“From the theoretical side, it has often been shown irrefutably to us practical physicians that all our conclusions about advantages and disadvantages of individual treatment methods, provided that they are based on the statistics of the actual success, are completely vague as long as we do not apply the strict rules of probability calculus. In fact, if the observed outcomes have been more favourable in one treatment than in another, this may just as well be based on chance as it is undoubtedly based on chance if, in a game of hazard with equal chances, one person wins today and the other tomorrow. The simple fact that 20 of 100 patients died with one treatment method and 10 of 100 patients died with the other one, does not in itself prove that the second treatment method deserves to be preferred, and does not give any certainty that perhaps 30 of 100 will die next time this second treatment method is used. If we want to draw conclusions from the actual successes, the inevitable prerequisite to do so is to examine how large the probability is that the observed differences in success are not simply due to chance. And for this question only probability calculus gives the necessary indication.”
Translation of whole article from German to English by Leonhard and Valentina Held.
Download the full article as a PDF
Liebermeister C (1877). Ueber Wahrscheinlichkeitsrechnung in Anwendung auf therapeutische Statistik [On the calculus of probabilities applied to therapeutic statistics]. In: Volkmann R (ed), Sammlung Klinischer Vortrȁge, Innere Medicin, No 39, pp 935-962.
Translation from German to English by Leonhard and Valentina Held
Whole article typeset by Leonhard Held
Carl von Liebermeister (1833-1901)