Measuring Glycemic Control
Measuring Glycemic Control
Elaena Quattrocchi, Pharm.D., FASHP
Associate Professor, Division of Pharmacy Practice, Arnold & Marie Schwartz
College of Pharmacy & Health Sciences, Brooklyn, NY
Diabetes affects 6%–7% of the population, or approximately 16 million Americans.1
The prevalence in people over 65 years of age is estimated to be 20%; half of these
patients are undiagnosed.1 Among those who have been diagnosed, too few are
receiving education and adequate care that can prevent the deadly complications of
diabetes. Secondary complications from this chronic, debilitating disease include stroke;
coronary artery disease; end-stage renal disease; peripheral vascular disease, which could
lead to amputation, neuropathy, gastroparesis and other GI complications; and retinopathy,
which could lead to blindness.
Fifteen percent of total medical costs in this country and 25% of the total Medicare
budget is spent on diabetic patients, mostly for the complications of the disease.1,2
Diabetes annually costs the U.S. economy more than $100 billion.3
While there is neither a clear understanding of the cause of diabetes nor a cure, recent
research has demonstrated the need for continuous, tight control of blood glucose to avoid
the long-term complications of the disease. The Diabetes Control and Complications Trial
(DCCT) in type 1 diabetes and the Kumamoto trial in type 2 diabetes demonstrated that
intensive control of blood glucose concentrations to near euglycemia (near normal range)
will ameliorate and possibly prevent the onset of diabetic nephropathy, retinopathy and
neuropathy.1 This would save the Medicare budget more than $17 billion
annually.3 The American Diabetes Association’s position statement on the
standards of medical care for patients with diabetes mellitus mandates treatment aimed at
lowering blood glucose levels to or near normal in all patients.4 Tight blood
glucose control should be a primary treatment goal for the majority of patients with
diabetes. TABLE 1 shows the ADA recommended target goals of glycemic control for people
|What type of glucose testing is best for an individual who has diabetes? Educate your
patients about self-monitoring |
and proper use of home testing products.
Target Glucose Levels for Diabetes Patients
|Preprandial glucose (mg/dL)
||<80 or >140
|Bedtime glucose (mg/dL)
||<100 or >160
|Hemoglobin A1c (%)**
|Note: These values are for
*Action suggested depends on individual patient circumstances. Such actions may include
enhanced diabetes self-management education, comanagement with a diabetes team, referral
to an endocrinologist, change in pharmacological therapy, initiation of or increased SMBG,
or more frequent contact with the patient.
**HbA1c is referenced to a nondiabetic range of 4.0%–6.0%
(mean 5.0%, SD 0.5%).
Source: reference 4
Patients with diabetes visit their community pharmacy an average of three to eight times
more per year than other patients.5 It is likely that the diabetic patient will
see the pharmacist more often than any other health professional.6 Pharmacists
should check if their patients with diabetes are consistently checking their glucose
Although the key to successful diabetes control is self-monitoring, many patients do not
test their glucose levels. The pharmacist must develop teaching programs that will reach
this population and stress the importance of self-monitoring to help avoid acute and
chronic complications of diabetes. There are three categories of products available for
self-monitoring: urine glucose and ketone products, blood glucose test kits, and HbA1c
Urine Glucose and Ketone Tests
Although urine glucose testing is a less desirable method than blood glucose monitoring,
many patients continue to use this method. Therefore, it is important for pharmacists to
be educated regarding these tests so they can properly counsel patients.
