Louis, P. C. A.
Louis, P. C. A.
Pierre Charles Alexandre Louis (1787–1872), who did much to introduce the use of statistics into medicine, was born the son of a vineyard proprietor in the small town of Ay (Marne) in Champagne. Although his father died when Louis was six, his mother saw to her son’s primary education. After his schooling was over he was sent to a law office to prepare for a legal career. In 1807, deciding that the law was not to his liking, Louis began to study medicine. He spent a year at Reims with a surgeon and then went to Paris to pursue his studies, graduating in 1813 with a medical degree.
A chance encounter during a brief vacation in his native town had an important effect on his career. The comte de Saint-Priest, an emigré noble in whose family Louis’s aunt had been a governess, was friendly with the Louis family and paid them a visit. As governor of Podolia, Saint-Priest was in the service of the tsar and persuaded Louis to accompany him to Russia. After traveling about that country for three years Louis settled in Odessa, where he acquired a substantial practice and even received a titular appointment as physician to the tsar. In 1820 Odessa experienced a diphtheria epidemic, an event which led Louis not only to realize the shortcomings in his knowledge of disease but to give up his practice and return to Paris for further study.
Six months in the Paris hospitals convinced Louis that clinical medicine required a more precise basis than it had and that this could be achieved by what he called the numerical method. A.-F. Chomel, his friend and fellow student, was attending physician at the Charite Hospital and gave Louis the run of two wards of his service, as well as the privilege of performing all the autopsies on the patients who died there. For six years Louis worked at the hospital from three to five hours a day, devoting at least two hours to each autopsy and collecting over two thousand observations. In 1827 he retired to Brussels, where the cost of living as lower, and spent a year tabulating and analyzing his statistical data.
Some of his observations had been published while he was still in Paris, and in 1825 Louis had brought out his Recherches anatomico-pathologiques sur la phthisie, based on 123 cases. (Some eighty were added to the second edition of 1843.) By the time he returned to Paris in 1828 he had acquired a reputation in medical ircles. That year he was sent, together with Armand Trousseau and icolas Chervin, to investigate a yellow fever epidemic at Gibraltar. In 1828 his work on typhoid fever was published and he became attending physician at La Pitié Hospital. Subsequently he was also appointed to the Hôtel-Dieu; he served in these institutions for many years. Louis’s last important publication was his critical analysis of the alleged therapeutic effects of bloodletting, Recherches sur les effets de la saignée dans quelques maladies inflammatoires (1835). In this book he employed the numerical method to refute the views of Francois Broussais on copious bloodletting in pneumonia and other inflammatory diseases.
Louis married a daughter of the marquis de Montferrier in 1832, when he was 45. For some twenty years he carried on an ample consultation practice in Paris. His only son, Armand, died of tuberculosis in 1854, a loss from which Louis never recovered.
The numerical method. Louis’s major contribution resides in his efforts to apply statistical analysis to problems of clinical medicine. His advocacy and use of the numerical method served this end. What Louis did was to study each patient as thoroughly as possible at the bedside and at autopsy, employing rigorously the methodology already established by G. L. Bayle, Rene Th. H. Laennec, and others of the Paris school of clinicians and path-ologists. Having carefully collected his observations he then grouped them in tabular form. From these grouped and tabulated data inferences might be drawn concerning the relations between diverse clinical phenomena, the probability of their occurrence, the value of a given therapy, and other items. Although he had used his approach in his earlier publications on phthisis and typhoid fever, the numerical method was first fully presented in 1835 in Louis’s therapeutic study, Recherches sur les effets de la saignée. .. , and in a special memoir De I’examen des malades et de la recherche des faits generaux (1837).
Essentially, Louis’s numerical method was not new. The procedure had been employed some three decades earlier by Philippe Pinel to prove the value of his “moral treatment” of mental patients. It was also being employed in the 1820s and 1830s by physicians concerned with such public health problems as the causes of differential mortality and the effect on health of such factors as economic and social class, occupation, race, imprisonment, intemperance, or lack of proper sanitation. Further-more, Louis’s handling of numerical data was basically simple. If, as it is said, Louis was familiar with the work of Laplace on probability, there is no evidence that he ever used such knowledge in his statistical thinking. Like so many of his contemporaries he dealt with small numbers of observations and had no knowledge of how to establish the precision or validity of his results. There is no doubt that Louis was himself conscious of these difficulties; et he did not seek statistical criteria of reliability, nor did he try to decide when the number of observations was large enough to avoid error.
Nevertheless, Louis has a significant place in the evolving application of statistical analysis to health problems. First of all, he recognized the basic importance of accurate observations. His insistence on good clinical records established a fundamental principle for statistical work in clinical medicine. Second, Louis made an important contribution as a teacher and a propagandist. Through his students, of whom a considerable pro-portion were foreigners (among them Americans), through the Société Médicale d’Observation (which several of his students founded in 1832), and through his writings Louis advocated and spread the idea of the numerical method, in spite of vigorous opposition. He envisaged the goal of a science of clinical medicine and pointed to the road that would lead to it.
(1825) 1843 Recherches anatomico-pathologiques sur la phthisic. 2d ed. Paris: Bailliere.
(1828) 1841 Recherches anatomiques, pathologiques ... sur la maladie connue sous les noms de fièvre typhoïde ... 2 vols., 2d ed. Paris: Baillière.
1835 Recherches sur les effets de la saignée dans quelques maladies inftammatoires. Paris: Baillière.
1837 De Texamen des malades et de la recherche des faits généraux. Société Médicale d’Observation, Mémoires 1:1-63.
Astruc, Pierre 1932 Le centenaire de la Société “Me”dicale d’Observation. Progrés médical, supplément illustré9:73-87.
BÉclard, Jules-auguste 1878 Notices et portraits: Éloges lus à I’Académie de Medecine. Paris: Masson. → See especially pages 228-257.
Greenwood, Major (the younger) 1936 The Medical Dictator and Other Biographical Studies. London: Williams & Norgate. → See especially pages 123-142.
Osler, William 1908 An Alabama Student and Other Biographical Essays. Oxford Univ. Press. → See especially pages 189-210.
Rosen, George 1955 Problems in the Application of Statistical Analysis to Questions of Health: 1700–1800 Bulletin of the History of Medicine 29:27-45.
Woillez, EugÈne J. 1873 Le docteur Pierre Charles Alexandre Louis: Sa vie— ses oeuvres (1787–1872). Paris: Dupont.