Patterns of weight change after the diagnosis of type 2 diabetes in Scotland and their relationship with glycaemic control, mortality and cardiovascular outcomes: a retrospective cohort study

OBJECTIVES: To determine weight change patterns in Scottish patients 2 years after diagnosis of type 2 diabetes and to examine these in association with medium-term glycaemic, mortality and cardiovascular outcomes. SETTING: Using a retrospective cohort design, ethical approval was obtained to link the Scottish diabetes care database to hospital admission and mortality records. PARTICIPANTS: 29 316 overweight/obese patients with incident diabetes diagnosed between 2002 and 2006 were identified with relevant information for ≥2 years. PRIMARY AND SECONDARY OUTCOME MEASURES: Weight records over time provided intrapatient weight change and variation and glycated haemoglobin (HbA1c) gave measures of glycaemic control. These characteristics and demographic variables at diagnosis were linked with notifications of death (2–5 years after diagnosis) and cardiovascular events (0–5 year after diagnosis). RESULTS: By 2 years, 36% of patients had lost ≥2.5% of their weight. Increasing age, being female and a higher body mass index at diagnosis were associated with larger proportions of weight lost (p<0.001). Multivariable modelling showed that inadequate glycaemic control at 2 years was associated with being younger at baseline, being male, having lower levels of obesity at diagnosis, gaining weight or being weight stable with weight change variability, and starting antidiabetic medication. While weight change itself was not related to mortality or cardiovascular outcomes, major weight variability was independently associated with poorer survival and increased cardiovascular outcome risks, as was deprivation. CONCLUSIONS: Our results suggest that weight loss or being weight stable with little weight variability early after diabetes diagnosis, are associated with better glycaemic control and we identified groups less able to lose weight. With respect to mortality and cardiovascular outcomes, although weight change at 2 years was a weak predictor, major weight variability appeared to be the more relevant factor.