We performed the cohort study to evaluate the association between BMI, high-sensitivity C-reactive protein (hs-CRP) and the conversion from metabolically healthy to unhealthy phenotype in Chinese adults.

Metabolically healthy was defined as participants without history of metabolic diseases and with normal fasting blood glucose level, glycated Hb A1c level, blood pressure, lipid profile, serum uric acid level and liver ultrasonographic findings at baseline. Participants were either classified into normal weight (18·5 ≤ BMI < 24·0 kg/m2) and overweight (BMI ≥ 24·0 kg/m2) based on baseline BMI, or low (<1 mg/l) and high (≥1 mg/l) groups based on baseline hs-CRP. The conversion from metabolically healthy to unhealthy phenotype was deemed if any of the metabolic abnormalities had been confirmed twice or more during 5 years of follow-up.

Included were 4855 (1942 men and 2913 women, aged 36·0 ± 8·9 years) metabolically healthy Chinese adults. We identified 1692 participants who converted to metabolically unhealthy phenotype during the follow-up. Compared with their counterparts, the adjusted hazards ratio of the conversion was 1·19 (95 % CI 1·07, 1·33) for participants with overweight, while it was 1·15 (95 % CI 1·03, 1·29) for those with high hs-CRP level (≥1 mg/l). Further adjustment of hs-CRP did not materially change the association between BMI and the conversion. However, the association between hs-CRP and the conversion was not significant after further adjustment of BMI. The sensitivity analysis generated similar results to main analysis.

BMI was associated with the risk of the conversion from metabolically healthy to unhealthy status in Chinese adults.

To study the demographic and clinical parameters of three different categories of obesity, with particular focus on a cohort of individuals with BMI ≥ 50 kg/m2, the fastest growing category of obesity.

Over 700 obese individuals were studied (186 with BMI = 30–39 kg/m2, 316 with BMI = 40–49 kg/m2 and 290 with BMI ≥ 50 kg/m2).

Median BMI was 51 kg/m2 for patients who reported onset of overweight before 15 years of age, 47 kg/m2 for patients who reported onset between 15 and 30 years, and 42 kg/m2 for patients who became overweight after 30 years of age. The BMI ≥ 50 kg/m2 group was notably younger than the group with BMI = 30–39 kg/m2 (44 (sd 11) years v. 50 (sd 15) years; P < 0·0001). Eighteen per cent of obese patients studied were considered metabolically healthy according to standard cut-off points for blood pressure, fasting glucose and lipid profiles. However, the proportion of metabolically healthy individuals was significantly higher in the BMI = 30–39 kg/m2 group than in the BMI = 40–49 kg/m2 and BMI ≥ 50 kg/m2 groups (31 % v. 17 % and 12 % respectively; P < 0·05 and P < 0·005). When compared with people of similar age in the general population, individuals with BMI ≥ 50 kg/m2 had lower rates of marriage (51 % v. 72 %) and a higher prevalence of unemployment (14 % v. 5 %).

The current study suggests that the increasing prevalence of childhood obesity worldwide will lead to many more individuals achieving a higher BMI at a younger age. Furthermore, an earlier onset of overweight does not appear to prevent the adverse metabolic health outcomes associated with extreme obesity.