A Clinical Dilemma: What is Normal?
Men with a PSA level above the median for their age group are at higher risk for prostate cancer.
Predetermined thresholds tell clinicians what normal is, but these values are arbitrary.
In this first of two parts, Renal & Urology News focuses on the problems with establishing a “normal” PSA value and how this impacts on testing.
The concept of what is “normal” enters into nearly every aspect of clinical decision making. For physicians to declare something abnormal and make a diagnosis, they must first know what normal is.
Often, this is a matter of thresholds, as with diagnostic tests. For any substance being measured in blood, for example, there must be a predetermined value that is considered normal. Above or below that level would be deemed abnormal and hence a cause for clinical concern.
But what is normal and who decides this? In an editorial written 35 years ago in Clinical Chemistry (1975;21:1873-1877), noted pathologist F. William Sunderman Jr, MD, PhD, observed: “Investigators who have explored the conceptual pitfalls in the definition of ‘normalcy' have concluded that there is no absolute method of defining ‘normal values' or ‘normal ranges' for laboratory tests.”
Dr. Sunderman had developed a pioneering method of measuring blood glucose and invented quality control techniques for clinical laboratories. He wrote the editorial to summarize “the problems and ambiguities that are causing many clinical chemists to discard the term, ‘normal values,'” and to describe and define two then-new terms to replace it: reference values and discrimination value.
Three and a half decades later, it appears that not too much has changed. “Thresholds are always going to be sort of arbitrary,” notes Thomas H. Hostetter, MD, Nephrology Division Chief at Albert Einstein College of Medicine in Bronx, N.Y. “If I say 140/90 [mm Hg] for hypertension, why isn't it 145/95 or 135/85?” Sometimes researchers do establish thresholds above or below which the risk of other medical problems is increased.
“We know that the risks of stroke, heart attack, and other vascular disease start to get higher and higher, the higher your blood pressure is,” says Dr. Hostetter, who recalls that as recently as the 1960s, there were people who believed that hypertension did not need to be treated. “And that's the way somebody decided that yes, this epidemiologic association can be influenced by therapy; therefore measuring blood pressure [and settling on a “normal” range] is worthwhile.”