Three Ways Practitioners Are Botching Blood Chemistry Interpretation [Part 1 of 3]

Jul 27 / Drs. Bryan & Julie Walsh
Listen, as an industry we can't mess up the most valuable and cost-effective test available today, a standard blood chemistry test.

We get it - it's poorly taught. Even in our 4-year doctorate program, we were taught only conventional methods. And as for the vast majority of online courses, seminars, and materials available today (except ours, of course), they're all the same, repackaged information that lacks scientific evidence to support their claims.

We want you to buckle up and come along as we share three ways that most practitioners are missing the boat on this incredible tool.

Before that, let us point out that the first way to mess up a blood chemistry test is by not using it at all and/or using other tests instead. Practitioners love organic acid tests, urinary hormone tests, stool tests, and genetic tests. While those tests are attractive to run, fun to talk about, and entertaining to look at, none have the scientific scrutiny or accuracy blood chemistry tests have. Clinically, they can't hold a candle to a blood chemistry test.

A standard blood chemistry test is the most cost-effective, scientifically validated, accurate, and most studied lab test on the planet, which no other lab can say, but who's counting.

If a practitioner IS using blood chemistry in their practice, here are the top three ways they are messing it up. (By the way, it's not anyone's fault. The fact is, this isn't being taught. We're trying to change it.)

#1 Practitioners Are Not Using Optimal, Evidence-Based Reference Ranges.

Reference ranges may not seem very exciting, but MAN is it important and a mess today, which you are about to see. 

Reference ranges matter because the most valuable skill in clinical practice is diagnosis, diagnosis, and diagnosis. It doesn't matter how nice of a person you are, what your office smells like, or how many degrees you have listed after your name, if a patient comes to you with an issue, your ability to diagnose them is everything.   

If the reference range is not accurate, a patient won't get diagnosed properly, won't get treated, and may continue to be sick or unwell.  

Having valid and accurate reference ranges is SO critically important, various initiatives around the world have been done to improve this issue. For example, there's:
  • Canadian Health Measures Survey (2015) used data from 12,000 Canadians to determine reference ranges for over 50 biomarkers
  • Aussie Normals (2014) studied just under 2000 people to establish reference ranges for 91 biomarkers.
  • The Nordic Reference Interval Project (2004) attempted to do the same for 25 biomarkers across five Nordic countries.
And there's more. Turkey has done this. Italy has done this. All to shed light on this incredibly important subject. 

So you think with all these initiatives, we'd have gotten somewhere, right?

Wrong.

Let's take a look at one marker, albumin, to demonstrate. Using three of the initiatives above and some other labs, universities, and medical clinics, here are some reference ranges for albumin: 
4.4 – 5.1 (CHMS)
3.6 – 4.8 (NORIP)
4.3 – 5.6 (Realab)
4.0 – 5.0 (LabCorp)
3.6 – 5.1 (Quest)
3.5 – 5.0 (Mayo Clinic)
3.4 – 5.4 (UCSF)
3.3 – 4.8 (AAFP)

In case you just skimmed that, take a look at them again. Does anyone else see this is a serious problem?

Based on this, you could have low albumin in one clinic and be treated according, while in another clinic you'd be considered totally fine?! 

This is an enormous problem. 

The wrong reference interval can result in a missed diagnosis, inappropriate treatment, incorrect information told to the patient, increased healthcare costs, and more.  

What should be done?

Until someone wants to dump millions of dollars into this to straighten it out, scientific papers are available to give us some insights into a more optimal reference range, if you're willing to take the time and look around a bit.   
Let's continue using albumin as an example.
First of all, one should look for the average, or median, amount of a given biomarker in healthy individuals. One semi-recent study suggests that median serum albumin is 4.7 g/dL. This type of information doesn't offer a range, but is helpful because it starts to give a ballpark of where a marker might be.   

The next thing to do is find more specific and targeted papers to help give us an idea of a healthy, optimal range for a given marker.  

Here's an example. This paper used 1.7 million "healthy" individuals and found an increasing incidence of mortality over a 12-year period as albumin crept below 4.3 g/dL. Specifically, if one's albumin was between 4.0-4.1 g/dL or 3.9-4.0 g/dL, the relative risk of mortality increased by 153% and 182%, respectively. That's pretty significant and considerably different than the low end of most of the ranges listed above.   

But that isn't the only paper. When you look around a bit, you find that other papers are suggesting a low-end of albumin might be somewhere in the mid 4's (g/dL). 

For example, all cause mortality was the lowest when albumin was above 4.7 g/dL for men and women, hospitalized patients with albumin greater than 4.5 g/dL had the greatest chance of survival, and albumin below 4.4 g/dL was associated with vascular dysfunction.   

Based on these and other studies, we believe a healthy albumin level is somewhere between 4.5 and 5.0 g/dL. Could arguments be made for 4.4. and 5.1 g/dL? Sure. This isn't black and white, nor is it an exact science. Still, given all the data we've looked at, albumin in the neighborhood of 4.5-5.0 g/dL is probably a good target to shoot for. It's a heck of a lot better than the ranges listed earlier and has some decent scientific evidence behind it.

In Closing

Make sure you are using evidence-based optimal reference ranges from studies based on healthy individuals. We shared the example above as one example of what you can learn in our Blood Chemistry Interpretation course.

Read about the #2 way practitioners are botching blood chemistry interpretation here.