Based on the Kailuan Study, this research systematically investigated the association of changes in weight, waist circumference, or both, with the risk of incident heart failure among middle-aged and older adults in China. The findings provide a scientific basis for the precise prevention of heart failure and the management of body composition. The corresponding authors are Assoc. Prof. Yu Yuan from the School of Public Health at Huazhong University of Science and Technology and Prof. Shouling Wu from the Kailuan General Hospital. Yu Yin and Rui Tang, PhD student from the School of Public Health, were the co-first authors.
Heart failure represents a major global public health challenge. Previous studies have largely focused on static baseline measures of body composition, such as body mass index and waist circumference, while overlooking their dynamic changes over time. Moreover, the relationship between the opposite changes in weight and waist circumference and heart failure risk in middle-aged and older adults remains unclear, indicating a significant research gap.
To address this gap and support evidence-based weight management strategies for aging populations, the Yu Yuan team conducted a large-scale epidemiological study using the prospective Kailuan Cohort, which included 45,620 middle-aged and older Chinese adults. With a median follow-up of 9.66 years, the study provided several key findings:
I. Risk Association of Isolated Changes in Body Weight or Waist Circumference: Correcting Misconceptions in Weight Management
1. Body weight change showed a U-shaped association with heart failure risk. Compared to the stable weight group, individuals with excessive weight gain had a 27% higher risk of heart failure. Although not reaching statistical significance, the excessive weight loss group also exhibited a trend toward higher risk. This finding challenges the conventional belief that weight loss is invariably beneficial.
2. Waist circumference change demonstrated a positive linear association with heart failure risk. Compared to the stable waist circumference group, the excessive waist circumference gain group had a 28% higher risk of heart failure while the excessive waist circumference loss group had a 24% lower risk of heart failure. The finding indicates that an increase in waist circumference is an independent risk factor for heart failure.
Table 1. Associations between weight change categories and heart failure.
Model 1: The multivariable models were adjusted for height and weight at cohort recruitment, waist circumference change (continuous variables) and stratified by age at risk (5‐year interval) and sex.
Model 2: The multivariable models were adjusted for height and weight at cohort recruitment, waist circumference change (continuous variables), smoking status, alcohol intake status, dietary pattern, educational attainment, physical activity and stratified by age at risk (5‐year interval) and sex.
Model 3: The multivariable models were adjusted for height and weight at cohort recruitment, waist circumference change (continuous variables), smoking status, alcohol intake status, dietary pattern, educational attainment, physical activity, occupation, hypertension, fasting blood glucose, total serum cholesterol level and stratified by age at risk (5‐year interval) and sex.
Abbreviations: CI, confidence interval; HR, hazard ratio.
Table 2. Associations between waist circumference change categories and heart failure.
Model 1: The multivariable models were adjusted for height and waist circumference at cohort recruitment, weight change (continuous variables) and stratified by age at risk (5‐year interval) and sex.
Model 2: The multivariable models were adjusted for height and waist circumference at cohort recruitment, weight change (continuous variables), smoking status, alcohol intake status, dietary pattern, educational attainment, physical activity and stratified by age at risk (5‐year interval) and sex.
Model 3: The multivariable models were adjusted for height and waist circumference at cohort recruitment, weight change (continuous variables), smoking status, alcohol intake status, dietary pattern, educational attainment, physical activity, occupation, hypertension, fasting blood glucose, total serum cholesterol level and stratified by age at risk (5‐year interval) and sex.
Abbreviations: CI, confidence interval; HR, hazard ratio.
Figure 1. The restricted cubic spline for the association between weight and waist circumference change categories and heart failure.
The curve represents adjusted HRs based on restricted cubic splines with knots at the 5th, 50th and 95th percentiles of the distribution of weight and waist circumference percent change (solid lines are HRs, dashed lines indicate 95% CIs), the reference values (HR = 1) were set at where weight and waist circumference changes were equal to zero. The multivariable‐adjusted models were stratified by age at risk (5‐year interval), sex and adjusted for height, weight change, waist circumference change, smoking status, alcohol intake status, dietary pattern, educational attainment, physical activity, occupation, hypertension, fasting blood glucose and total serum cholesterol level. For the analysis of weight change, we additionally adjusted for weight at cohort recruitment. For the analysis of waist circumference change, we adjusted for waist circumference at cohort recruitment.
