Home A1c Search

Updated: September 16, 2013

Testing blood Sugar

Content on this page requires a newer version of Adobe Flash Player.

Get Adobe Flash player

Details of Testing

Check that strips are not outdated.

Check that control solution is not outdated. Use control solution with a strip (at least once per new bottle of strips, additional if you get unusual results with a blood sugar test). When you use a control solution, the result should be in the “Control Range”. If not, call the phone number on the back of the meter. Control Range in mg/dl in US, mmol/L in Europe and other areas.


WASH HANDS thoroughly. Why? Modern glucose meters use a very small amount of bloodwasning_hands. Therefore any microscopic, unseen sugar-containing substance on the skin may have a large effect on blood sugar. Testing sugar after handling a banana peel has been shown to increase the blood sugar results significantly. Even hand lotion has been shown to change the blood sugar. If hand washing is not immediately available, a thorough cleansing with an alcohol pad is recommended.

Top of page


CHECK CODE. Why? Some meters require coding. If the code on the strip bottle doemeter_codes not match the code on the meter, blood glucose (sugar) results will be inaccurate.

STRIP STORAGE. Strips should be stored with closed lid in the container they come in.

RECORD RESULTS. Why? Recording results allow you to see patterns. That allows you to adjust your lifestyle to improve control of diabetes. Do NOT change your medication without checking with your physician or diabetes team.

WHEN TO TEST. Your diabetes team will best determine when you should test. Some general rules: Test when you feel that sugar is low, you have a headache, develop shakes or feel different in any way. Re-test immediately if the sugar is unusually low or high. Ideally test before meals and 2 hours after the largest 2 meals. But there are large individual differences in testing need, consult your diabetes team.

ACCURACY. How accurate are glucose meters? “ Accuracy targets established by the International Organization for Standardization (www.iso.org) indicate that 95% of meter measurements should be within ±20% for glucose concentrations ≥ 75 mg/dL and within ± 15 mg/dL for levels <75 mg/dL." The U.S. Food and Drug Administration requires all meters to have an error rate of <20% of the reference value when glucose is >100 mg/dL and within 20 mg/dL when glucose is <100 mg/dL. Most major meter manufacturers meet ISO standards. Data from some of the newer meters (such as those with pharmacy and strip supplier brand names are difficult to find). We prefer meters by major players who provide meters with memory, download capacity with analytic data reports and automatically coded. In general those by Bayer, some by Roche and Abbott meet these criteria. LifeScan meters are very good but still require coding.

Sampling Blood Glucose (sugar).
Testing blood glucose is essential if one is to maintain a close watch on the progress of diabetes. This is especially important for those taking insulin, but also important for those with type 2 diabetes taking oral agents (pills). Remember, type 2 diabetes is progressive, no matter what is used to treat it, it gets worse. We often say to our patients with type 2, “whatever was working to control blood sugar 2 years ago, probably won’t be working now.”  The frequency of testing is subject to controversy. Each person with diabetes should discuss this with the Health Care Provider or Diabetes Team.  Our rules of thumb for people with stable diabetes are these:

  1. If using an insulin pump, test 4 – 6 times daily.
  2. Those taking insulin injections, test 1 more time per day than insulin injections taken. For example:
    1. If taking insulin once daily e.g. Lantus®, or Levemir®, at bedtime.
      1. Test blood glucose twice daily, in the AM and 2 hrs after the largest meal. Occasionally test at about 2 AM to check on overnight glucose. The morning test is used to judge the effect of the bedtime insulin and the post meal one to judge the effect of food on the blood glucose. Generally our target in the morning is 80 -120 mg/dl and less than 140 mg/dl, 2 hrs after the meal.
    2. If taking premixed insulin (e.g. 75/25 or 70/30).
      1. Test before each injection and 2 hours after the largest meal.  Occasionally test at about 2 AM to check on overnight glucose.
    3. If taking rapid insulin   (e.g. Humalog®, Novolog®, or Apidra®,) before meals and a basal insulin (e.g. Lantus®, or Levemir®,) at bedtime
      1. Test before each meal, 2 hours after the largest meal and at bedtime. Occasionally test at about 2 AM to check on overnight glucose.
  3. For those with type 2 diabetes taking only oral agents (pills), it depends:
    1. If A1c is at target (generally less than 6.5%, but this must be individualized), there are no episodes of hypoglycemia (low blood sugar) and everything else is stable, perhaps 2 to 3 times a week is adequate.
    2. If A1c is not at target, twice daily testing at least (morning and 2 hours after the largest meal).

