How are reference ranges determined?
The short answer to this question is by testing a large number of people who have key similarities and observing what appears to be "typical" for them.
To determine ranges, labs may conduct their own studies for the tests they perform, they may adopt reference ranges from test manufacturers or other labs, or they may derive reference ranges from existing patient data.
- The most important step in determining a reference range for any test is to define the reference population – the group of people who will be represented in the reference range. Depending on the test and factors that may influence its results, reference populations may be chosen based on age, sex, race, general health, and/or medical history.
- Next, a large number (minimum of 120) of people who fit the profile of the reference population are tested under nearly identical conditions, and the results are analyzed.
- For many tests, reference ranges include the values that are statistically analyzed and reported for the middle 95% of the reference population.
Why might there be more than one reference range for a test?
For some tests, there is no single reference range that applies to everyone. By far, the most common factors that cause variations in reference values are age and/or sex.
For example, the range used to evaluate a creatinine test result depends on a person's age and sex as shown in the table below.
|Age||Sex||Creatinine Reference Range|
|18-20 years||Male||0.9-1.3 mg/dL|
|60-90 years||Male||0.8-1.3 mg/dL|
Creatinine is produced as a natural byproduct of muscle activity and is removed from the blood by the kidneys. Creatinine is often measured as a gauge of how well the kidneys are functioning. Creatinine levels are affected by a person's muscle mass. Because men have greater muscle mass than women, the reference values for men are higher than for women.
Reference ranges for children who are still growing and developing are often very different than those for adults and may change significantly as they age. Alkaline phosphatase (ALP), for instance, is an enzyme found in the cells that make bone, so its concentration in the body rises in proportion to new bone cell production. The reference values for ALP in children and adolescents, who should be growing healthy bones, are higher than they are for adults in whom high ALP levels may signify bone or liver disease.
Some reference ranges apply only to certain subsections of a population. For example, pregnancy changes many aspects of a woman's body chemistry, so what is typical for pregnant women is not typical for those who are not pregnant. Thus, pregnancy-specific reference ranges are established for a variety of tests.
A very short list of examples of additional factors that might cause a lab to establish more than one reference range for a test includes:
- The type of sample used (for example, whole blood, plasma, serum, urine, saliva, or other body fluid)
- The time of day when a sample is collected
- Whether the test is performed on a patient who has been fasting
- The phase of a female patient's menstrual cycle at the time of testing or whether she is in perimenopause or menopause
The reason a test is performed sometimes guides which reference range is used to interpret results and guide treatment decisions. For example, the hemoglobin A1c (HbA1c) test can be used to screen for diabetes risk, aid in confirming a diagnosis of diabetes, or monitor how well a person with diabetes has controlled his or her blood sugar over a three-month period. The table below shows how a provider might use the results of the hemoglobin HbA1c test to guide treatment decisions.
|Diabetic Status||HbA1c Level|
|Nondiabetic||Less than 5.7% (39 mmol/mol)|
|Prediabetic||5.7% - 6.4% (39 - 46 mmol/mol)|
|Diabetic||6.5% (47 mmol/mol) or greater|
|Well-controlled diabetes||Less than 7.0% (52 mmol/mol)|
Let's look at two people who have the same HbA1c result of 6.6%, one who is being screened for diabetes and another who is already known to have diabetes.
- For the person who is being screened, a result of 6.6% would lead a healthcare provider to suspect that the person has diabetes. The HbA1c test must be repeated at a later date and if the second result is again greater than 6.5%, a diagnosis of diabetes is confirmed.
- For the person who has already been diagnosed with diabetes, an HbA1c result of 6.6% is good news. It means the condition is being well controlled by the current treatment (whether diet and exercise and/or medication).