Primary prevention of type 2 diabetes (T2DM) is an important strategy for curbing its rising global burden. Though lifestyle modification has provided an effective method of preventing/delaying incidence of diabetes in high-risk individuals, it has not been widely implemented even in developed countries due to its high-cost, need for expertise and difficulties in translating the benefits of lifestyle intervention to the community at large. Hence, there is an urgent need to identify an alternate mode of delivery to transmit healthy lifestyle information to high-risk individuals. In this trial, we sought to determine whether lifestyle advice through mobile phone text messaging could reduce incident diabetes compared to standard lifestyle advice in Asian Indian men with prediabetes. The study showed for the first time that mobile phone messaging is an effective and acceptable method to deliver advice and support towards lifestyle modification to prevent T2DM in men at high risk.

The identification of novel predictors for T2DM is an arduous task. The glycaemic markers of diabetes (fasting plasma glucose, 2hr post glucose load and HbA1c) are, in fact, risk factors for microvascular complications of diabetes and it was on this basis that diagnostic cut-offs’ for diabetes were arrived at. Nevertheless, elevated levels of glycaemic markers in the sub-clinical, or pre-diabetic, range are associated with increased risk of progression to diabetes. However, there is already considerable deterioration of beta cell function by the time diabetic dysglycaemia occurs. The way forward is clearly to identify biomarkers that serve as reliable predictors of progression to diabetes rather than simply reflecting accompanying levels of glycaemia. In this thesis using the database of the above mentioned trial it was aimed to identify the predictors of T2DM in Asian Indian cohort with prediabetes at baseline.

The classical risk factors studied here are: 1) increased prevalence of the hypertriglyceridemic waist phenotype, 2) a combination of HbA1c and gamma glutamyl transferase and 3) a measure of beta cell compensation (disposition index) predicted incident diabetes. Among these, the disposition index was the most powerful predictor in the cohort.

In addition to these classical risk factors mentioned above, in a small nested-case control, cross sectional study, the association of adipokines (adiponectin, leptin, interleukin-6 (IL-6), retinol-binding protein4 (RBP4)) and vitamin D3 were assessed to study the mechanistic link of novel biomarkers with diabetes. In this cohort, lower levels of baseline adiponectin, and higher IL-6 and RBP4 were associated with diabetes. Though, many of these provided a novel mechanistic pathogenic link with diabetes they did not improve prediction over and above that of glycaemic measures in identifying individuals with diabetes. However, the non-glycaemic biomarkers appear to have a role in the underlying pathogenesis of diabetes.