Transition Patterns of Weight Status and Their Predictive Lipid Markers Among Chinese Adults: A Longitudinal Cohort Study Using the Multistate Markov Model

Abstract

Xiao Tang,1 Hongya Zhang,1 Yanxiang Zhao,2 Fang Lei,1 Qigui Liu,1 Dongmei Hu,1 Guorong Li,1 Guirong Song1 1Department of Health Statistics, School of Public Health, Dalian Medical University, Dalian City, Liaoning Province, 116044, People’s Republic of China; 2Department of Mathematics, George Washington University, Washington, DC, USACorrespondence: Guirong SongDepartment of Health Statistics, School of Public Health, Dalian Medical University, No. 9 South Road, Lvshun District, Dalian City, Liaoning Province, 116044, People’s Republic of ChinaTel [phone]Email [email protected]: Obesity is well recognized as a risk factor for cardiometabolic diseases. The development of obesity is a dynamic process that can be described as a multistate process with an emphasis on transitions between weight states. However, it is still unclear what convenient biomarkers predict transitions between weight states. The aim of this study was to show the dynamic nature of weight status in adults stratified by age and sex and to explore blood markers of metabolic syndrome (MetS) that predict transitions between weight states.Methods: This study involved 9795 individuals aged 18 to 56 at study entry who underwent at least two health check-ups in the eight-year period of study in the health check-up centre of our institution. Weight, height and biochemical indices were measured at each check-up. The participants were divided into four groups based on age and sex (young male, middle-aged male, young female and middle-aged female groups). A multistate Markov model containing 3 states (healthy weight, overweight and obesity) was adopted to study the longitudinal weight data.Results: Young people were more likely to transit between weight states than middle-aged people, and middle-aged people were more resistant to recover from worse states. The mean sojourn time in obesity was greatest in the middle-aged male group (6.23 years), and the predicted rate of obesity beginning with healthy weight was greatest in the young male group (13.7%). In multivariate models, age group and triglyceride (TG) and high-density lipoprotein cholesterol (HDL) levels were significant for specific transitions in females, whereas age group and HDL levels were significant in males. In females, if HDL levels increased 1 mmol/L, the probability of progression from healthy weight to overweight decreased by 37.0% (HR= 0.63), and the probabilities of recovery (overweight to healthy weight and obesity to overweight) increased by 62.0% (HR= 1.62) and 1.23-fold (HR= 2.23), respectively. In males, if TG levels increased 1 mmol/L, the risk of progression from healthy weight to overweight increased by 24.0% (HR= 1.24). Each unit increase in HDL levels was associated with a 0.99-fold (HR= 1.99) increase in the chance of recovery from overweight to healthy weight and with a 0.37-fold (HR= 0.63) decrease in the risk of progression from healthy weight to overweight.Conclusion: The weight status of young people was less stable than that of middle-aged people. Males were more likely to become overweight and more resistant to recover from worse states than females. Young males with healthy weight were more likely to develop obesity than other healthy weight groups. Blood lipid levels, especially HDL, were predictors of weight transitions in adults. Prevention and intervention measures should be applied early.Keywords: obesity, multistate Markov model, triglyceride, high-density lipoprotein cholesterol, low density lipoprotein cholesterol

Keywords

• obesity
• multistate markov model
• triglyceride
• high-density lipoprotein cholesterol
• low density lipoprotein cholesterol