The most serious disadvantage of urine glucose testing is that it correlates poorly with
blood glucose concentration. Urine glucose concentration is influenced by a patient’s
renal threshold for glucose. Glucose is filtered at the glomerulus and reabsorbed back
into the circulation in the proximal tubule.7 When the filtered glucose load
exceeds the tubular resorptive capacity, called the renal threshold, the nonabsorbed
glucose appears in the urine.7 The normal renal threshold is approximately 180
mg/dL.7 However, the renal threshold varies among individuals and is affected
by age, presence of renal disease, and pregnancy.7 Some older people have a
high renal threshold and therefore can have high blood sugar levels without much glucose
appearing in the urine. Heart failure, hypertension, and renal disease can also cause high
renal thresholds. On the other hand, a child or a pregnant woman may have a low renal
threshold, resulting in evidence of glucose in the urine when the blood sugar is normal or
Factors not directly influencing renal threshold may also influence urine glucose values.
Results of urine glucose tests can be adversely affected by the concentration of the
urine. If the urine is concentrated, as in dehydrated patients, test results will be
inflated. If the urine is diluted, as in overhydrated patients, test results will be
deflated. Incomplete bladder emptying causes a mixture of recent and residual urine, which
fails to reflect recent urine glucose values.
Another drawback to urine glucose testing is that these tests cannot identify a
hypoglycemic state. Hypoglycemic patients do not “spill glucose into the urine”
(i.e., exceed renal threshold). The last problem with urine testing is that the patient
must read a color chart, making it difficult for patients with color-blindness to obtain
Even though there are limitations to urine glucose testing when compared with blood
glucose testing, urine testing is inexpensive, painless, useful in patients who will not
monitor blood glucose levels, and a valuable indicator of glucose over a period of time.
Blood glucose testing is a single-point determination, and it is difficult to determine if
the blood glucose is increased or decreased over a span of time. Urine glucose testing can
be used to complement blood glucose monitoring. However, in order to obtain accurate
results, patients should have a normal renal threshold and complete bladder emptying.
Patients must also follow the correct procedures when performing the tests.
There are two methods of testing for glucose in the urine: copper reduction tests and
glucose oxidase tests. The copper reduction tests utilize cupric sulfate, which becomes
reduced to cuprous oxide in the presence of glucose and other sugars, and results in a
color change from green to orange. The glucose oxidase tests utilize the enzyme glucose
oxidase or peroxidase, which reacts specifically with glucose to produce a color change.
Copper Reduction Testing: Clinitest is the only product on the market
using the copper reduction method. This test can be performed using either a 5-drop method
or a 2-drop method. Separate instructions for each method are provided in the package. The
key difference between the two methods is the amount of urinary glucose detected. The
5-drop method is more accurate for lower concentrations of glucose (up to 2%).7,8 The
2-drop method is more accurate for higher glucose concentrations and can read up to 5%
Because of problems associated with this test, it is rarely used. The method is
quantitative, but it is not specific for glucose. It may produce false-positive results
when other sugars are present, such as lactose, which is found in nursing mothers’
urine. Certain drugs that reduce cupric sulfate to cuprous oxide may also cause
false-positive results. TABLE 2 lists substances that interfere with urine tests.
Substances that Interfere with
Urine Tests for Glucose
|+= False positive
*Except with Chemstrip UG or K
-= False negative Source: reference 14
Several other disadvantages are associated with the copper reduction method. The copper
reduction reaction is exothermic and could cause severe burns, especially in a patient who
has peripheral neuropathy and may not be aware of the temperature. Clinitest tablets and
solution are caustic and poisonous, so caution must be taken, especially with children,
when using these products. This test requires equipment and the dilution of urine, which
can be cumbersome and inconvenient to the patient. The patient should observe the reaction
as it is occurring and read the color exactly 15 seconds after the tablet was placed in
the tube.8 The proper color chart for the method being used (5- or 2-drop) must
be consulted when interpreting the results.8 Positive reactions can sometimes
be read as low if not watched closely, due to the “pass through phenomenon.”
During the “pass through phenomenon” a bright orange color will be seen during
the first few seconds, and this color will fade to a greenish brown when the reaction
ceases and the color is read.8 This occurs with the 5-drop test when there is a
high percentage of glucose present in the urine being tested.8 If this occurs,
the patient should retest his/her urine using the 2-drop test.
