Doctors have long tried to find new uses for old drugs, and in this quest, centuries may pass before someone discovers hidden benefits in standard medications.
Yet even when important new uses are found, it may be decades before the benefits are confirmed and widely recommended.
A case in point is aspirin, which is one of the oldest and most commonly used drugs.
In a new use documented only in recent years, aspirin is now recommended for millions of Americans who are at risk for the prevention and treatment of heart attacks and strokes.
And only in recent weeks has the modern generation of doctors begun to appreciate that the new use derives from observations made by Dr. Lawrence L. Craven, a general practitioner in Glendale, Calif., more than 40 years ago.
Dr. Craven noted an unusual incidence of bleeding complications among children who chewed gum containing aspirin to relieve the pain from a tonsillectomy.
Probably no other new use for an old drug took longer to discover, has had a greater impact on public health, and has had as fascinating a history.
The history of aspirin and related compounds can be traced to two Greek physicians, Hippocrates, who lived from 460 B.C. to 377 B.C., and Galen, who lived from A.D. 130 to A.D. 200.
These physicians recommended that their patients treat a variety of ailments by chewing willow bark, which contains salicylic acid, a close chemical cousin of aspirin.
Aspirin, which is acetyl salicylic acid, was first synthesized in the 19th century, but it lay dormant on the shelves of the Bayer Co. in Germany until it began to be used in a few patients just before the turn of the century.
But no one seems to have seen aspirin's potential against heart attacks and strokes until Dr. Craven began his studies in the late 1940s, according to an article in the June issue of the Archives of Internal Medicine by the journal's editor, Dr. James E. Dalen of the University of Arizona in Tucson.
One reason such a link could not have come much earlier is that doctors began to diagnose heart attacks in patients who lived through them only after 1912.
Before then, non-fatal heart attacks were passed off as indigestion, gallbladder attacks or other problems.
Little is known about Dr. Craven, who was a graduate of the University of Minnesota medical school.
The crux of his discovery was his observation of the bleeding among children who chewed aspirin gum after tonsillectomies.
Noting reports that aspirin affected blood's ability to clot, he inferred that aspirin would be safer than dicumarol, an anti-coagulant that doctors had begun prescribing for heart attack patients in the 1940s.
Dr. Craven recommended an aspirin a day for healthy men between the ages of 45 and 65, particularly those who were overweight and led sedentary lives.
He made the same recommendation for people who had recovered from heart attacks.
His advice was based on studies of nearly 8,000 patients. He reported the studies, from 1950 to 1956, in obscure medical publications, including the Mississippi Valley Medical Journal. Dr. Craven was astonished by his own findings: no detectable cases of heart attacks and strokes among those studied.
It was a time when the statistical techniques used in modern clinical trials were just being pioneered, and Dr. Craven was aware that he lacked a comparison group.
Before he died in 1957 at the age of 74 from angina, a year after publication of his final article, Dr. Craven invited others to test his thesis with more sophisticated techniques. Larger studies did come, but it took more than a decade for them to start.
The reasons for the delay are not clear but they partly reflect the tendency of scientists to insist on understanding the biological mechanism for a treatment before studying it.
Dr. Craven's reports help refute the dogma that discoveries come only from full-time scientists working at research institutions and demonstrate that inquisitive doctors can do research in any setting.
But observations alone are not enough to change medical practice because they may be wrong and may have resulted from unrelated and confounding factors.
Also, therapies based simply on observation may turn out to be more hazardous than beneficial.
Paradoxically, aspirin poses risks of bleeding that can cause strokes, and aspirin can also cause gastrointestinal symptoms. But studies have shown that the benefits generally outweigh the risks.
Although Dr. Craven noticed the bleeding that his tonsillectomy patients suffered, many doctors overlook similar clues that abound in everyday medical practice.
Doctors do detect many other clues, but they usually do not take the next steps of reporting the observation and starting studies to document the effect. Those are the crucial steps that help distinguish the gadfly from the true scientist.
The link between rubella (German measles) and birth defects is among the crucial findings that astute practitioners have made.