Research in brief

Abstract

The addition of a genetic score to the Framingham Offspring Study T2D risk model improves the prediction of incident of type 2 diabetes, conclude investigators for the UCLEB Consortium. They used a genetic risk score based on 65 risk alleles for type 2 diabetes to predict disease risk in individuals included in seven UK prospective studies. 804 of 13 294 people developed type 2 diabetes during the 10 year follow-up. The risk score predicted 20% of cases (area under the receiver operator characteristic [ROC] curve 0·60; 95% CI 0·58–0·62) and the phenotypic Framingham model predicted 31% of cases (0·75; 0·73–0·77), whereas the combined risk score predicted 37% of cases (0·76; 0·75–0·78). Researchers for the JUPITER trial have examined whether rosuvastatin reduces the risk of fractures. JUPITER was a clinical trial comparing rosuvastatin with placebo for the primary prevention of cardiovascular disease in people with a serum high sensitivity C-reactive protein concentration ≥2 mg/L; incident fracture was a prespecified secondary endpoint. 221 of 8901 patients receiving rosuvastatin and 210 of 8901 receiving placebo had confirmed fractures (adjusted hazard ratio [HR] 1·06, 95% CI 0·88–1·28; p=0·53) over the follow-up period. Diabetes and prediabetes could be linked to cognitive decline in later life, suggest the investigators of the ARIC study. In their prospective study of 13 351 individuals, 1779 had diabetes at baseline (mean age at baseline was 57 years [SD 5·7]). In people with diabetes at baseline, the change in global cognitive Z score was −0·92 (95% CI −1·00 to −0·85) over the 20-year follow-up, compared with −0·78 (−0·80 to −0·75) in people without diabetes (difference −0·15; 95% CI −0·22 to −0·08). In people without diabetes and with an HbA1c of 5·7–6·4% at baseline, the reduction in global cognitive Z score over 20 years was greater when compared with people who did not have diabetes and who had an HbA1c <5·7% at baseline (p=0·005). Researchers for the Rx Weight Loss Trial have studied the effect of an internet behavioural weight loss programme, including an automated feedback system and weekly multimedia behavioural sessions, in obese individuals. In the trial, adult patients with a BMI of 25–45 kg/m2 were randomly assigned to the 12-week internet-based intervention (n=77), or to control (n=77; comprising an internet-based newsletter). Mean weight loss at 3 months was higher in the intervention group than in the control group (5·5 kg [SD 4·4] vs 1·3 kg [2·1]), with the difference being sustained at 6 months follow-up (p<0·001). New research provides insight into the risk of mortality associated with type 1 diabetes. Using data from the Swedish National Diabetes Register, mortality in people with type 1 diabetes was compared with matched controls by researchers. After roughly 8 years' follow-up, the incidence of mortality was 8% in people with diabetes (2701/33 915) and 3% (4835/169 249) in controls (adjusted hazard ratio 3·52; 95% CI 3·06–4·04). Compared with controls, people with type 1 diabetes and an HbA1c ≤6.9% had more than double the risk of dying from any cause (adjusted HR 2·36; 95% CI 1·97–2·83). Results from a genome-wide association study in a Japanese population provide new information on the genetic basis of Hashimoto's thyroiditis and its distinction from Graves' disease. Researchers included 265 patients with Hashimoto's thyroiditis and 261 patients with Grave's disease in the discovery stage of the study. After logistic regression analysis, 35 SNPs were included in the replication stage, comprising 181 patients with Hashimoto thyroiditis and 286 patients with Graves' disease. Combination of results from the two stages, and subsequent association analysis in controls, showed the rs7537605 locus to be specifically associated with Hashimoto's thyroiditis (odds ratio 1·60, 95%CI 1·30–1·97; p=1·24 × 10−5). According to researchers for the FACTOR-64 trial, CT angiography screening and subsequent therapy recommendations for cardiac disease in patients with diabetes at high cardiac risk—but no symptoms of cardiac disease—does not confer an advantage compared with standard care. In the trial, patients with type 1 or type 2 diabetes were randomly assigned to screening with CT angiography followed by treatment recommendations based on screening results (n=452), or to care based on national guidelines targets (n=448). After 4 years' follow-up, the primary outcome (composite of all-cause mortality, non-fatal myocardial infarction, or hospital admission for unstable angina) did not differ between the intervention and standard care groups (p=0·38).