In 1874, eight years before Robert Koch presented his discovery of the bacterial agents that cause tuberculosis, a lesser-known microbe hunter, Arthur Boettcher, published a paper on a small, curved bacterium that he found repeatedly in ulcers of the stomach. Over the next half century, several other scientists confirmed Boettcher’s finding. Some also extended the research by experimentally transmitting the bacterium in lab animals. By the late 1940s peptic ulcers were being successfully treated with antibiotics in New York City hospitals. Paul Fremont-Smith was a young intern in 1948 at Manhattan’s New York Hospital. He remembers the orders he was given there by his no-nonsense supervisor, Connie Guion. She told him, “The people over at Mount Sinai have found that Aureomycin [a trade name for chlortetracycline] is effective against peptic ulcers. Use it. It works.” He did, and it worked. Then, around 1950, discussions of infectious causation of ulcers disappeared from the literature and from the treatment regimen. The medical texts from 1950 through the early 1990s attributed peptic ulcers to gastric acidity, stress, smoking, alcohol consumption, and genetic predispositions—everything but infection. Generally there was not even a reference to the possibility of infectious causation.
Around 1980, after three decades of medical impotence against ulcers, researchers in Perth, Australia, were back on the infection trail. The driving force was a young internist named Barry Marshall. He was not guided by the long history of research on the subject, which began with Boettcher and ended mysteriously around 1950. In fact, Marshall was unaware of this
history—medical education in the 1970s, like medical education in the year 2000, did not waste time making medical students learn about the mistaken theories of the past.
Like Boettcher over a century before, Australian pathologists during the late 1970s noticed the curved bacteria in tissue samples from ulcer patients. Like Connie Guion over three decades before, Marshall in 1981 saw his patients with ulcers and gastritis improve after tetracycline treatment. He got the attention of microbiologists and histologists in his hospital who were needed to resolve the matter. They found the curved bacilli in all their duodenal ulcer patients and in 80 percent of their peptic ulcer patients. In April 1982 they cultured the bacterium that caused the trouble, now known as Helicobacter pylori.
Infectious causation of peptic and duodenal ulcers is now widely accepted by medical authorities, though this acceptance has grown only gradually since the mid-1980s. Marshall helped the process along by intentionally drinking an infective dose of H. pylori, getting gastritis, and then curing it with an antibiotic. Still, it was only in the mid-1990s that the medical establishment finally generally accepted the idea that peptic and duodenal ulcers are infectious diseases.
It is not clear why almost all medical experts ignored for four decades the evidence that ulcers could be caused by infection. Several attributes of ulcers made infectious causation inconspicuous: a loose correlation between infection and ulcers, the internal site of infection, and variable delays between the onset of infection and the onset of overt disease. Partly as a result of these attributes, evidence of infectious causation could be overlooked or dismissed. H. pylori is so cryptic in its disease causation that its transmission mode is still unclear today.
Another reason for the delay in solving the ulcer puzzle stems more from psychology than medicine. Some researchers reported in the late 1940s that they could not find the curved bacteria in fresh specimens; they attributed the associations that were found by others to bacteria that were growing in the ulcers after death. Perhaps this kind of negative evidence was all that was needed in a time when medical thinking was influenced by Freudianism, which emphasized the power of mind to influence disease and the fragility of that powerful mind in response to stressful influences of the social environment. If the mind, altered by the stress of
twentieth-century life, could make a person mentally self-destruct, certainly it could cause a little old ulcer.
Besides, infectious causation was beginning to be old hat; genetics had become the biological answer to disease, especially after the structure of DNA was discovered in 1953. This discovery would eventually allow an understanding of what went wrong brick by brick, from the nucleic acid bricks that were used to build the DNA and RNA, to the amino acid bricks that were used to build the proteins the DNA and RNA encoded. Using this metaphor, genetically minded medical researchers thought that they could understand why biological buildings had problems and eventually solve those problems. The history of the subsequent half century shows that this general approach was often correct, though sometimes misleading. Genes that cause human
disease were found, but often those genes were not human; viral genes were found inside cells, sometimes even integrated into human DNA. The structures these genes encoded were not buildings at all, but tanks, missiles, and sinister devices of sabotage and manipulation, such as the E6 and E7 proteins of human papillomaviruses.
Still another reason for the slow acceptance of infectious causation of ulcers stems from the standards of evidence used in medical science. Simply, the standards of evidence were too high. When standards of evidence are set too high, scientific rigor declines. It is easy to recognize this problem when the standards are set so high that they can never be met. If, for example, scientists demanded experimental demonstration of humans evolving from ancestral apes before accepting the hypothesis that humans evolved from apes, the hypothesis could never be accepted. This
too-high standard of evidence would have sapped the rigor of scientific inquiry into human origins. The same logic applies when the standards are extremely difficult but not impossible to reach, though the trade-offs between standards of evidence and scientific rigor is more difficult to assess. The required shift in standards seems especially difficult for researchers to accept when the standard has been reached for somewhat similar hypotheses. This is the state of affairs in studies of infectious causation.
Standards for identifying infectious causation had been established during the 1880s when infectious causes of acute diseases were being rigorously identified at a remarkable pace. But the early success of these standards led health scientists to overlook the importance of a simple fact: diseases caused by infection will vary greatly in the degree to which their infectious causation can be demonstrated by any particular set of standards. Applying one set of standards will allow early recognition of diseases that are easily recognized by the standards, and leave in the wake of that recognition those diseases that cannot be so identified. Throughout most of the twentieth century the experts maintained a stalwart grip on has-been standards that had largely outlived their usefulness.
*37\225\2*