Criterion Standard

The expression criterion standard, according to the AMA Manual of Style, represents the “diagnostic standard for a particular disease or condition, used as a basis of comparison for other (usually noninvasive) tests. Ideally, the sensitivity and specificity of the criterion standard for the disease should be 100%.” This definition on its face seems a fairly straightforward way to identify the best method for making a diagnosis or the best treatment plan for a given disease.

A controversy, however, emerges in a parenthetical phrase that suggests the alternate expression gold standard be avoided because it “is considered jargon by some.” This assertion is supported by a reference to A Dictionary of Epidemiology, third edition, by John M. Last, published in 1995. Treating it as slang, his entry not only presents gold standard in quotes but, to be sure the reader understands his meaning, follows it with the word “jargon” in parentheses before defining it. Last provides no alternate expression and does not include criterion standard in his dictionary. He does, however, include criterion, which he defines as “[a] principle or standard by which something is judged. See also STANDARD.” And that is defined as “[s]omething that serves as a basis for comparison.”

In his fourth edition, however, Last does not include a gold standard entry. His definitions for both criterion and standard remain the same, and because they have nearly identical definitions could account for his not including an entry for criterion standard.

The gold standard entry returns in the fifth edition, edited by Miquel Porta. It is again presented in quotes but omits the parenthetical naming of it as jargon and defines it as “[a] method, procedure, or measurement that is widely accepted as being the best available. Often used to compare with new methods of unknown effectiveness (e.g., a potential new diagnostic test is assessed against the best available diagnostic test).”

Looking at another source, Annals of Internal Medicine Editor Hal Sox in his book Medical Decision Making never mentions the expression criterion standard. He does, however, talk about gold standard, which he takes out of the realm of epidemiology and into clinical practice by defining the gold standard test as “[t]he procedure that is used to define the true state of the patient.”

Although the “when” of its adoption as the preferred expression for JAMA and the Archives Journals seems to have escaped memory, the “why” remains among most of the medical editors. Some suggest avoidance of gold standard because it crosses disciplines from economics to medicine. As an economic term, it had served as the basic support of paper money. Another consideration, offered by former JAMA Deputy Editor Richard Glass, is that “gold standard …implies more of a sense of permanence than is appropriate for scientific topics.” With new knowledge, he reasons, comes new standards.

Practice, however, defies style preference. A search of JAMA articles in 1998 shows that criterion standard was used 7 times while gold standard was used 42 times. Jumping ahead 10 years, the trend holds: criterion standard was used 9 times; gold standard, 35 times.

Yet the practice of using gold standard over the style recommendation may all boil down to what JAMA Deputy Editor Drummond Rennie wrote in an e-mail. “If we are prepared to consider using ‘criterion standard,’ we should really prefer ‘criterion criterion’ (though ‘standard standard’ sounds a tad less pompous, even if just as meaningless). We all know what ‘gold standard’ means. It has the merit of being customary, memorable, understandable.”

And isn’t that the job of editors? — Beverly Stewart, MSJ

One thought on “Criterion Standard

  1. Totally agree with Rennie and with Stewart (end of post). Slang that doesn’t die off will eventually become standard register. The good neologisms or borrowings (ones that succinctly fill a need in the communicative toolbox) are the ones that last, and they eventually lose their novelty. Since that happens pretty fast in STM nowadays, editors are better off keeping up. Trailing a little bit to stay on the conservative side is appropriate to the role; but it’s important not to trail /too/ far back, lest one take on the role of the annoying laggard that pesters the group to wait up. Glass has a great point (“implies more of a sense of permanence than is appropriate for scientific topics”), but Rennie’s point (“the merit of being customary, memorable, understandable”) will override it in the end anyway. And that’s OK. Thanks for the great post.

Leave a Reply

Your email address will not be published. Required fields are marked *