Glucose Oxidase Testing: There are several different glucose oxidase
urine tests on the market. All these tests are qualitative and specific for glucose. When
recommending a test to a patient or healthcare professional one should consider shelf
life, the color chart, how quantitative the product is, and ease of use.8 TABLE
3 compares the various urine glucose tests on the market. Among them, Chemstrip UG is the
most widely used. Chemstrip UG strips have two reagent pads, which improves reading
accuracy, and the color of the strip is stable for one minute after the 2-minute read
time.8 The 2-minute read time may be a disadvantage when compared to a
30-second or less read time with other strips. Chemstrip UG, however, combines the
advantage of high sensitivity, a range from 1/10% up to 5%, and discernible color changes
as urinary glucose concentrations gradually increase from the lowest to the highest
Urine Glucose Tests
| Glucose oxidase
||1/10, 1/4, 1/2,
1, 2, 3, 5
||Date on bottle
1/4, 1/2, 3/4, 1, greater than or = 2
Less than or = 1/2, 1, 2, 3, 5
||Date on bottle
||6 months after
||1/10, 1/4, 1/2,
1, greater than or = 2
||6 months after
With all urine glucose tests, patients should be counseled to use only the color chart on
the vial. A color chart from another lot can indicate inaccurate results. Patients should
also be counseled on expiration dates of products. Chemstrip UG and Clinitest are stable
until the expiration date on the bottle.8 Clinistix and Diastix are stable only
for 6 months after opening the vial.8 The primary disadvantage of glucose
oxidase urine tests is that some substances interfere with results.
Ketone Testing: When patients with diabetes are ill, injured or not able
to eat, glucose and ketones must be monitored. This is particularly important for type 1
patients, who are especially susceptible to ketoacidosis. Urine ketone testing is not
needed in well-controlled type 1 patients and is used only for those type 2 patients who
are under severe stress (e.g., sepsis, myocardial infarction). Pregnant women also need to
test for ketones in their urine because ketones can adversely affect the fetus.
Patient Counseling for Urine Glucose Tests
|• Test freshly voided urine. Urine that has been allowed to stand will give an inaccurate result due to bacterial consumption of glucose.|
• Do not touch the reagent pad or tape with your fingertips. The oil on your skin can give an inaccurate result.
• Time test accurately and read the test result right after. Some tests fade immediately.
Source: reference 8
Ketones are metabolic acids. They are comprised of beta-hydroxybutyrate, acetone, and
acetoacetate. The basis for ketone testing is that sodium nitroprusside alkali or the
sodium nitroferricyanide in the tests will react with certain ketones and cause a color
change. Ketone tests do not react with beta-hydroxybutyrate, but are specific for
acetoacetate and/or acetone. TABLE 4 lists urine ketone test products and which ketones
they detect. It is important to realize that beta-hydroxybutyrate may be present in the
urine of diabetic patients who are on low-calorie or high-fat diets or in times of
fasting, stress or exercise.8 Patients may be given a false sense of security
by a negative test result since urine ketone tests will not detect beta-hydroxybutyrate.8
Chemstrip UK is the easiest product to use, has the longest shelf-life, and will detect
both acetone and acetoacetate in the urine. Acetest will also detect both, but it is not
easy to use; it is messy and time-consuming. It can be used with urine or blood. Ketostix
will only detect acetoacetate and give a false-negative result if acetone is present. It
is only good for 6 months after opening the vial. There are also products that monitor
glucose and ketones (Chemstrip UK and Keto-Diastix).
