Objective   To investigate the correlation of waist circumference and its change with hypertension incidence among Chinese residents aged 18 – 65 years for providing evidence to the prevention and control of hypertension.

  Methods  A prospective dynamic cohort was established among the participants of 7 waves of the China Health and Nutrition Survey (CHNS) conducted from 1993 to 2015 across the mainland of China. After screening out pregnant women and those with abnormal waist circumference (< 50 or >150 centimeters cm), the study enrolled 9 980 community residents aged 18 to 65 years who had participated in at least two waves of the survey and were not identified as having hypertension at the baseline survey. The study period was from 1997 through 2015. The individual period of follow-up was from the baseline survey to the last follow-up for non-hypertensives; but for hypertensives, the follow-up ended by the identification of hypertension incidence. Multivariate Cox proportional hazard regression model was used to analyze the influence of baseline waist circumference and its change on hypertension incidence.

  Results  During the follow-up among 5 060 male and 4 920 female participants aged 36.88 ± 11.42 years and with a median follow-up of 7 years, totally 3 132 hypertension incidences were observed and the incidence density was 32.60/1 000 person-years; among the participants at the baseline survey with normal waist circumference (< 85 cm for males and < 80 cm for females), pre-obesity (85 – 89 cm for males and 80 – 84 cm for females), and central obesity ( ≥ 90 cm for males and ≥ 85 cm for females), the number (incidence density) of observed hypertension incidence were 1 983 (26.71/1 000 person-years), 496 (45.78/1 000 person-years), and 653 (59.45/1 000 person-years), respectively. After adjusting for sex, age, education, marital status, residence place, smoking, current alcohol consumption, total physical activity level, fat intake ratio, baseline chronic disease, baseline waist circumference, and the year at first and last follow-up, multivariate Cox proportional hazard regression showed that an increase of 1cm in waist circumference was associated with a 1.7% increased risk of hypertension (hazard risk HR = 1.017, 95% confidence interval 95%CI: 1.011 – 1.024) and an increase of ≥ 5 cm in waist circumference was associated with a 29.5% increased risk of hypertension (HR = 1.295, 95%CI: 1.130 – 1.484) in comparison to normal waist circumference. The results of restricted cubic spline analysis revealed that the increase of 4.9 cm and higher in waist circumference was correlated with the increased risk of hypertension. In the same way after adjusting for confounding factors listed above, the results of multivariate Cox proportional hazard regression indicated that the risk of hypertension increased by 27.2% (HR = 1.272, 95%CI: 1.117 – 1.449) and 77.5% (HR = 1.775, 95%CI: 1.546-2.038) for all the participants with the waist circumference changed from normal to abnormal level and those with persistent abnormal waist circumference compared with the participants with persistent normal waist circumference; the increased risks were also significant for both male participants (HR = 1.191, 95%CI: 1.004 – 1.412 and HR = 1.485, 95%CI: 1.218 – 1.810) and female participants (HR = 1.374, 95%CI: 1.121 – 1.684 and HR = 2.061, 95%CI: 1.691 – 2.513).

  Conclusion  Both baseline abnormal waist circumference and increased waist circumference over time can increase the risk of hypertension incidence. Keeping a waist circumference within the normal range with an increase of less than 5 cm is of great significance for the prevention and control of hypertension among adult population.