In lieu of an abstract, here is a brief excerpt of the content:

c h a p t e r t h r e e Fundamental Tensions Clinical “Proof” and Clinical Resistance The treatment of pneumonia—on account of the disease’s marked variability in severity from patient to patient and from year to year—had long provided a challenging yet alluring subject for those concerned with the methodology of “proving ” therapeutic eªcacy. Pierre Louis, for example, had in the 1830s used “pneumonitis ” as the principal object of his “numerical method” (comparing the mortality rates of patients divided into unequal treatment groups) in challenging the eªcacy of bloodletting.1 By the era of serotherapy, the disease retained its enigmatic variability. Such variability, by the turn of the twentieth century, was felt to “render general statistics wholly unreliable in determining the eªcacy of any plan of treatment” if such statistics were based on either small case reports or the use of historical controls.2 Yet by the end of World War I the concentration of pneumonia patients in hospitals in America provided clinical researchers with a unique opportunity. It soon became apparent to a core group of researchers that the experimental immunological science underlying serotherapy should be accompanied by an equally reproducible clinical science capable of separating the inherent clinical variability of pneumonia from the e¤ects of therapy. Controlled clinical pneumonia studies would hence serve at the vanguard of a gradually changing approach to therapeutic evaluation; and the treatment of pneumonia would likewise serve as a litmus test regarding clinicians’ evolving “therapeutic perspectives” themselves. Antipneumococcal Serotherapy and the Propagation of the Controlled Clinical Trial The degree to which studies of antipneumococcal serotherapy—as well as the general interwar development of the alternate controlled trial—have been ig- nored in most recountings of the evolving clinical trial is surprising. The conventional account begins with biblical references to comparative studies, proceeds through James Lind’s eighteenth-century evaluation of scurvy remedies, and advances through such nineteenth-century applications of the numerical method as Louis’s researches and Ignaz Semmelweis’s studies of the prevention of puerperal fever.3 Critically, it next fixes on Johannes Fibiger’s use of alternate controls (treating patients admitted on alternate days di¤erently, to be precise) in establishing the eªcacy of diphtheria antitoxin in 1898, as well as the contemporary e¤orts of Karl Pearson in Britain to establish biometry as a medical science capable of rendering reproducible statistical distinctions between treatment groups.4 From Fibiger and the early biometricians, the story next moves to the advocacy of “randomization” by Ronald Fisher for agricultural experiments in the 1920s (i.e., through using random strips of land to allocate treatment versus control groups, local environmental influences would be eliminated); to the popularization of such techniques for clinical medicine through Fisher and A. Bradford Hill in the 1940s (i.e., through the selection of treatment versus control groups on a random basis, the bias incurred through selecting patients to treat would be eliminated); and, ultimately, to Hill’s subsequent combination of randomization with “blinded” evaluation of clinical outcome (i.e., ignorance on the part of the assessors to which therapeutic group a particular patient had been assigned ) in the British Medical Research Council study of streptomycin for tuberculosis in 1948. In moving from Fibiger to Hill, linkages are generally depicted as stuttering and fragmentary,5 or even nonexistent.6 The rise during the intervening era of the structural elements that would make possible the advent of large-scale randomized controlled trials has been the subject of more sophisticated recent analyses . These include Harry Dowling’s and Harry Marks’s depictions of the Cooperative Clinical Group study of syphilis in the late 1920s and early 1930s as the origin of multicenter collaborative e¤orts in America, Nick Rasmussen’s investigation of the evolving American clinician-industry arrangements in the conduct of trials, Desiree Cox-Maksimov’s relating of the rise of state “machinery” and legitimization for the running of large-scale trials in Britain, and J. Rosser Matthews’ focusing of attention on the rise of statisticians (such as Raymond Pearl in the United States and Major Greenwood in Great Britain) and their attempts at establishing a professional stronghold in medical research.7 Such accounts , however, still miss critical elements contributed and embodied by the intense evaluation of antipneumococcal serotherapy in America. When serotherapy is mentioned at all in the above accounts, it is done so dismissively at best,8 in36 Serotherapy, 1891–1930 [18.119.253.93] Project...

Share