The Concept of Normal

The result of a laboratory test is compared with a reference standard, which traditionally has indicated values that are seen in healthy persons. Using the terms "normal results" or "normal range" implies that there is a clear distinction between healthy and diseased persons, when in reality there is considerable overlap.

The current standard of comparison for laboratory results is the reference range, which is frequently defined by results that are between chosen percentiles (typically the 2.5th to 97.5th percentiles) in a healthy reference population. Several problems are encountered when deriving a reference range. Often, the reference population is not representative of persons being tested. Differences in gender, age distribution, race, ethnicity, or the setting (hospitalized vs. ambulatory patients) between the reference population and the person receiving the test may be present. The person being tested

Table 15-1 Biologic Variables that Affect Test Results Table 15-2 Probability that a Healthy Person Will Have an

Abnormal Result with Multiple Tests

Biologic Rhythms

Circadian

Ultradian

Infradian

Constitutional Factors

Gender

Genotype

Extrinsic Factors

Posture

Exercise

Diet: caffeine

Drugs and pharmaceuticals: oral contraceptives

Alcohol use

Pregnancy

Intercurrent illness

From Holmes EA. The Interpretation of laboratory tests. In McClatchey KD (ed). Clinical Laboratory Medicine, 2nd ed. Philadelphia, Lippincott-Williams & Wilkins, 2002, p 98.

Number of Independent Tests

Probability of Abnormal Result (%)

1

5

2

10

5

23

10

40

20

64

50

92

90

99

Infinity

100

From Burke MD. Laboratory tests: basic concepts and realistic expectations. Postgrad Med 1978;63:55.

should be tested under similar physiologic conditions (e.g., fasting, sitting, resting) as the reference population. The size of the reference population may be too small to include a representative range of the population.

Two statistical methods, parametric and nonparametric, are generally used to define the reference intervals. The parametric method applies when the results of the sample population fits a normal gaussian distribution, with a bell-shaped curve around the mean. In this case, the 2.5th and 97.5th percentiles can be calculated using statistical formulas. When the reference values do not follow a normal distribution, nonparamet-ric methods are used, arranging the results from the reference subjects in ascending order, and identifying values between the 2.5th and 97.5th percentiles as within the reference range.

Reference ranges for a particular test can be the manufacturer's suggested reference range or may be modified because of differences in the population using the laboratory. The Clinical Laboratory Improvement Act of 1998 (CLIA) has defined three requirements for reference values: the normal or reference ranges must be made available to the ordering physician; the normal or reference ranges must be included in the laboratory procedure manual; and the laboratory must establish specifications for performance characteristics, including the reference range, for each test before reporting patient results. Using the manufacturer's reference range is valid when the analytic processing of the test is the same as that done by the manufacturer and when the population being tested is similar to the reference population used to define the reference range. When selecting a reference range that includes 95%

of the test results, 5% of the population will fall outside the reference range for a single test. When more than one test is ordered, the probability increases that at least one result will be outside the reference range. Table 15-2 compares the number of independent tests ordered with the probability of an abnormal result being present in healthy persons.

Alternative Medicine

Alternative Medicine

Discover How to Heal Yourself Naturally! Safe, All-Natural Alternatives to Taking High-Cost, Side-Effect-Laden Prescription Drugs! If you are tired of paying the high cost of prescription drugs if you are tired of worrying about prescription drug side affects or if you just dont want to risk becoming dependent on prescription drugs then have I got great news for you!

Get My Free Ebook


Post a comment