Urine Ketone Tests
|| Ketones Detected
||Acetoacetate & acetone
|Date on bottle
||Can use urine or serum
||Acetoacetate & acetone
||Date on bottle
||Easy to use
||6 months after opening
if acetone is present
|Urine Glucose and
|Chemstrip UGK, Keto-Diastix
Source: reference 8
Self-Monitoring of Blood Glucose
Home blood glucose monitoring revolutionized the management of diabetes mellitus by
providing a simple and portable method for periodic and repeated measurement of blood
glucose in the ambulatory care setting.9 It is generally accepted that
monitoring blood glucose levels is the best way to manage diabetes. The ADA recommends
self-monitoring for the following groups: pregnant patients with diabetes; unstable
diabetic patients; patients prone to severe ketosis or hypoglycemia; patients who do not
experience warning symptoms when hypoglycemic; patients on intensive treatment regimens,
such as those using insulin infusion devices or multiple daily insulin injection; and
patients with an abnormal renal threshold for glucose.4
Self-monitoring provides information about current blood glucose levels so that the
patient can make necessary adjustments in medication, diet and exercise based upon his/her
knowledge and direction from a healthcare provider. A log of blood glucose values helps
the provider make decisions about medication dosage, diet and exercise. The log is more
helpful than a routine fasting blood sugar, which is just one point in time.
Self-monitoring is a way of establishing a patient’s individual response to insulin
(peak action and duration of activity) and documenting the influence of diet and exercise
on blood glucose levels. It gives the patient more confidence and knowledge in the
management of his/her disease, and allows the patient to achieve tighter blood glucose
control to prevent diabetic complications.
Patients who use oral diabetes medication or follow a meal plan exchange program should
monitor blood glucose levels twice daily (alternating times during the day), in cases of
poor glycemic control (4 times a day), during illness (every 2–4 hours, depending on
severity), and in cases of hypoglycemia.6 Patients requiring insulin need to
self-monitor more frequently: before meals and bedtime, before and after exercise, in
cases of hypoglycemia, in cases of hyperglycemia, in cases of nocturnal hypoglycemia
(check 3 am blood glucose level), and during illness (every 2–4 hours, depending on
Over the years technology has expanded to provide a variety of meters with various
features. TABLE 5 lists a representation of currently available meters. These products are
smaller, faster and easier to use than the devices first introduced in the early 1980s.
Two types of blood glucose meters are used for SMBG: color reflectance meters and those
that use sensor technology.10-12 With reflectance meters, the glucose in a drop
of blood reacts with glucose oxidase or hexokinase on the test strip and changes the color
of the strip. The color of the strip is then accurately measured by the meter, which gives
a numerical readout.10,11 Sensor-type meters measure the electronic charge
generated by the reaction of the glucose and the glucose oxidase or hexokinase.10-12
It should be kept in mind that plasma and serum have a slightly different glucose
concentration from that of whole blood.13 Many self-monitoring blood glucose
systems use whole blood; laboratories use plasma or serum.13 Self blood glucose
monitoring values are generally 10%–15% (not 10–15 points) less than concurrent
plasma glucose values reported by the lab.9,13 Some meters such as Glucometer
Elite and SureStep are calibrated to give values equivalent to plasma concentrations even
though the sample measured is whole blood.7,12 Different meters from the same
company are calibrated to give either whole blood or plasma equivalent values (e.g.,
LifeScan).7 This characteristic can be identified by contacting the company.
|Table 5. Selected Blood Glucose Monitors
Glucometer Dex Diabetes Care System
Glucometer Elite XL
Boehringer Mannheim 1-800-858-8072
Accu-Chek Instant Plus
Supreme II Blood Glucose
Assure Blood Glucose Meter
Home Diagnostics Inc. 1-800-342-7226
LXN Corporation 1-888-596-8378
Duet Glucose Control System
One Touch Basic
One Touch Basic Enhanced
One Touch Profile
ExacTech, ExacTech RSG, Medisense 2,
Medisense 2 Card, Medisense 2 Pen,
Precision QID, Precision Pen
All meters have a hematocrit range, within which patients will get accurate results from
the meter.11 Hematocrit refers to the percentage of the patient’s total
blood volume that consists of red blood cells.11 If a patient’s hematocrit
is not within a range that the meter is designed to handle, the meter can give an
inaccurate result.11 Other patient variables that could affect blood glucose
readings include uric acid >13 mg/dL, bilirubin >10 mg/dL, triglyceride >3,000
mg/dL and total cholesterol >500 mg/dL.
