The potential for a two-stage diabetes risk algorithm combining non-laboratory-based scores with subsequent routine non-fasting blood tests: results from prospective studies in older men and women.

摘要: P>AimTo develop strategies based on simple clinical assessment and blood markers to identify older individuals at high risk for Type 2 diabetes.MethodsA prospective study of non-diabetic men (n = 3523) women 3404) aged 60-79 years followed 7 years, during which there were 297 incident cases diabetes. Logistic regression was used scores predict cases, starting with predictors adding that predicted the incidence Receiving operating characteristic analyses assess improvement in prediction.ResultsThe area under curve a score, included age, sex, family history diabetes, smoking status, BMI, waist circumference, hypertension recall doctor diagnosis coronary heart disease 0.765 (0.740, 0.791); sensitivity specificity top quintile score 50.3 81.4%, respectively. Addition fasting HDL cholesterol, triglyceride glucose improved prediction [area 0.817 (0.793, 0.840), P < 0.0001; 63.8%; 82.0%]. An alternative model not dependent yielded similar results. Further addition C-reactive protein made no improvement. Blood measurements small differences reclassification those lowest three quintiles non-laboratory score.ConclusionIn large population settings, assessments could be first instance adults who would benefit from further testing routine (non-fasting) most likely elevated diabetes risk.