Eighteen year weight trajectories and metabolic markers of diabetes in modernising China

AIMS/HYPOTHESIS: Although obesity is a major risk factor for diabetes, little is known about weight gain trajectories across adulthood, and whether they are differentially associated with metabolic markers of diabetes. METHODS: We used fasting blood samples and longitudinal weight data for 5,436 adults (5,734 observations, aged 18–66 years) from the China Health and Nutrition Survey (1991–2009). Using latent class trajectory analysis, we identified different weight gain trajectories in six age and sex strata, and used multivariable general linear mixed effects models to assess elevated metabolic markers of diabetes (fasting glucose, HbA(1c), HOMA-IR, insulin) across weight trajectory classes. Models were fitted within age and sex strata, and controlled for baseline weight (or baseline weight by weight trajectory interaction terms), height, and smoking habit, with random intercepts to control for community-level correlations. RESULTS: Compared with weight gain, classes with weight maintenance, weight loss, or a switch from weight gain to loss had lower values for metabolic markers of diabetes. These associations were stronger among younger women (aged 18–29 and 30–39 years) and men (18–29 years) than in older (40–66 years) men and women. An exception was HOMA-IR, which showed class differences across all ages (at least p < 0.004). CONCLUSION: Trajectory analysis identified heterogeneity in adult weight gain associated with diabetes-related metabolic markers, independent of baseline weight. Our findings suggest that variation in metabolic markers of diabetes across patterns of weight gain is masked by a homogeneous classification of weight gain. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1007/s00125-014-3284-y) contains peer-reviewed but unedited supplementary material, which is available to authorised users.