WHY DO I HAVE ABNORMAL LABS?
“Doc, I feel fine, but when I looked on the portal, my labs are abnormal. What’s wrong with me?”
I hear this several times a week, and even more common are the folks who do not bother to look at the portal, but become very disturbed when they come in to review the labs with me and see highlighted “abnormal” values. So, let me explain how the labs decide what normal values are and why, in many cases, “abnormals” do not mean there is anything at all wrong with you. You may be perfectly healthy and “normal” even if you have lab values that are not listed as being in the “normal range.”.
To begin with, let’s talk about what we mean by “normal” and why I have been putting the term in quotation marks. We all have a visceral understanding of what we feel is “normal” in our daily lives. We may not all agree on what this means, but we all have a feeling for it. When talking about lab values we all use that term and discuss “normal” values. However, if you look at the lab printout, the word normal is never used. The term you are likely to see is “Reference Range.” This term is used because it avoids the value judgements associated with “normal” or “abnormal.”
Most people intuitively think that having a lab value that falls outside of the “reference range” makes that value “abnormal,” and by extension, it makes them or their body “abnormal.” This is not at all the case.
So, how do the labs determine the Reference Range? Surprisingly, there are totally different methods for different lab tests, which can lead to confusion. For the majority of the lab values you see, the method is statistical. The Reference range is chosen to include 95% of some population group, usually young, healthy people.
How is this done? They might, for example, choose a group of 200 young healthy adults to study. The 200 number is suggested by some academic pathologists, although larger numbers will give more meaningful results. The blood is drawn and analyzed, and the 200 values that result are lined up in numeric order from the lowest to the highest. The lowest 5 numbers, or 2.5%, are deleted, as are the highest 5 numbers. The Reference Range, then, runs from the value for person number 6 to the value for person number 995. In most cases, that is all there is to it.
The remaining 190 people are the 95% who fall in the Reference range. The ten who were not in the range are usually perfectly healthy, but have a lab value that appears “abnormal.” This should not worry you, or even surprise you. Some simple statistics will show why most people have one or more abnormal value on their lab report.
If you are healthy and only one lab test is drawn, your chance of being in the reference range is 95%, by definition. If two tests are drawn, assuming the two tests do not affect each other, the chance of having both tests in range is 0.95 times 0.95, or 90.2 %, mathematically referred to as 0.95 to the second power.
If this is carried out to ten tests, the chance of having all ten normal is 59.9% (0.95 to the tenth power). Most of the people I see have about 20 tests drawn. For them the chance of having all be normal is 35.8% (0.95 to the 20th power).
In short, the majority of healthy people have one or more abnormal tests. So do not fret. Your physician should tell you if the number that is not is the reference range is anything that should be further evaluated or treated. Usually it is not.
Another issue with defining “normal” is deciding who should be in the study group. Sometimes the laboratory supervisors attempt to look at a random group of the population they serve. More commonly, they try to get a group of “young healthy” individuals, often half male and half female. But how do you know they are healthy? Usually the potential subjects are given a health questionnaire. Anyone who has tried to use health questionnaires know how often people fail to report something important. The supervisors then have to decide which conditions would exclude a person from the test group. Which illnesses might matter? Which medications, given the fact that many people do not know exactly which medications they are taking? Which supplements or over the counter meds, given the fact that these are sold under a huge number of brand names, and it is often hard to determine exactly what is in them. These decisions can skew the range substantially.
I remember that many years ago, some genius decided that the best way to be sure they were studying young healthy people was to use boot camp graduates. After all, they are all young with an average age of about 22. They were all healthy or they probably would not have made it through boot camp. What could go wrong? Well, how closely do you think a 22-year-old male who is walking ten to twenty miles a day reflects your current body? Modern ranges are not so severely skewed, but the principle remains. Additionally, some tests will be influenced by age, gender, ethnicity, climate, prevailing diet in that population, and a slew of other factors.
Finally, you need to recognize that there are some conditions that are totally innocuous that will routinely give lab values out of the normal range. One example is called Gilberts Syndrome. Let me explain. Many standard lab scans routinely look at what are referred to as liver function tests. One of these is Bilirubin. This is a blood breakdown product that is cleared by the liver. If the liver starts to fail, as with cirrhosis, the bilirubin level will usually rise. However, there is another congenital condition, Gilberts Syndrome, in which the liver is absolutely normal, but these people have one enzyme involved with bilirubin metabolism that is less active than normal. They will routinely have an elevation of bilirubin but are actually perfectly healthy. They have nothing to worry about.
I have talked about the population statistics used to determine most reference ranges for lab tests. However, there is another group of lab tests for which the reference ranges on the lab reports have absolutely nothing to do with any sort of average of a population. They are actually what are called “consensus decision points.” In other words, they reflect what some group of experts believe peoples numbers should be, as opposed to what they actually are. Unfortunately, the labs rarely if ever flag the fact that these are different than the statistical reference ranges.
These “consensus” ranges include many of the labs we care the most about: Cholesterol and triglycerides, blood glucose and Hemoglobin A1c, and Vitamin D.
These “normals” bring up a whole list of other concerns, which will be the subject of another post.
Stay tuned.