Situations that might create a need to test more often.

  1. Having low blood sugar (hypoglycemia).
  2. Not in control.
  3. Wide swings in blood sugar (glucose).
  4. Frequent stressful situations.
  5. Other illnesses (e.g. heart attack, stroke, kidney problems).
  6. Any surgery.
  7. Any infection or inflammation.
  8. Dental problems (especially gingivitis).
  9. Any change in diabetes medications.
  10. The addition of other medications that may change blood glucose (sugar). Common ones include steroids (either injected or taken by mouth), blood pressure medications, fluid pills, thyroid medicines and medications used to treat HIV.

Top of page


CGM (Continuous Glucose Monitoring)
At the current state of research CGM measures glucose in the fluid beneath the skin (subcutaneous fluid). It has several advantages over glucose meter testing. One might consider the finger stick test a snapshot of the state of glucose here and now. The meter test however does not provide a complete pattern of the glucose changes even is the person is testing frequently.

In this example the person is testing four times daily. It appears that morning glucose is elevated but there is good control for the rest of the day.

Meter Glucose

When one looks at the wider picture with CGM it is obvious that glucose is sometimes very elevated in the morning and again in the evening, probably related to food intake. Continuous Glucose Monitoring can thus be used to adjust therapy to avoid unrecognized high and low glucose. Using CGM dose not take away the need to test with finger stick. All devices now available require finger stick blood sugar to calibrate the CGM device. Click here to see CGM devices.


Click here to see the device used to create this graph.

Many studies show that diabetes complications (kidney failure, amputation, heart attack, and stroke) can be prevented or delayed by good control of blood glucose (sugar). One also needs to control other risk factors such as blood pressure and cholesterol. It doesn’t matter HOW it is controlled as long as the control method itself is not harmful.  In those of us requiring insulin, it is nearly impossible to gain control of blood glucose (sugar) without data collection.
In the “normal” person without diabetes, blood glucose is kept within a narrow range. This is done by the secretion of insulin in two fashions:

  1. Basal Insulin. A steady flow of insulin, required primarily because of liver production of sugar. This basal insulin may vary due to exercise, illness, stress, etc.
  2. Bolus Insulin. A temporary squirt of insulin in response to something the raises the blood glucose (sugar). The common cause for a need of bolus insulin is food.

If the pancreas is not able to provide for these types of insulin, something must be done to help it along. In the person with Type 1 diabetes, both basal and bolus insulin will be required. In the person with type 2 diabetes, medication other than insulin may improve insulin secretion, help the response to insulin, reduce liver sugar production, slow stomach emptying, reduce carbohydrate absorption, etc.  All of these require some internal insulin production to be effective. Eventually most people with type 2 diabetes will require insulin in some form.
In the insulin-requiring person with diabetes (type 1 or type 2), the collection of blood sugar information is crucial for insulin dose adjustment to gain control. Somehow basal insulin needs (e.g. Lantus®, Levemir®) must be established (we generally estimate this based on body weight) and bolus insulin must be matched to blood glucose (sugar) and food intake. Below is a example of a form we use to collect information.
In this example several things become obvious:

  1. The high glucose at 3:00 AM suggests that the Lantus dose is too low (would like to see this pattern for several days before making a change).
  2. The amount of Humalog give for the 3:00 AM high blood glucose was too much – it caused the 7:00 AM blood sugar to be too low.
  3. The person felt bad, ate a large breakfast and did not take any Humalog for the food. The result is high sugar at noon. In this case we would probably recommend taking half the amount of Humalog normally required for the food.
  4. The amount of Humalog taken for the high sugar and food at lunch and dinner was on target. This conclusion is reached because the 6:00 PM and 11:00 PM sugars are at the desired range.
  5. Blood sugar 2 hours after meal is at target.

This example only shows one day. We like to see several days before making changes,

This Excel file can be downloaded by clicking here , or in pdf format for printing.

Top of page



Designed and Copyright by Charles H. Raine, III, M.D.

Mission StatementPrivacy PoliiesPrlduct Policy