By PulsHealth. Updated
Research reviewed .
What does blood glucose measure?
Blood glucose is the concentration of glucose, or blood sugar. HealthKit uses this discrete quantity type for readings written by compatible devices, apps or manual entry. A source may represent a fingerstick meter, laboratory result or continuous glucose monitor (CGM); these measurement methods should not be treated as interchangeable.
Apple Watch does not measure glucose directly. Data appearing in Apple Health can have been supplied by another device or app. Check the original source before interpreting a graph.
Which units should I use?
Apple’s HealthKit documentation supports regional display preferences and warns against saving meter control-solution tests as glucose observations. PulsHealth’s canonical export unit is mg/dL.
| Conversion example | Equivalent |
|---|---|
| 90 mg/dL | Approximately 5.0 mmol/L |
| 126 mg/dL | Approximately 7.0 mmol/L |
| 180 mg/dL | Approximately 10.0 mmol/L |
Divide mg/dL by 18.0182 to obtain mmol/L. Convert the quantity, not just its label; values already exported in mg/dL should not be converted again.
What did the database research find?
The October 2025–September 2026 snapshot contained no blood-glucose observations, including the July–September focus window; the inventory also found no glucose aggregates or workout-series points. This means ranges, extreme readings, source-specific cadence and workout/sleep changes could not be measured here. It does not mean glucose was zero or healthy, or that any particular device or permission was absent. No measured-data figure is presented.
Why does recording frequency matter?
A CGM can estimate glucose every few minutes; a fingerstick or manually entered laboratory result is a point observation. NIDDK explains that CGM sensors measure fluid between cells. Sensor and blood values can differ, especially as glucose changes; follow the particular device’s instructions when readings conflict with symptoms or other measurements.
A raw average weights every stored row equally. A dense sensor period can dominate a sparse meter history. Before reporting time in range, verify source modality, contiguous coverage, missing periods and the target specified for that person. Counts of samples inside a band are not automatically percentages of elapsed monitored time.
What ranges are meaningful?
Diagnostic cutoffs require a defined test. NIDDK’s diagnostic guidance distinguishes fasting plasma glucose, a standardized oral glucose tolerance test and other laboratory tests.
| Laboratory context | Reference interpretation |
|---|---|
| Fasting plasma glucose below 100 mg/dL | Normal diagnostic category |
| Fasting plasma glucose 100–125 mg/dL | Prediabetes category |
| Fasting plasma glucose 126 mg/dL or higher | Diabetes threshold; clinical confirmation as appropriate |
| Two-hour 75-g oral glucose tolerance test 140–199 mg/dL | Prediabetes category |
| Two-hour 75-g oral glucose tolerance test 200 mg/dL or higher | Diabetes threshold |
Those oral-glucose-test cutoffs do not classify an arbitrary after-meal or CGM reading. Diabetes-management targets are also a different question and should follow the care plan. Timing, pregnancy, medications and measurement method matter.
How do meals, workouts and sleep fit in?
HealthKit can include meal-relative metadata, but absence of that key is unknown context. A morning timestamp does not establish fasting, and a “before meal” tag does not establish an eight-hour fast. Nutrition entries need actual timing and quantity to support a meal comparison.
Workout and asleep intervals can provide temporal context when glucose records exist. Keep source-specific coverage and sensor timing in view, avoid joining unrelated sources by proximity, and compare like conditions. Dietary Carbohydrates and Insulin Delivery can support richer descriptions when present; none of these relationships was measurable in this snapshot.
What makes an unusual reading worth checking?
Retain extreme readings for review instead of automatically deleting them as statistical outliers. Check original units, sensor alerts, control-solution tests, duplicate imports, meal context and symptoms. Do not change treatment using an exploratory export chart; use the prescribed monitoring and safety plan.
Understanding your records
Recording frequency, source and coverage affect what you can conclude from this data. Read the recording guide.
Reference ranges
These source-specific ranges provide context; they are separate from the personal values in any data examples and do not diagnose a condition.
Fasting plasma glucose laboratory test
- normal
- <100 mg/dL
- prediabetes
- 100–125 mg/dL
- diabetes threshold
- >=126 mg/dL
Not a diagnosis from an arbitrary sensor or after-meal reading; confirm clinically as appropriate.
Standardized 75-g oral glucose tolerance test, two-hour plasma result
- normal
- <140 mg/dL
- prediabetes
- 140–199 mg/dL
- diabetes threshold
- >=200 mg/dL
These cutoffs do not apply to arbitrary post-meal measurements.
| Population | normal | prediabetes | diabetes threshold |
|---|---|---|---|
| Fasting plasma glucose laboratory test Not a diagnosis from an arbitrary sensor or after-meal reading; confirm clinically as appropriate. | <100 mg/dL | 100–125 mg/dL | >=126 mg/dL |
| Standardized 75-g oral glucose tolerance test, two-hour plasma result These cutoffs do not apply to arbitrary post-meal measurements. | <140 mg/dL | 140–199 mg/dL | >=200 mg/dL |