Meter performance involves both accuracy (extent to which the glucose value agrees with
the laboratory reference) and precision (reliability of the test result or consistency of
measurement).12 Important health decisions are based on the results of blood
glucose testing, so accuracy and precision are essential. Most monitors have the
capability of giving results within 15% of a matching laboratory test.14
Although the American Diabetes Association (ADA) now recommends strict criteria for meter
accuracy, some individuals consider SMBG results acceptable and reasonable if they are
within 15% of the reference method.12 The consistency of blood glucose readings
is as meaningful as absolute accuracy because changes in therapy are often based on
emerging trends or patterns rather than on individual values.12
Unfortunately, several factors can affect the accuracy of meter readings, including
equipment malfunction and human error. The reagents (strips or sensors) are one source of
potential error.10,11 Pharmacists should make sure patients are using the
correct strips with their meters. Patients should also be instructed to check the
expiration date before use; keep the strip in the vial or foil wrapping until they are
ready to use it; and be careful not to touch the test area of the reagent strip. SureStep
and Precision QID are strips that are touchable.10-12 Test strips are sensitive
to temperature, light and moisture, and should not be exposed to light for long periods of
Quality control checks with control solution is the best way to verify that the meter and
strips are working together properly.6,10,11 Control solutions are solutions of
known glucose concentrations.12 Every meter has at least one control solution,
and some have three levels of control solutions (low, normal and high) so patients can
test the meter at the extremes.10,11 Control solution is used with the meter
the same way a patient tests a drop of blood.11 Using control solution on the
first strip out of a new vial or package is a great way to assure the accuracy of results.11
Calibrating the meter is another way to ensure accurate results. Some of the newer meters
automatically calibrate the reagent with the meter, unlike other meters that require
setting a code or inserting a code chip or strip to calibrate the meter. Patients should
also properly clean and maintain their meters in order to prevent inaccurate results.
Patients should be cautioned that extreme temperatures, humidity or altitude can affect
meter function and blood glucose readings.12
User error is the most common reason for inaccurate results.10 Not putting
enough blood on the reagent is frequently the cause of an inaccurate result.10
Patients should be instructed on how to obtain a blood sample. (See box.)
There are numerous blood glucose meters available, each offering its own advantages and
disadvantages. Although the cost of meters is a major consideration for many patients,
meters have become more affordable. The real cost in monitoring is the strips, which are
not reimbursed by many insurance companies.
There are many features to consider when choosing the most appropriate blood glucose meter
for a particular patient. Meter size and weight may be an important consideration, since
some meters are smaller than others and can be more convenient to carry. Patients prefer
systems that require smaller blood samples, report user errors, are low maintenance, and
are non-wipe and self-timing.15 The wipe system requires the blood to be on the
strip for a measured amount of time, and a result is obtained after the blood is wiped
from the strip.11,14 The patient must wipe all the blood off the strip or the
results will be inaccurate.11 With non-wipe technology, the patient puts a drop
of blood on the strip and waits for the result.6,11 The non-wipe method of
glucose monitoring is thought to provide the most accurate results.14
Test strips are specific for each meter and some are unique. LifeScan’s SureStep has
a test strip that is an absorbent test. Once there is enough blood on the strip, a blue
color appears on the back of the strip. Glucometer Elite’s Active Strip draws blood
from any angle for an accurate result. Glucometer Elite also uses a small amount of blood
(3 µL), so it is not necessary to draw a hanging drop of blood. Strips that need a small
amount of blood are good for individuals with poor circulation and who have difficulty
getting an accurate result. Some test strips also provide visual back-up for the meter,
where the test strip could give an approximate reading. Meters also differ in testing
times, how they are calibrated, method of cleaning, type of batteries used, and memory
features. Some meters have no memory and only provide the last result, while others store
hundreds of results. The memory feature allows patients and their healthcare providers to
track trends of blood glucose values. There are also computer systems/data management
systems that help to track the blood glucose, maintain long-term records of the results,
and provide other information. Some meters have special features, such as an audio readout
for patients who are visually impaired (Diascan Partner). There are also meters that give
directions to patients in different languages.
