Association between weight change and remission of type 2 diabetes: a retrospective cohort study in primary care

Purpose: to quantify the association between weight change and the likelihood of remission of type 2 diabetes in a population-based cohort without intensive interventions. Methods: a retrospective analysis of adults with type 2 diabetes registered across 150 primary care practices in Southern England, United Kingdom, within the Electronic Care and Health Information Analytics (CHIA) database between 2013 and 2020. Stepwise mixed logistic models were constructed to examine the association between percentage weight change over five-years, and the likelihood of remission in the subsequent two years using four categories: weight gain ≥2.5%; weight loss ≤ 2.5–5% or ≤5–10% or ≥10%. Results: the study cohort included 56,120 people with type 2 diabetes with a mean (SD) disease duration of 8.0 (6.7) years. 17,035 (30.4%) had microvascular complications and 10,661 (19.0%) had macrovascular complications. A total of 19.4% (10,896) lost ≥10% of their baseline weight with 1646 (15.1%) of these achieving remission. Overall, those who lost ≥10% in weight had a significantly higher likelihood of remission in both unadjusted and adjusted models; RR: 1.53 (95% CI: 1.40-1.68); and RR:1.51 (95% CI: 1.37-1.66), respectively. Remission was still achievable but less likely amongst younger men with a longer diabetes duration (< 5 years), higher baseline HbA1c level, or pre-existing microvascular or macrovascular complications. Conclusion: weight loss of >10% is associated with remission of type 2 diabetes even amongst those with advanced disease and established microvascular or macrovascular complications. Our findings could motivate people with diabetes to lose weight in order to increase likelihood of remission.