The post-trial analysis of the Indian SMS diabetes prevention study shows persistent beneficial effects of lifestyle intervention.

AIMS: We had shown that mobile phone based text messaging was an effective tool to deliver lifestyle changes among Asian Indian men with a 36% relative risk reduction in incident diabetes over two years. The present analysis investigated whether beneficial effects of intervention on diabetes prevention persisted for an additional three years after withdrawal of active intervention. METHODS: The primary two year randomized controlled trial (2010-2012) compared lifestyle changes with use of automated text messaging reminders in the intervention (n = 271) versus standard care advice (n = 266) at baseline. At the end of the study, both groups received additional advice on lifestyle changes by a trained dietician. Participants free of diabetes (n = 394) were invited three years later to ascertain the sustained effect of intervention. The primary outcome was incidence of type 2 diabetes. This trial is registered with ClinicalTrials.gov,number NCT02848547. RESULTS: During the mean follow-up of 5 years, 346 out of 394 (87.8%) men were reviewed. Incidence of diabetes was reduced by 30% in the intervention group, with declining gap between-group differences over time (Kaplan-Meier analysis). Significant improvement in dietary adherence occurred in the intervention group at 2nd and 5th year follow up (trend χ2 = 21.35, p < 0.0001). Cox regression analysis showed that the 5th year incidence of diabetes was significantly reduced in the intervention group. Higher body mass index and 2 h plasma glucose at 24 months increased the incidence of diabetes. CONCLUSIONS: Sustained reduction in incident diabetes was apparent after cessation of active lifestyle intervention. This was possibly associated with continuing practice of improved lifestyle.

Incidence of Type 2 Diabetes is Higher among Men with Persistent Impaired Glucose Tolerance than in Transient Impaired Glucose Tolerance - A 5 year Follow up Study.

OBJECTIVE: This was a 5 year comparative analysis of the incidence of type 2 diabetes in men who had persistent impaired glucose tolerance (P-IGT) versus transient impaired glucose tolerance (T-IGT). P-IGT (positive IGT on two oral glucose tolerance tests (OGTT), T-IGT (IGT in first OGTT and normal glucose tolerance (NGT) in the 2nd OGTT). METHODS: The samples were collected from a randomized controlled diabetes prevention study. The prevention study was done using lifestyle modification (LSM) promoted by use of mobile short message services (SMS) for 2 years. The control group of the randomized study who received advice on LSM at only the baseline formed the P-IGT group for the 3 years follow up study (n=236). T-IGT (n=569) were available from those who had NGT on the 2nd OGTT while screening for the prevention study. The total diabetes incidence at 5 years in the study groups were compared using standard OGTT (WHO criteria). RESULTS: The conversion rate to diabetes in 5 years was significantly lower among T-IGT than among P-IGT, OR=0.202 (95% CI, 0.145-0.296,p< 0.0001). P-IGT had higher rate of risk factors for diabetes than T-IGT. CONCLUSION: The risk of conversion to diabetes was 80 percent lower in T-IGT than in P-IGT. Identification of P-IGT will help in selecting persons who require early intervention for diabetes.

Baseline level of 30-min plasma glucose is an independent predictor of incident diabetes among Asian Indians: analysis of two diabetes prevention programmes.