A new meter that will be available this winter, At Last is the first blood glucose system
to eliminate painful fingersticks. It allows patients to use less sensitive body sites
(upper thighs, lower arms) for testing. In the near future, meters will measure the
concentration of blood noninvasively.
|Appropriate Technique for
Obtaining a Blood Sample
Wash hands with warm water and soap. Dry thoroughly. A wet finger causes blood to spread
rather than form a drop. Milk the finger from the base to insure an adequate flow of
blood. Some patients use a rubber band as a tourniquet.
• Lance the tip or sides of the finger. Avoid the balls of the finger, where nerves
• Hold the finger at a level below the heart with the lanced area pointing toward the
floor. Always use a new lancet and dispose of it properly.
• Once a dangling drop of blood is formed, quickly cover the entire pad. Newer
machines allow you to use smaller amounts of blood and do not require wiping, thereby
eliminating the need for accurate timing.
• Record results in blood glucose diary.
Source: reference 9
Glycosylated Hemoglobin Test
Home blood glucose monitoring is useful in determining the glucose value at the moment of
sampling. Blood glucose values could flucuate throughout the day, and the values between
samplings remain unknown. There are still many who do not self-monitor blood glucose
levels. The glycosylated hemoglobin test or hemoglobin A1c is a measurement of
overall diabetes control for the previous two or three months. The red blood cell has a
life span of 120 days, and this test reflects the blood glucose concentration over that
period of time.14
The glycosylated hemoglobin test measures how much glucose is attached to hemoglobin
cells. Glucose freely passes through the cell membranes of erythrocytes and becomes
covalently bound to hemoglobin.14 This process is irreversible. As the
hemoglobin floats around in the blood, it picks up glucose in about the same proportion as
the glucose that exists in the bloodstream. The greater the amount of sugar in the blood,
the higher the HbA1c results will be. High blood sugars over a long period of
time may increase the risk for diabetic complications. TABLE 1 shows HbA1c
target measurements for people with diabetes. The American Diabetes Association recommends
HbA1c testing at least two times per year in patients who are meeting treatment
goals and who have stable glycemic control, and four times per year in patients whose
therapy has changed or who are not meeting glycemic goals.16
Various laboratory techniques have been developed over the years to assay HbA1c,
including high performance liquid chromatography (HPLC), affinity chromatography,
electrophoresis and immunoassays.17 HPLC, the method employed in the DCCT, is
considered the reference or “gold standard” of HbA1c quantification
methods.18,19 The National Glycohemoglobin Standardization Program certifies
individual clinical laboratories to relate their assay results to those of large-scale
studies such as the DCCT.20
The B-D A1c kit provides a convenient, at-home test for accurately measuring
HbA1c levels. The test allows patients to draw a blood sample at home and send
it to a laboratory for measurement. The kit contains everything the patient needs to
conduct the test: lancet; test paper; instructions; and a prepaid, preaddressed laboratory
mailer. Patients simply put 2 drops of blood on the test paper, let it dry overnight, fill
out the appropriate information, and mail the sample back to the lab within 24 hours. A
central analytical laboratory, which is certified by the National Glycohemoglobin
Standardization Program, receives the sample and performs the assay of HbA1c.
Results are mailed within two weeks to both the patient and the physician. A current HbA1c
measurement should be obtained at least two weeks before an office visit so that therapy
can be adjusted. The B-D A1c At-Home test kit is a valuable tool for enhancing
monitoring of HbA1c in the therapeutic management of diabetes.
