Table 3.
Weighted diagnostic performance of 1,5-anhydroglucitol, fructosamine, and glycated albumin to identify cases of diabetes comparing four definitions of diabetes
Total study population (N= 1,719)
|
Excluding persons taking diabetes medications (N=1,421)
|
Excluding persons with a history of diabetes* (N = 1376)
|
|
---|---|---|---|
AUC (95% CI)** | AUC (95% CI)** | AUC (95% CI)** | |
Diabetes defined by hemoglobin A1c ≥ 6.5% (N = 252) | |||
1,5-Anhydroglucitol | 0.74 (0.69 to 0.78) | 0.66 (0.58 to 0.74) | 0.65 (0.56 to 0.73) |
Fructosamine | 0.83 (0.79 to 0.87) | 0.71 (0.63 to 0.78) | 0.69 (0.61 to 0.77) |
Glycated Albumin | 0.87 (0.83 to 0.90) | 0.73 (0.64 to 0.81) | 0.71 (0.62 to 0.78) |
Fasting Glucose | 0.83 (0.79 to 0.87) | 0.76 (0.69 to 0.83) | 0.74 (0.66 to 0.81) |
Diabetes defined by fasting glucose ≥ 126 mg/dL (≥6.99 mmol/L) (N = 297) | |||
1,5-Anhydroglucitol | 0.70 (0.65 to 0.75) | 0.60 (0.53 to 0.67) | 0.59 (0.51 to 0.66) |
Fructosamine | 0.83 (0.79 to 0.87) | 0.78 (0.72 to 0.83) | 0.77 (0.71 to 0.83) |
Glycated Albumin | 0.86 (0.82 to 0.89) | 0.80 (0.74 to 0.85) | 0.80 (0.74 to 0.85) |
Hemoglobin A1c | 0.87 (0.85 to 0.90) | 0.82 (0.76 to 0.86) | 0.81 (0.75 to 0.86) |
Diabetes defined by history of diabetes* (N = 343) | |||
1,5-Anhydroglucitol | 0.69 (0.65 to 0.74) | 0.60 (0.54 to 0.67) | - |
Fructosamine | 0.81 (0.77 to 0.85) | 0.74 (0.66 to 0.82) | - |
Glycated Albumin | 0.85 (0.81 to 0.88) | 0.74 (0.66 to 0.83) | - |
Hemoglobin A1c | 0.87 (0.84 to 0.89) | 0.73 (0.64 to 0.82) | - |
Fasting Glucose | 0.80 (0.76 to 0.84) | 0.73 (0.64 to 0.82) | - |
Diabetes defined by hemoglobin A1c ≥ 6.5%, fasting glucose ≥ 126 mg/dL (≥6.99 mmol/L), or history of diabetes* (N = 491) | |||
1,5-Anhydroglucitol | 0.66 (0.63 to 0.70) | 0.59 (0.54 to 0.64) | 0.58 (0.52 to 0.64) |
Fructosamine | 0.80 (0.76 to 0.83) | 0.73 (0.69 to 0.78) | 0.72 (0.67 to 0.78) |
Glycated Albumin | 0.82 (0.79 to 0.85) | 0.75 (0.70 to 0.80) | 0.74 (0.68 to 0.79) |
Self-reported history of diabetes or diabetes medication use.
AUC represents the area under the receiver operating characteristic curve. Confidence intervals were calculated using bias corrected bootstrapping with 1,000 replications.