What is the best screening test for diabetes?

Best Screening Test for Diabetes

The fasting plasma glucose (FPG) is the best screening test for diabetes because it is easier and faster to perform, more convenient and acceptable to patients, less expensive, and more reproducible than other screening tests. 1

Why FPG is Preferred Over Other Tests

The FPG test has several practical advantages that make it superior for screening purposes:

• Greater convenience and lower cost compared to the 2-hour oral glucose tolerance test (OGTT), which requires multiple blood draws and prolonged patient time 1
• Better reproducibility with less intraindividual variation than the 2-hour plasma glucose test 1
• Similar predictive value for development of microvascular complications of diabetes compared to OGTT 1
• Standardized methodology using enzymatic assay techniques (hexokinase or glucose oxidase) with good standardization across laboratories 2

Diagnostic Threshold

• FPG ≥126 mg/dL (7.0 mmol/L) indicates diabetes and should be confirmed with repeat testing on a different day 1, 2
• If FPG is <126 mg/dL but there is high clinical suspicion for diabetes, an OGTT should be performed 1

Alternative Tests: A1C and OGTT

While the 2014 American Diabetes Association guidelines expanded acceptable screening tests to include A1C and OGTT alongside FPG 1, important caveats exist:

A1C Testing (≥6.5% diagnostic threshold)

• Advantages: Greater preanalytic stability, no fasting required, less day-to-day variation 1
• Significant limitations:
  • Misses a substantial number of people with type 2 diabetes, including some with fasting hyperglycemia 3
  • Misses most people with impaired glucose tolerance 3
  • Susceptible to systematic error from factors affecting red blood cell turnover (anemia, hemoglobinopathies) that may not be clinically evident 3
  • Less sensitive for detecting mild glucose intolerance 4

OGTT (2-hour value ≥200 mg/dL diagnostic)

• May identify more individuals as diabetic compared to FPG 1
• Major disadvantages: More expensive, time-consuming, less convenient, and has the highest intraindividual variability of all tests 1
• Should be reserved for cases where FPG is normal but clinical suspicion remains high 1

Who Should Be Screened

Screen adults 40-70 years of age who are overweight or obese (BMI ≥25 kg/m²), repeating every 3 years if normal 1, 5

Additional screening criteria include 1:

• Age ≥45 years regardless of weight
• Physical inactivity
• First-degree relative with diabetes
• High-risk ethnicity (African American, Latino, Native American, Asian American, Pacific Islander)
• History of gestational diabetes or delivering a baby >9 lb
• Hypertension (≥140/90 mmHg or on therapy)
• HDL cholesterol <35 mg/dL or triglycerides >250 mg/dL
• Previous A1C ≥5.7%, impaired glucose tolerance, or impaired fasting glucose
• Polycystic ovarian syndrome
• History of cardiovascular disease

Critical Technical Considerations

To ensure accurate FPG results 2:

• Use proper collection tubes containing glycolytic inhibitors (citrate buffer) to prevent falsely low values
• Process samples promptly: If appropriate tubes unavailable, place sample immediately in ice-water slurry and centrifuge within 15-30 minutes
• Measure in accredited laboratory using enzymatic methods with analytical imprecision ≤2.4%, bias ≤2.1%, and total error ≤6.1%
• Fasting defined as no caloric intake for at least 8 hours 1

Confirmation Testing

• Always confirm abnormal screening results with repeat testing on a different day, especially for borderline values 1
• If two different tests (e.g., A1C and FPG) are both above diagnostic thresholds, diagnosis is confirmed without repeat testing 1
• Exception: A single random plasma glucose ≥200 mg/dL with classic symptoms (polydipsia, polyuria) likely indicates diabetes without need for confirmation 5

Common Pitfalls to Avoid

• Do not use A1C as the sole diagnostic test due to risk of systematic bias and missed diagnoses 3
• Do not perform community screening outside healthcare settings—this has not been shown beneficial and may cause harm 1
• Do not use random capillary blood glucose meters for screening—these lack standardization and precision needed for diagnosis 1
• Do not confuse normal reference ranges (74-110 mg/dL) with diagnostic criteria for diabetes 2