The metabolically healthy obese (MHO) phenotype may express typical characteristics on long-term follow-up. Little is known about the initiation of this phenotypes and its future stability.

The Clinical Practice Research Datalink (CPRD) is a large-scale primary care database. The aim of this study was to assess the stability of, and evaluate the factors associated with a transition into an unhealthy outcome in, a MHO population in the UK.

The CPRD was interrogated for a diagnosis of ‘obesity’ and cross-referenced with a body mass index (BMI) ≥35 kg/m²; participants were further classified as MH using a clinical diagnostic code or a relative therapeutic code. A hazard cox regression univariate and multivariate analysis evaluated the time to transition for independent variables.

There were 231,399 patients with a recorded BMI of 35 kg/m² or greater. Incomplete records were eliminated and follow-up limited to 300 months, the cohort was reduced to 180,560 patients. The prevalence of MHO within the obese population from the CPRD was 128,191/180,560 (71%). MHO individuals, who were of male gender (hazard ratio (HR) 1.23 (1.21–1.25), p = < 0.01), older age group (HR 3.93 (3.82–4.04), p = < 0.01), BMI of 50–60 kg/m² at baseline (HR 1.32(1.26–1.38), p = 0.01), smokers (HR 1.07(1.05–1.09), p = < 0.01) and regionally from North West England (HR 1.15(1.09–1.21), p = < 0.01) were more prone to an unhealthy transition (to develop comorbidities). Overall, of those MH at baseline, 71,485/128,191(55.8%) remained healthy on follow-up, with a mean follow-up of 113.5 (standard deviations (SD) 78.6) months or 9.4 (SD 6.6) years.

From this unique large data set, there is a greater prevalence of MHO individuals in the UK population than in published literature elsewhere. Female gender, younger age group, and lower initial weight and BMI were found to be significant predictors of sustained metabolic health in this cohort. However, there remains a steady progressive transition from a healthy baseline over the years.