Understanding the limitations of laboratory reference ranges

Most patients and doctors mistakenly believe that a laboratory’s reference range defines the “normal range” of a particular measured lab value within a population of healthy individuals.  Any value within the normal range should be normal for a healthy person and any value above or below the reference range should be abnormal for a healthy person.

In reality, reference ranges define something else entirely: they merely represent the lab values of 95% of the patients who are tested by the lab.  Reference ranges vary from lab to lab and change over time as new patients are added to the sample and as different diseases become more prevalent in the general population.

For example since the advent of cholesterol lowering statin drugs, “normal values” for cholesterol have continued to go down representing a larger portion of the population taking statin drugs.

Reference ranges are based on arbitrary statistics, not logic or health.  A lab result falling within the “normal reference range” does not necessarily represent a healthy value.  There is no universally accepted definition of health, just a comparison to the mean or average value measured by the lab.  A “normal” lab test result simply means that the result fall less than 2 standard deviations from the mean on a Standard Normal Curve.

Mean: The average.  The red line in the picture below represents the mean value.

Normal reference range: includes all values that are +2 or -2 standard deviations from the mean.  The normal range includes 95% of all values measured.  It excludes the upper and lower 2.5% of values measured.

Each laboratory determines it’s own reference range for each test.   The actual range is determined by the lab results of patients who are tested: 95% of patients who get a particular test run fall less than 2 standard deviations from the mean and therefore by definition their results make up the lab’s normal range.  5% of patients results are greater than 2 standard deviations from the mean so their lab results are by definition “abnormal.”  Of these abnormal values, 2.5% fall above the reference range and 2.5% fall below.

Laboratory reference ranges represent what is normal for a sick population.

Laboratories don’t use thousands of optimally healthy subjects to determine their reference ranges for each test.  Instead, they use the data obtained by testing patients whose doctors have requested that they have a particular test run. These are people who’s doctors already suspect have a problem in the area being tested and do not represent a healthy popluation.

Insurance contracted physicians must restrict their testing to what is absolutely necessary and can be justified as essential to the patient’s insurance company.  In other words, physicians may only order tests that have a high chance of being abnormal and only test patients who are obviously sick.

Today the percentage of sick individuals who help determine a lab’s reference ranges is higher than ever before because of the tremendous pressure that insurance companies place on physicians to minimize the number of tests ordered in order to reduce costs.

Laboratory reference ranges are often too broad to be meaningful.

All reference ranges contain the values obtained from both healthy and unhealthy people.  Remember that reference ranges are purely statistical: by definition, 95% of the people who walk into the lab to be tested will be normal whether they are healthy or not.

The reference range for thyroid stimulating hormone, TSH is a good example of an excessively broad reference range.  Labcorp’s reference range for TSH is: 0.450 uIU/ml – 4.500 uIU/ml.  Patients with TSH of 4.500uIU/ml have a 10 times more TSH than individuals with 0.450 uIU/ml it is absurd to consider both as equally “normal” or “healthy”

Why does my doctor say that everything is fine when I don’t feel good?

Many people can relate to this scenario. You go to the doctor every year and are told: “all your labwork is normal”.  All of the sudden, after 10, 20, or 30 years of being normal, you all of the sudden are diagnosed with a severe disease such as cancer or diabetes which we know does not happen over night.  Another common scenario is “I have been feeling bad for a long time and my doctor insists that nothing is wrong because my labwork is normal.”  What this really means is that your labwork does not fall within the 5% of cases that are so severe that your insurance company will allow him to diagnose and treat you.

How should the reference ranges for variations of normal between healthy individuals be determined?

Lab data obtained from young adults who do not have health concerns severe enough to make them seek medical help is more likely to represent the optimal, healthy levels of a particular lab value.

However, each patient’s optimal range for each lab value is unique.  Because each patient is an individual, lab testing alone cannot diagnose or rule out an imbalance in most cases.  Since each person is unique, physicians should base their diagnosis on all of the information available: presenting symptoms, family history, lifestyle, physical exam findings, and lab results.

This comprehensive approach allows not only the detection of advanced disease: the relatively rare, severe, and obvious imbalance that causes lab results to fall outside of the lab’s normal reference range.  It also allows the detection of the much more common and subtle clues that indicate the start of a disease process that is not yet obvious enough to cause a test value to fall outside of the lab’s reference range.  Ultimately, correcting a health imbalance before it becomes a diagnosable disease by conventional medicine’s low standards of “normal” is true disease prevention.