China Chronic Disease Surveillance Project in 2010. The China Chronic Disease Surveillance Project data in 2010 were used as baseline data. A total of 27 762 people were followed up from 2016 to 2017. Cox proportional risk regression model was used to analyze the risk ratio of waist-to-height ratio for the overall and type specific incidence of stroke. Subgroup analyses were performed based on baseline characteristics such as age and sex, and sensitivity analysis was performed by excluding those who died and those with diabetes at baseline survey. A total of 27 112 subjects were included in the stroke analysis, and 1 333 stroke events were observed. A total of 26 907 subjects were included in the ischemic stroke analysis, and 1 128 ischemic stroke events were observed. A total of 25 984 subjects were included in the hemorrhagic stroke analysis, and 205 cases of hemorrhagic stroke were observed. After adjusting for relevant confounders and taking group with waist-to-height ratio of 0-0.45 as a reference, the stroke analysis indicated that in groups with waist-to-height ratio of 0.46-0.49, 0.50-0.54 and ≥0.55 the risk for stroke increased by 21% (=1.21, 95%:1.00-1.46), 26% (=1.26, 95%:1.04-1.53) and 60% (=1.60, 95%:1.29-1.99) respectively. Subgroup analysis revealed that age specific waist-to-height ratio had modification effect on the risk for stroke (interaction =0.001). Ischemic stroke analysis indicated that in groups with waist-to-height ratio of 0.46-0.49, 0.50-0.54 and ≥0.55 the risk for ischemic stroke increased by 30% (=1.30, 95%: 1.05-1.60), 33% (=1.33, 95%: 1.07-1.64) and 61% (=1.61, 95%: 1.26-2.05) respectively. Subgroup analysis revealed that age specific waist-to-height ratio had modification effect on the risk for ischemic stroke (interaction =0.024). Hemorrhagic stroke analysis indicated that in group with waist-to-height ratio of ≥0.55 the risk for hemorrhagic stroke increased by 73% (=1.73, 95%: 1.02-2.94), but the differences in the risk increase in groups with waist-to-height ratio of 0.46-0.49 and 0.50-0.54 were not significant. The sensitivity analysis showed no changes. In the prevention and control of stroke by body weight control, it is necessary to take waist to height ratio as one of the indicators of body weight control. Particular attention needed to be paid to the people aged <50 years with waist-to-height ratio of ≥0.55 as well as those with waist-to-height ratio of <0.5 (i.e., 0.46-0.49).