BACKGROUND: The objective was to study the ability of the 30-min plasma glucose (30-min PG) during an oral glucose tolerance test to predict the future risk of type 2 diabetes among Asian Indians with impaired glucose tolerance. METHODS: For the present analyses, we utilized data from 753 participants from two diabetes primary prevention studies, having complete data at the end of the study periods, including 236 from Indian Diabetes Prevention Programme-1 and 517 from the 2013 study. Baseline 30-min PG values were divided into tertiles: T1 < 9.1 mmol/L (<163.0 mg/dL); T2 9.2-10.4 mmol/L (164.0-187.0 mg/dL) and T3 ≥ 10.4 mmol/L (≥188 mg/dL). The predictive values of tertiles of 30-min PG for incident diabetes were assessed using Cox regression analyses RESULTS: At the end of the studies, 230 (30.5%) participants developed diabetes. Participants with higher levels of 30-min PG were more likely to have increased fasting, 2-h PG and HbA1c levels, increased prevalence of impaired fasting glucose and decreased beta cell function. The progression rate of diabetes increased with increasing tertiles of 30-min PG. Cox's regression analysis showed that 30-min PG was an independent predictor of incident diabetes after adjustment for an array of covariates [Hazard Ratio (HR):1.44 (1.01-2.06)] CONCLUSIONS: This prospective analysis demonstrates, for the first time, an independent association between an elevated 30-min PG level and incident diabetes among Asian Indians with impaired glucose tolerance. Predictive utility of glycemic thresholds at various time points other than the traditional fasting and 2-h PG values should therefore merit further consideration. Copyright © 2016 John Wiley & Sons, Ltd.

Adiponectin, leptin, interleukin-6 and HbA1c in the prediction of incident type 2 diabetes: A nested case-control study in Asian Indian men with impaired glucose tolerance.

AIMS: The aims of this study were: (1) to assess the association of adiponectin, leptin and interleukin-6 (IL-6) with incidence of type 2 diabetes (T2DM) in Asian Indian men with impaired glucose tolerance (IGT) and (2) to evaluate the additional contribution of these with the well-established glycaemic marker HbA1c. METHODS: This is an ancillary analyses of a nested case-control study derived from a prospective, prevention trial in India. All the participants had IGT at baseline. For this subanalysis a total of 147 (T2DM: 71; nondiabetic: 76) participants were selected based on the final glycemic outcomes. Association of these selected adipokines with T2DM were assessed using logistic regression analyses. Clinical usefulness of adding adipokine markers with HbA1c on prediction of T2DM was assessed using the area under the curve (AUC) of the receiver operating characteristics. RESULTS: Baseline levels of adiponectin were lower and the levels of IL-6 were higher in T2DM cases when compared with non-diabetic cases (P<0.05). Levels of leptin were similar in both groups. In fully adjusted models, adiponectin (odds ratio (OR): 0.55 [95%CI: 0.33-0.91]; P=0.019) and IL-6 (OR: 2.27 [95%CI: 1.40-3.691]; P=0.001) were associated with diabetes. Addition of adiponectin to HbA1c improved the AUC (ΔAUC: 0.0619; P=0.0251), whereas addition of IL-6 did not improve the predictive power of HbA1c alone. CONCLUSIONS: Adiponectin and IL-6 are independently associated with incident diabetes. However, they are unlikely to serve as simple tools to predict future risk of diabetes but may have a role in understanding the pathogenesis.

Retinol binding protein-4 predicts incident diabetes in Asian Indian men with prediabetes.

The association of retinol binding protein-4 (RBP4) with incident type 2 diabetes (T2DM) in Asian Indian middle-aged men with impaired glucose tolerance (IGT) was studied. This was an ancillary analysis of a subsample from a cohort of participants with IGT in a 2 year prospective diabetes prevention program in India. For this analysis, 71 incident T2DM and 76 non-diabetic cases (non-progressors) based on the final glycemic outcome were selected. Baseline serum RBP4 was measured using competitive enzyme immunoassay. Correlations of RBP4 with relevant anthropometric and biochemical variables and also its association with diabetes were assessed using appropriate statistical analyses. Participants who developed T2DM had higher levels of serum RBP4 (21.3 [IQR: 17.7-24.9] µg/mL) compared with non-progressors (17.3 [IQR: 13.1-21.0] µg/mL; P = 0.001). Levels of RBP4 were lower than in Caucasians. Stepwise linear regression analysis showed that body mass index (BMI), systolic blood pressure, triglycerides, and HbA1c had independent associations with RBP4 levels. Multiple logistic regression analyses showed that RBP4 was independently associated with incident diabetes (odds ratio [OR] [95%confidence interval (CI)]: 1.69 [1.18-2.41]; P = 0.004). Adjustment for study group, age, BMI, waist circumference, 2 H plasma glucose, triglycerides, gamma glutamyl transferase, and insulin resistance weakened the significance of its association (OR [95%CI]: 1.65 [1.03-2.66]; P = 0.038).The results of this preliminary analyses showed that baseline serum RBP4 levels were independently associated with incident diabetes in Asian Indian men with IGT. It may be used as an additional predictor of future diabetes.