A new tool in diabetes management is a glucoprotein/glucose test called Duet Glucose
Control System. One fingerstick indicates if a diabetes treatment plan is effectively
controlling glucose levels. The system tests for both glucose and glucoprotein
(fructosamine). Fructosamine is a measurement of overall diabetes control for the previous
two to three weeks. This allows the physician more flexibility in adjusting the
1. Davidson MB, ed. Diabetes Mellitus Diagnosis and Treatment, 4th ed. Philadelphia, PA:
WB Saunders Company; 1998: VII.
2. Pfeifer MA. Chronic Complications of Diabetes an Overview. In: Funnell MM, ed. A Core
Curriculum for Diabetes Education. 3rd ed. Chicago, Il: American Association of Diabetes
Educators; 1998: 659-675.
3. Simon WA, Howard AR. Can Type 2 Diabetes Be Prevented? U.S. Pharmacist.
4. American Diabetes Association. Position Statement: Standards of Medical Care for
Patients with Diabetes Mellitus. Diabetes Care. 1997;20(1):22-31.
5. Van Velbuizen-Scott MK, Popovich NG. Caring for the Patient with Diabetes. U.S.
6. Dunne N, Zanowiak P, Mellen K. Blood Glucose Monitoring:
What Your Patients Must Know. U.S. Pharmacist. 1994;19(11):81-94.
7. Davidson MB, ed. Diabetes Mellitus Diagnosis and Treatment.
4th ed. Philadelphia, PA: WB Saunders Company; 1998: 195-258.
8. Engle JP. Checking Out Diabetes Home Monitoring Products.
Drug Topics. 1990;134:65-73.
9. Koda-Kimble MA, Carlisle BA. Diabetes Mellitus. In: Koda-Kimble MA, Young LY, eds.
Applied Therapeutics the Clinical Use of Drugs. 6th ed. Vancouver, WA: Applied
Therapeutics Inc; 1995: 48-18-48-21.
10. Peragallo-Dittko V. Monitoring. In: Funnell MM, ed. A Core Curriculum for Diabetes
Education. 3rd ed. Chicago, IL: American Association of Diabetes Educators; 1998: 363-381.
11. Peragallo-Dittko V. Buyers Guide to Blood Glucose Meters. Diabetes Self Management;
12. Kirk JK, Rheney CC. Important Features of Blood Glucose Meters.
J Am Pharm Assoc. 1998;38(2):210-219.
13. Daniel S. Counseling the Diabetic: From Stem to Stern. Drug Topics.
14. Davidson MB, ed. Diabetes Mellitus Diagnosis and Treatment.
4th ed. Philadelphia, PA: WB Saunders Company; 1998: 365-375.
15. White JR. Blood Glucose Monitoring and The Pharmacist. Pharmacy Power-Pak; 1994: 1-34.
16. American Diabetes Association. Standards of Medical Care for Patients with Diabetes
Mellitus. Diabetes Care. 1998;21(suppl 1):S23-S31.
17. Colman PG, et al. Glycohaemoglobin: A Crucial Measurement in Modern Diabetes
Management. Progress Towards Standardization and Improved Precision of Measurement. Med J
18. Little RR, et al. Interlaboratory Standardization of Measurement of Glycohemoglobins.
Clin Chem. 1992;38(12):2472-2478.
19. Halwachs-Baumann, et al. Comparative Evaluation of Three Assay Systems for Automated
Determination of Hemoglobin A1c. Clin Chem. 1997;43(3):511-517.
20. National Glycohemoglobin Standardization Program website: htttp://www.missouri.edu/~diabetes/ngsp/1.html
21. DiPiro JT, ed. Pharmacotherapy a Pathophysiologic Approach.
3rd ed. Stamford, CT: Appleton & Lange; 1997:1505.