Posted by Sten Westgard, MS
Earlier this week, I had the pleasure of attending the AACC webinar on Biologic Variation. It was a presentation by Dr. Callum Fraser, the noted international expert on biologic variation.
In the webinar, Dr. Fraser covered a wide variety of topics, ranging from within-subject biologic variation, reference intervals, reference change values (RCV), and result reporting. He used examples that included creatinine, troponin, and glucose.
I thought he made several interesting points:
- Population-based reference intervals are not as useful as we desire/expect, particularly with analytes that have a lot of biologic variation and biologic individuality (quantified by the Index of Individuality, II). A person could be experiencing a significant change from their own homeostatic set-point, but still be well within the population reference interval (and thus the lab and clinician might not pick up on the change). Stratified reference intervals (by age, gender, etc.) can help mitigate this issue. RCVs are even better.
- While there are some expected differences in biologic variation between healthy and sick patient populations, Ricos' latest update on Within-Subject Biologic Variation in Patients with Disease shows that for the majority of the analytes studied, the biologic variation is the same order in disease and health. That means we can have more confidence in using the larger database on within-subject biologic variation for healthy individuals.
- For troponin, while some guidelines state that a 20% change is significant, the calculated RCV from one 2009 study for troponin I suggested that it would require at least a 46% increase in serial results to indicate a significant change in the short-term, and a 32% increase on the long-term. (A 81% decrease in the short term, and a 45% decrease in the long-term, was significant). Guidelines with nice, round-number recommendations are probably not wholly evidence-based.
Here are a few links to articles Dr. Fraser discussed or referenced during his talk.
Ricos Group Database on Within-Subject Biologic Variation within Healthy Individuals(and Desirable Specifications for Imprecision, Bias, and Total Allowable Error): http://www.westgard.com/biodatabase1.htm
Ricos Group Database on Within-Subject Biologic Variation in Patients with Disease:
One of Fraser's essays on Biologic Variation: http://www.westgard.com/biological-variation-data-for-setting-quality-specifications.htm
Fraser's essay on the use of Biologic Variation data: http://www.westgard.com/are-scientific-statements-the-scientific-truth.htm
The introduction to Fraser's book, Biologic Variation: http://www.westgard.com/biologic-variation-principles-and-practice.htm
Finally, here's the link to purchasing Fraser's essential book from the AACC store: http://direct.aacc.org/productcatalog/product.aspx?id=1541
Dr. Fraser's work has the potential to transform how we interpret, report, and even auto-verify test results. The implications of the work on biologic variation, for some tests, show that our methods need to significantly improve to match the desired clinical decision-making.