The oral disposition index is a strong predictor of incident diabetes in Asian Indian prediabetic men.

AIMS: In this analysis, we sought to examine the prospective association of the disposition index (DIo) derived from oral glucose tolerance test with incident diabetes in Asian Indian men with impaired glucose tolerance (IGT). METHODS: These post hoc analyses used data from a 2-year prospective study in primary prevention of diabetes using lifestyle intervention among 517 men with IGT. All the participants received standard lifestyle advice at baseline. The surrogate insulin sensitivity and insulin secretion measures were tested for their hyperbolic relationship. Predictive associations of various surrogate measures with incident diabetes were determined using receiver operating characteristic curves. RESULTS: The combination of total area under the curve of insulin-to-glucose ratio (AUCinsulin/glucose) and Matsuda's insulin sensitivity index was the best equation to depict DIo [ß: -0.954 (95 % CI -1.015 to -0.893)] compared to other measures tested in this cohort. There was an inverse association between change in DIo at the final follow-up and development of incident diabetes. Among the surrogate insulin measures studied, DIo [AUC (0.717 (95 % CI 0.675-0.756))] as a composite measure was superior than other surrogate indices. CONCLUSIONS: Among the surrogate indices studied, DIo was the best measure associated with incident diabetes.

Improvement in diet habits, independent of physical activity helps to reduce incident diabetes among prediabetic Asian Indian men.

AIMS: To assess the beneficial effects of the components of lifestyle intervention in reducing incidence of diabetes in Asian Indian men with impaired glucose tolerance (IGT) in India. METHODS: This analysis was based on a 2 year prospective, randomized controlled primary prevention trial in a cohort of Asian Indian men with IGT (n=537) (Clinical Trial No: NCT00819455). Intervention and control groups were given standard care advice at baseline. Additionally, the intervention group received frequent, mobile phone based text message reminders on healthy lifestyle principles. Dietary intake and physical activity habits were recorded by validated questionnaires. The lifestyle goals were: reductions in consumption of carbohydrates, oil, portion size and body mass index of at least 1 unit (1 kg/m(2)) from baseline and maintenance of good physical activity. The association between diabetes and lifestyle goals achieved was assessed using multiple logistic regression analyses. Changes in insulin sensitivity (Matsuda's insulin sensitivity index) and oral disposition index during the follow-up were assessed. RESULTS: At the end of the study, 123 (23.8%) participants developed diabetes. The mean lifestyle score was higher in the intervention group compared with control (2.59 ± 1.13 vs. 2.28 ± 1.17; P=0.002). Among the 5 lifestyle variables, significant improvements in the 3 dietary goal were seen with intervention. Concomitant improvement in insulin sensitivity and oral disposition index was noted. Higher lifestyle score was associated with lower risk of developing diabetes (odds ratio: 0.54 [95% CI: 0.44-0.70]; P<0.0001). CONCLUSIONS: Beneficial effects of intervention were associated with increased compliance to lifestyle goals. The plausible mechanism is through improvement in insulin sensitivity and beta cell preservation.

Early improvement predicts reduced risk of incident diabetes and improved cardiovascular risk in prediabetic Asian Indian men participating in a 2-year lifestyle intervention program.

