The Desperate Pre-Antibiotic Era
Before 1947, a diagnosis of progressive pulmonary tuberculosis was a grim prospect. The disease, caused by the bacterium Mycobacterium tuberculosis, was a leading cause of death for young adults in Europe. Treatment consisted of admission to a sanatorium, a facility dedicated to providing rest, fresh air, and a nutritious diet. These institutions were built prolifically across Europe between 1900 and 1950. Some surgical interventions like artificial pneumothorax, which involved collapsing a lung, were attempted, but their effectiveness was uncertain and carried substantial risks. Other historical treatments were entirely ineffective, including cod liver oil and vinegar massages.
The discovery of penicillin in 1928 offered hope for bacterial infections, but it had no effect on Mycobacterium tuberculosis. The scientific community urgently sought an effective antibiotic. In 1943, Albert Schatz, a PhD student in Selman Waksman's lab at Rutgers University, isolated streptomycin from a soil fungus. Early tests on animals and then humans showed promise, but supplies of the drug were extremely limited and expensive. This scarcity created an ethical and scientific dilemma: how to determine if the drug truly worked without wasting a single dose on a method that might produce ambiguous results. The medical world needed a new, rigorous way to test treatments.
A Revolution in Method
The answer came from a study design that became the foundation of modern medical research: the randomized controlled trial (RCT). The first such trial for a curative treatment was organized by the British Medical Research Council (MRC) in 1946 to test streptomycin. The trial's design, spearheaded by statistician Sir Austin Bradford Hill, was revolutionary in its simplicity and rigor. Patients with "acute progressive bilateral pulmonary tuberculosis" between the ages of 15 and 30 were selected.
The core of the method was random allocation. To prevent selection bias by doctors, patients were assigned to one of two groups using a table of random numbers. The details were kept in sealed, numbered envelopes, so no investigator knew which treatment the next patient would receive. One group received streptomycin plus the standard treatment of bed rest. The control group received only bed rest. This comparison to the existing standard of care was an important feature. To prevent assessment bias, the chest X-rays were read by experts who were "blind" to which group each patient belonged. The results, published in the British Medical Journal in 1948, were clear. In the first six months, only 4 of the 55 patients in the streptomycin group died, compared to 15 of the 52 patients in the control group. This trial established a new standard for evidence-based medicine, and its methodology is still used today in clinical trials for nearly every new medical treatment.