Cause and effect can generally be established only by an intervention in which all extraneous variables are controlled. Usually, this means creating the artificial conditions of a randomised trial. Occasionally, the intervention comes about naturally.

For more than 50 years, mankind has been the subject of one such natural experiment—an intervention of unprecedented scale which has proved both a serious threat to health and an unexpected source of fundamental new understanding. The intervention has been that of extreme weight gain, amounting in adults to some 9 kg over the past generation.¹ The corresponding increase in children has been even greater.² The gains have not been in bone, muscle, or water, but in fat. Body fat—most particularly visceral fat—leads to insulin resistance, and insulin resistance to increasing demands on β-cell reserve.³

Diabetes is a disorder of β-cell failure in which insulin reserves are no longer sufficient to meet demand.⁴ The rise of insulin resistance in contemporary society has served both to increase the incidence of diabetes and to accelerate its onset, so much so that type 2 diabetes, a disorder of middle age just a generation ago, now ranks among the fastest growing disorders of paediatric practice across the industrialised world.⁵ The impact has been extraordinary, but it is not confined to type 2 diabetes. There has been a parallel increase in the incidence of type 1 diabetes,⁶ and an acceleration of its onset corresponding to that observed in type 2.^7,8 Most striking has been the convergence of their phenotypes, making it increasingly difficult for paediatricians to distinguish the types at presentation.⁹

Thirty years ago, type 1 diabetes seemed distinct from type 2. It was predominantly a disorder of childhood, …