National Institutes of Health, Bethesda, MD 208922092, USA.
In 1835, the Paris Academy of Science commissioned a report on the statistical research that had been conducted by the surgeon and urologist Jean Civiale (17921867). By collecting statistical data on a wide scale throughout Europe, Civiale argued that a new bloodless procedure for removing bladder stones, a lithotripsy, was superior to the more widely-used technique of surgically cutting to remove the stones known as a lithotomy.1 Although the specific therapeutic and surgical interventions that motivated the commission's report may no longer be directly relevant to the treatment of bladder stones, the broader issues that this report engaged (namely, the cultural authority of appeals to quantitative evidence) are clearly relevant to the contemporary medical world. For this reason, putting this report into a broader historical context illustrates the 19th-century antecedents to au courant debates concerning evidence-based medicine.
In order to situate the commission's report historically, it is necessary to realize that, in some respects, Civiale's work builds on research traditions that had developed in the 18th century while, in other respects, it is very culturally specific to the second quarter of the 19th century. As Ulrich Tröhler has demonstrated, the recording of successes and failures in removing bladder stones had been practiced since the beginning of the 18th century; because practitioners received more patients if they had a higher success rate, they began to list the number of their successes and failures annually in tabular form.2 In this respect, Civiale's method of quantitative comparison was not fundamentally new. However, he was able to carry out his research on a much larger scale than his 18th century predecessors because of support from the French Ministry of Public Instruction.3 This government support illustrates how the appeal to aggregate data had now come to be seen as key method of settling questions deemed to be public issues'.
By the second quarter of the 19th century, several factors had converged that gave quantitative evidence, like that espoused by Civiale, an increasingly high level of cultural cachet. This era in European history was characterized by a marked interest in the collection of numerical information about societywhat the philosopher of science Ian Hacking has called the avalanche of printed numbers'.4 Based on the work of reform-minded physicians and political economists, European nations now had at their disposal vast amounts of aggregate data about their citizenry. Also, the political upheavals associated with the French Revolution in the last decade of the 18th century had been the crucible in which modern clinical medicine was formed.5 In Paris, the new hospital-based methods of instruction placed an emphasis on clinical observation, autopsy, and the use of statistical datawhat the French clinician P-C-A Louis famously described as the numerical method.6 Viewed against the backdrop of these developments, Civiale's report could be seen as another manifestation of this larger focus on quantification to adjudicate questions of medical uncertainty.
In commenting on Civiale's report, however, the Commission of the Academy of Science did more than speculate on whether his statistics were conclusive for the specific issue at hand; they used the report as an occasion to engage the general question of the proper role of statistical reasoning in deciding therapeutic questions. As the report noted, We hasten to seize this opportunity to broach the question of the application of the calculation of probabilities to medicine ... Medicine ... has all too often sought to hitch on to ideas that are fashionable in the opinion of the day. Thus, at present, statistics are constantly applied to most of the major questions in therapeutics. Yet in this case statistics are no more than an attempt at application of calculation of probabilities. Let us try to see what opinion we should form.7 Although the commission criticized applying the calculus of probabilities' to medicine for a myriad of reasons, their main concern was that the clinician (inevitably) focused on the diagnosing and on treating the individual who would present a completely unique and idiosyncratic group of symptoms. By contrast, the commissioners noted that, in statistics, the first task is to lose sight of the individual seen in isolation, to consider him only as a fraction of the species. He must be stripped of his individuality so as to eliminate anything accidental that this individuality might introduce into the issue at hand.8 It was this fundamental distinction (the individual versus the statistical aggregate) that caused the commission to question the applicability of the numerical method to medicine.
Ultimately, this report is historically significant primarily because it framed the issue of the role of quantification in medicine in ways that still resonate with contemporary medicine at the dawn of the 21st century. Like their 19th-century ancestors Civiale and P-C-A Louis, contemporary supporters of evidence-based medicine herald the use of quantitative methods as the way that medicine will finally be transformed into a science.9 By contrast, contemporary critics of this approach, like the clinician and commission reporter François Double, question the excessive reliance on such quantitative methods; they still fear that such cookbook medicine may cause physicians to lose sight of the unique and individuating features of the patients under their care.10
References
1 Poisson SD, Double FJ et al. Rapports: Recherches de Statistique sur l'affection calculeuse, par M. Le docteur Civiale. Comptes Rendus Hebdomadaires des Séances de l'Académie des Sciences 1835;1:17172.
2 Tröhler U. Quantification in British Medicine and Surgery 17501830, with Special Reference to its Introduction into Therapeutics. PhD Dissertation, University of London, 1978.
3 Lecuyer B-P. The Statistician's Role in Society: The Institutional Establishment of Statistics in France. Minerva 1987;25:5253. Civiale J. Traité de laffection calculeuse, ou recherches sur la formation, les caractères physiques et chimiques, les causes, les signes, et les effets pathologiques de la pierre et de la gravelle suives d'un Essai de statistique sur cette malade. Paris: Crochard et Comp, 1838.
4 Hacking I. Was there a probabilistic revolution 18001930? In: Krüger L, Daston LJ, Heidelberger M (eds). The Probabilistic Revolution. Vol. 1: Ideas in History. Cambridge, MA: The MIT Press, 1987.
5 Ackerknecht E. Medicine at the Paris Hospital, 17941948. Baltimore: The Johns Hopkins University Press, 1967.
6 Louis P-C-A. De lexamen des malades et de la recherche des faits généraux. Mémoires de la Société Médicale d'Observation. Vol. 1. Paris: Chochard, 1837.
7 Poisson SD, Double FJ et al. Rapports: Recherches de Statistique sur laffection calculeuse, par M. Le docteur Civiale. Comptes Rendus Hebdomadaires des Séances de lAcadémie des Sciences 1835;1:17273.
8 Poisson SD, Double FJ et al. Rapports: Recherches de Statistique sur laffection calculeuse, par M. Le docteur Civiale. Comptes Rendus Hebdomadaires des Séances de lAcadémie des Sciences 1835;1:173.
9 Eddy D. Clinical Decision Making From Theory to Practice: A Collection of Essays from The Journal of the American Medical Association. Sudbury, MA: Jones and Bartlett Publishers, 1996.
10 Liang M. From America: Cookbook medicine or food for thought: Practice guidelines development in the USA. Ann Rheum Dis 1992; 51:125758.[ISI][Medline]