A1C reflects average blood glucose over approximately three months; a fasting glucose test measures glucose at a particular time; an oral glucose tolerance test measures the response to a defined glucose drink. They answer related but different questions. A result does not by itself determine whether diabetes is type 1 or type 2. 12
Repeated vomiting, abdominal pain, fruity-smelling breath, difficulty breathing or marked illness can be signs of diabetic ketoacidosis and need urgent care. Do not wait for a routine A1C appointment when an acute emergency is possible. 3
Diagnostic ranges for nonpregnant people
The following laboratory thresholds summarize NIDDK guidance. They are not individual glucose-management targets and are not the pregnancy-testing criteria. 2
| Test | Below the prediabetes range | Prediabetes range | Diabetes range |
|---|---|---|---|
| A1C | Below 5.7% | 5.7–6.4% | 6.5% or above |
| Fasting plasma glucose | 99 mg/dL or below | 100–125 mg/dL | 126 mg/dL or above |
| Two-hour OGTT plasma glucose | 139 mg/dL or below | 140–199 mg/dL | 200 mg/dL or above |
In someone without clear symptoms, a second test is usually used to confirm diabetes. A random plasma glucose of 200 mg/dL or above has a diagnostic role when symptoms are present; it should not be detached from that context. 2
Do not compare a value in mmol/L directly with this mg/dL table. Record the units from the actual laboratory report. A consumer device reading is not automatically the same as the diagnostic laboratory test described here.
What needs fasting?
NIDDK describes fasting for at least eight hours before the fasting plasma glucose test. The OGTT has its own preparation and timed sampling requirements. A1C does not require fasting. Follow the testing service's instructions about preparation and medicines instead of modifying a treatment plan yourself. 21
Record whether the test was actually fasting and whether the sample was taken at the specified time. “A glucose result after a meal” is not automatically the same thing as the standardized two-hour OGTT.
This is especially important when collecting results from different facilities. Preserve the exact test names and dates rather than relabeling every glucose measurement as “blood sugar.”
Why results can disagree
A1C and a single glucose measurement capture different time windows. A1C can also be affected by conditions that alter red blood cells or interfere with the test, including certain hemoglobin variants, recent blood loss or transfusion and some anemias. A clinician may use a different or repeated test when the result does not fit the situation. 1
Do not respond to disagreement by averaging the values. An A1C percentage and a plasma glucose concentration are not measurements that can be added and divided to create a diagnosis.
Ask, “Which result needs confirmation, could a known condition affect this test, and what is the next step?” This preserves the uncertainty without deciding that one laboratory is necessarily wrong.
A worked result-record example
Consider a fictional record with an A1C report, a fasting result from another date and a home reading after dinner. A useful table keeps all three, but labels them correctly:
Date → exact test → result and units → fasting/timing details → clinician interpretation → follow-up required.
Do not omit an inconvenient result or turn a home reading into an OGTT. Conversely, do not diagnose the diabetes type from a high value. NIDDK explains that additional testing, including selected autoantibody testing, can be used to clarify type. 2
The results tracker provides an empty record and a place for unresolved questions. It has no diagnostic scoring function.
Diagnosis is not the same as the treatment target
A value used to establish a diagnosis is not automatically the goal selected for every person already receiving care. Targets and the treatment plan are individualized. Discuss an unexpected result or symptoms with the treating team rather than increasing or stopping medicines based on this table. 1
Also distinguish evidence about glucose control from evidence about replacing insulin-producing cells. The site's islet-cell therapy comparison concerns specific research and approved-treatment contexts; an elevated test result is not eligibility for an experimental intervention.
The most useful next step is to leave the testing process with an interpreted result, a clear confirmation plan where needed, and an identified person responsible for follow-up. Collecting numbers without that handoff leaves the central clinical question unresolved.