Guest post: Defining “normal”

CDI Blog - Volume 11, Issue 134

Howard Rodenberg,

By Howard Rodenberg, MD, MPH, CCDS

When my son was in middle school, I would call him every evening to review the school day. I would want to know all about the things he was learning and the friends he was making, and he would grudgingly oblique me with a one-word answer if he was in a good mood. I asked about each class in turn. “How was English? How was Math? Did you have fun in science? Anything going on in social studies?” After a while, his responses being reduced to a series of incomprehensible tones, I tried to spice up the questions in order to generate some kind of response. “Did you eat lunch? Did you use any oxygen?" It got to the point where all I would have to say was "How was school?” and he would come back with, “No, no, no, no, yes, yes, no.” Even in college, those are the answers I pretty much still get.

The girlfriend has a son in middle school, and as a Concerned Individual (not my official title...that would be The Guy Who Lives With My Mom, which is how he introduces me), I’ll inquire as to his school day as well. His preferred response is, “Normal,” except when he was taking Spanish, in which case it was “norr-MAAL.” (My keyboard doesn’t have an accent mark.) I have no idea what a “normal” school day actually is, but I suspect it spans a range of educational activities that suffused the environment while he was daydreaming of Face Time with his iPad.


The simple fact is that most of the time, I don’t know what normal means. Your normal is most likely different than mine. But different versions of normal aren’t necessarily bad, right? One’s normal does not mean others are abnormal, mostly because often nobody really knows what “normal” means. On the other hand, there can be abnormalities within the realm of normal. Got that?

I was thinking about this last week as I dipped a toe in the endless swamp those who work in the CDI world call “clinical validation.” I’ve got my own rant about why we’ve been forced into this process, which someday I’ll share if I can find enough words to describe my feelings.

For now, the question of “normal” arose when looking to clinically validate charts for acute kidney injury (AKI). The specific question concerned a patient whose creatinine went from 1.2 mg/dl on admission to .08 mg/dl by time of discharge. The physician had documented AKI, but both values were within the normal range for the patient’s age and sex. Can you have AKI with normal values?

The short answer, if you’re running to catch that late bus home, is yes.  But if you’ve got a minute, I’d like to explain why.

To start, it’s important to know what “normal” really means in the context of a laboratory test. In physiology, there is no fixed value for “normal.” There is no single volume or quantity for any lab test that is shared by 51% of the healthy population to establish a standard. Instead, think of lab values as a graph, with each individual value being plotted against a numerical axis. Some will be higher than others, some lower, and still more will cluster towards the middle. But what we call a “normal range” is purely a statistical calculation of the range of values that includes 95% of the population, and which concurrently has a predictive value of 95% that the value for that test in a healthy person will fall within that range. One person’s creatinine may be 1.2, another’s may be 0.6, and both can fall within the normal range for the population as a while. What “normal” means for a particular individual, however, is a different kettle of fish, and differs with height, weight, sex, nutritional status, and a billion other parameters to determine an individual’s number.

Those working in CDI need to understand the difference between population numbers and individual values. A good way to think about it is with an analogy of cardiac risk factors. We know that if we examine an entire population, those who smoke, who are obese, and who have medical conditions such as hypertension and diabetes are more prone to heart attack. However, on an individual basis, risk factors mean nothing. You can have a heart attack with all the risk factors, or with none. Your "normal" person can still have significant disease.

Using AKI as an example, we can clearly see how “normal” values within an individual may actually represent pathology. I happen to use the Kidney Disease Improving Global Outcomes (KDIGO) Clinical Practice Guideline for AKI criteria as my gold standard reference for AKI. One of the definitions of AKI within KDIGO states that AKI can be diagnosed if there is an increase in serum creatinine of > 0.3 mg/dl within 48 hours. Of course, in order to demonstrate change you need to know a baseline, and KDIGO allows you to determine the baseline either from historical values or from the response to treatment. Understanding that everyone’s baseline is different helps explain why two values within the “normal” range, when properly interpreted, can still demonstrate evidence of AKI.

What about the circumstance where the creatinine drops iatrogenically because of aggressive fluid therapy? Maybe it wasn’t AKI after all? That's a question for your clinical validation team, who can look at the gestalt of the chart to get to the answer. But a "normal” creatinine, or before and after values both within the "normal” range, do not preclude the diagnosis of AKI.
I still ask my collegiate son about school, and I still get one-word answers. However, his love of academics is such that now his answers have changed from two letter words to all fours. Higher education in America: Value for money, to be sure.

Editor’s note: Rodenberg is the adult physician advisor for CDI at Baptist Health in Jacksonville, Florida. Contact him at or follow his personal blog at Advice given is general. Readers should consult professional counsel for specific legal, ethical, clinical, or coding questions. Opinions expressed are that of the author and do not represent HCPro or ACDIS.