II. Risk Stratification Based on Combined Changes in Weight and Waist Circumference: Three High-Risk groups Require Focused Attention
1. Excessive weight loss with stable waist circumference: The risk of heart failure increased by 53%, which suggested that unintentional weight loss in middle-aged and older adults may reflect muscle loss (sarcopenia), potentially increasing cardiac workload.
2. Mild weight loss with excessive waist circumference gain: The risk of heart failure increased by 119%. This is the pattern with the highest risk among all combinations. This phenotype characterizes "sarcopenic obesity", where muscle mass decreases while fat mass accumulates.
3. Excessive weight gain with excessive waist circumference gain: The risk of heart failure increased by 48%, which indicated the synergistic detrimental effects of overall obesity and abdominal obesity on the development of heart failure.
Figure 2. Adjusted hazard ratios for heart failure based on the combined changes in weight and waist circumference.
The multivariable‐adjusted model included the combined categories of weight and waist circumference changes, weight, height and waist circumference at cohort recruitment, smoking status, alcohol intake status, dietary pattern, educational attainment, physical activity, occupation, hypertension, fasting blood glucose, total serum cholesterol level; and stratified by age at risk (5‐year interval) and sex.
Building upon previous work from the Dongfeng-Tongji and Kailuan cohorts, the Yu Yuan team found that among middle-aged and older adults, compared with the group with stable weight and waist circumference, both gain and loss in weight and waist circumference were associated with elevated risk of all-cause mortality. Furthermore, both weight changes and waist circumference changes exhibit the U-shaped associations with the risk of all-cause mortality. Notably, weight gain was consistently associated with higher mortality risk, regardless of changes in waist circumference. Similarly, increases in waist circumference elevated mortality risk across all patterns of weight change. Most strikingly, individuals experiencing weight loss coupled with waist circumference gain demonstrated the highest all-cause mortality risk among all subgroups. These findings were published in JAMA Network Open (5-year Impact Factor: 11.2) under the title "Analysis of changes in weight, waist circumference, or both, and all-cause mortality in Chinese adults" [2]. Prof. Tangchun Wu and Prof. An Pan served as co-corresponding authors, with Assoc. Prof. Yu Yuan and Dr. Kang Liu as co-first authors.
This conclusion corroborates the findings of the current heart failure study, collectively forming a complete evidence chain for the "body composition changes–health risk" relationship. It aligns with the annual weight management goal of "long-term monitoring and continuous management." In body composition management for middle-aged and older adults, equal emphasis should be placed on "weight stability" and "waist circumference control" to avoid blind weight loss, with particular attention to abdominal fat accumulation. It is recommended that waist circumference changes should be incorporated into routine health monitoring indicators, and targeted nutritional interventions and exercise guidance should be provided for high-risk populations.
Reference
[1] Yu Yin, Rui Tang, Xi Wang, Mengyi Zheng, Jingli Qu, Shouhua Chen, Shouling Wu∗, Yu Yuan∗. Changes in weight, waist circumference or both with incident heart failure in Chinese middle-aged and older adults. Journal of cachexia, sarcopenia and muscle. 2025 Oct;16(5):e70059.
[2] Yu Yuan, Kang Liu, Mengyi Zheng, Shouhua Chen, Hao Wang, Qin Jiang, Yang Xiao, Lue Zhou, Xuezhen Liu, Yanqiu Yu, Jiachen Wu, Xiong Ding, Handong Yang, Xiulou Li, Xinwen Min, Ce Zhang, Xiaoming Zhang, Meian He, Yan Zheng, Dianjianyi Sun, Lu Qi, Elena C Hemler, Shouling Wu, Tangchun Wu∗, An Pan∗. Analysis of changes in weight, waist circumference, or both, and all-cause mortality in Chinese adults. JAMA Network Open. 2022 Aug 1;5(8):e2225876.
