Objective Incidence of type 2 diabetes mellitus (T2DM) in the UK has increased by two-thirds in the last decade. Understanding the reasons for the unfavourable trend may help inform efforts to curb future increases. The association between T2DM and adiposity has been established. However, few attempts have been made to quantify the contribution of adiposity changes to the observed time trend in T2DM. We estimated the proportion of the time trend in T2DM incidence in British men over 24 years that may be explained by increasing population body mass index (BMI) levels.

Design Longitudinal study.

Setting 24 British towns.

Participants 6460 men from a socially and geographically representative cohort of older British men, followed up for doctor-diagnosed T2DM incidence between 1984 (aged 45 to 65 years) and 2007 (aged 67 to 89 years) inclusive. Men with a diagnosis of diabetes before 1984 were excluded.

Main outcome measures The age-adjusted increase in the hazard of T2DM between 1984 and 2007 and the contribution of changing BMI to this increase, derived from Cox proportional hazards modelling with time-updated covariates.

Results Between the periods 1984–1992 and 1999–2007, the age-adjusted hazard of T2DM more than doubled (hazard ratio 2.33, 95% CI 1.75 to 3.10). An estimated 26% (95% CI 17 to 38) of this hazard increase could be statistically explained by a population-averaged age-adjusted increase in BMI from 1984 to 1999 of 1.42 kg/m² (95% CI 1.10 to 1.74). Earlier and later portions of the follow-up were examined separately. Between 1984–1992 and 1992–1999, T2DM incidence increased by half (hazard ratio 1.59, 95% CI 1.23 to 2.05). 22% (95% CI 12 to 48) of this increase could be explained by rising BMI. Between 1992–1999 and 1999–2007, a similar increase in T2DM incidence was observed (hazard ratio 1.47, 95% CI 1.17 to 1.84). 31% (95% CI 17 to 81) of this increase was explained by BMI.

Conclusion BMI changes can account for an appreciable portion of the increase in T2DM. Control and reversal of rising adiposity levels is therefore an important priority in controlling the diabetes epidemic. There remained a large “unexplained” portion of the T2DM increase. This may reflect imprecision of BMI as a measure of visceral fat or the independent contributions of changes in other determinants, such as physical activity and diet. The presence of other contributing factors would suggest the need for a more multi-factorial approach to combat rising T2DM in the population.