OBJECTIVE: Objectives of this ancillary analysis of a prospective, prevention study among Asian Indians with impaired glucose tolerance (IGT) were a) to quantify the reduction in incident diabetes at 24 months in participants who achieved normal glucose tolerance (NGT) at 6 months (NGT-6 m) compared with the other participants, b) the factors influencing the reversal to NGT at the end of the study at 24 months (NGT-24 m), and c) to assess changes in cardiometabolic risk factors in different categories of dysglycemia at 24 months. RESEARCH DESIGN AND METHODS: Data from a 2-year primary prevention trial were used. Effect of reversion to NGT-6 m on incidence of type 2 diabetes mellitus (T2DM) was analyzed using the Cox proportional hazards model. Predictive variables for reversal to NGT were identified using multiple logistic regression analysis. Changes in cardiometabolic risk factors were estimated according to the final glycemic status using fixed-effect, mixed-linear regression modeling. RESULTS: The risk of T2DM in 2 years was lower by 75% in NGT-6 m group (hazard ratio 0.25 [95% CI 0.12-0.52]). Predictive variables for reversal to NGT-24 m were good baseline ß-cell function (odds ratio [OR] 2.79 [95% CI 2.30-3.40]) and its further improvement (OR 5.70 [95% CI 4.58-7.08]), and NGT-6 m (OR 2.10 [95% CI 1.14-3.83]). BMI decreased in those who reverted to NGT. Deterioration to T2DM was associated with an increase in the levels of cardiometabolic risk factors. CONCLUSIONS: Early reversion to NGT by lifestyle intervention in prediabetic men was associated with a significant reduction in subsequent incidence of diabetes. Good baseline ß-cell function and its further improvement and NGT-6 m were associated with reversion to NGT-24 months. Reversion to NGT was associated with modest improvements, whereas conversion to T2DM was associated with significant worsening of the cardiometabolic risk profile.

Effectiveness of mobile phone messaging in prevention of type 2 diabetes by lifestyle modification in men in India: a prospective, parallel-group, randomised controlled trial.

BACKGROUND: Type 2 diabetes can often be prevented by lifestyle modification; however, successful lifestyle intervention programmes are labour intensive. Mobile phone messaging is an inexpensive alternative way to deliver educational and motivational advice about lifestyle modification. We aimed to assess whether mobile phone messaging that encouraged lifestyle change could reduce incident type 2 diabetes in Indian Asian men with impaired glucose tolerance. METHODS: We did a prospective, parallel-group, randomised controlled trial between Aug 10, 2009, and Nov 30, 2012, at ten sites in southeast India. Working Indian men (aged 35-55 years) with impaired glucose tolerance were randomly assigned (1:1) with a computer-generated randomisation sequence to a mobile phone messaging intervention or standard care (control group). Participants in the intervention group received frequent mobile phone messages compared with controls who received standard lifestyle modification advice at baseline only. Field staff and participants were, by necessity, not masked to study group assignment, but allocation was concealed from laboratory personnel as well as principal and co-investigators. The primary outcome was incidence of type 2 diabetes, analysed by intention to treat. This trial is registered with ClinicalTrials.gov, number NCT00819455. RESULTS: We assessed 8741 participants for eligibility. 537 patients were randomly assigned to either the mobile phone messaging intervention (n=271) or standard care (n=266). The cumulative incidence of type 2 diabetes was lower in those who received mobile phone messages than in controls: 50 (18%) participants in the intervention group developed type 2 diabetes compared with 73 (27%) in the control group (hazard ratio 0·64, 95% CI 0·45-0·92; p=0·015). The number needed to treat to prevent one case of type 2 diabetes was 11 (95% CI 6-55). One patient in the control group died suddenly at the end of the first year. We recorded no other serious adverse events. INTERPRETATION: Mobile phone messaging is an effective and acceptable method to deliver advice and support towards lifestyle modification to prevent type 2 diabetes in men at high risk. FUNDING: The UK India Education and Research Initiative, the World Diabetes Foundation.