New York Medical College - Fleuroscope Yearbook (Valhalla, NY)

Page 198 text:

HEAALTI-IY SKEPTICISM BY JOHN ABAJIAN, 112. THE purpose of this paper is to consider briefly some technical difficulties which are inherent in investigations pertaining to medicine and its allied sciences. One is profoundly impressed with the deluge of literature concern- ing scientific investigations in all branches of medicine. On closer examina- tion of various articles in different journals, apparently diametrically opposed results and conclusons, pertaining to the same subject, often leaves one in Wonderment as to the accuracy of these different papers and conclusions. Therefore, in order that one may avoid the acceptance of scientifically unsound results and theories in his ordinary routine of medical reading, it is necessary to acquaint oneself with some of the loopholes and fundamental Weaknesses which are present in a goodly proportion of the medical writings of today. It is essential to realize that we can seldom attain the simplification of rela- tions which is possible to the physicist or the pure chemist, who has his ma- terial so well under control that he can carry out a crucial experiment so that with proper care his particular results can be repeated at the next trial either by himself or by some other Worker. However, in the fields of experimental physiology, biochemistry, infection immunity and the like such precision is be- yond our reach: and the only hope of immediate progress is to realize the limitations under which our investigations are carried out, and to make due allowances for the various sources of error. The first of these technical difficulties to be discussed is the consideration of just exactly how quantitative our methods of quantitative analysis of blood, urine, body metabolism etc. are? In consideration of this point one must realize that the personnel of the average hospital laboratory is made up of technicians who are forced to do not only all descriptions of Work, from serology to fine micro-analytical determinations, but also work on a basis of mass pro- duction. With all due respect to the excellent and conscientious laboratory technicians, they are as a rule without training in the purely theoretical as- pects of quantitative chemistry, which of course makes them more susceptible to the pitfalls of quantitative analysis. Another point to consider is the usual type or method of analysis, this being the colorimetric method which is more of- ten not specific in the chemical sense and open to greater limits of error. The above points are sufficient for thought, to say nothing of the much too frequent absence of routine duplicate analysis. The Writer recalls one instance in which the same specimen of blood was sent to three different hospitals labora- tories as a check on present day methods of routine cholesterol determinations and was only mildly surprised to find as much as 25 to 5O'hv differences in the reports. Another glaring error in this field is the usage of B. M. R. reports based on methods which utilize only oxygen consumption, and the assump- tion of a normal respiratory quotient of .82 for all patients. This may account for the many discrepancies between the clinical pictures and the B. M. R. One hundred ninety nine

Page 197 text:

UNFINISHED sYMPHoNY BY WILLIAM L. EINHORN FOREW ORD YOU doctors of such vast experience would do well to reflect back upon the visit to your first patient. Perhaps then you will realize that, on the whole, your actions were quite' similar to those of this particular student . . . entirely excusable and attributable to a high degree of exhaltation associated with the realization of a goal within reach .... , Exactly seven days had passed since this student had completed the di- dactic part of his course in Physical Diagnosis . . . Seven-thirty, A. M. finds him freezing in the waiting room of the Municipal Ferry . . . decides it would be rather unprofessional to visit his first patient that early, so endeavors to make himself comfortable . . . He realizes that for the first time in some twenty-odd years, he has no books about his person. CED.-NOTE-Uribe lievablel which fact tends to make him disregard the hard bench . . . Glancing down at his overcoat pocket he notices the bright glare of stethoscope ear- pieces, so carefully fixed that but a wee bit protrudes, but enough, mind you, so that there can be no mistake as to the contents of that pocket. This fact ten-ds to make him disregard the cold . . . Carefully, he pulls his jacket to one side in order to make certain that thermometer and flashlight are well to the mesial side of his upper right vest pocket .... He rises, places his hand into his trouser pocket, and being assured that this maneuver will positively reveal the T and F, he buttons his coat and sits down ..,. by this time he closely ap- proximates satisfaction ....' N 'ow he decides to quickly scan over the ques- tions he was told to ask his patient CED.-NOTE-what's the use, he'll never re- member theml .... Finishing this, he has about three or four cigarettes, and then boards the ferry .... Arriving at the Island he takes a few deliberate steps towards the nearest building but stops short, removes his hat, an-d scatch- es his head, for it has just dawned upon him that the particular destination he had decided upon was only one of many buildings of a similiar kind .... Pushing the stethoscope way down and out of sight, he turns up his coat collar and walks back to the ticket collector KNOTE-he's playing the part of an or- dinary visitor now-how embarrassingl, and inquires as to the whereabouts of the ward he had been assigned to .... Having been satisfactorily directed, he turns down the collar of his coat, brings up the stethoscope and continues on his way .... If you are interested-any third year student is qualified to further en- lighten you. One hundred ninety-eight

Page 199 text:

reports. This difficulty may be surmounted by utilization of methods which involve not only oxygen consumption but the carbon dioxide output: thus de- termining the actual R. Q. for each individual. As a point of constructive criticism more specialization among hospital laboratory technicians, more duplicate determinations and more frequent checks on our methods, is rec- ommended. A second source of error lies in the word statistics. Sir Francis Galton ill some 50 years ago expressed a very apt thought which can be best expressed in his actual Words. Some people hate the very name of statistics, but I find them full of beauty and interest. Whenever they are not brutalized, but delicately han- dled by the higher methods, and are warily interpreted, their power of dealing with complicated phenomena is extraordinary. They are the only tools by which an opening can be cut through the formidable thicket of difficulties that bar the path of those who persue the Science of Man. Thus when one makes the statement- There are liars, damn liars and statistics, the capacity of that individuals to deal with his material should be questioned and not the statistical method itself . To the clinical investigator and the laboratory worker who constantly deal with variables affected by a multiplicity of factors, a knowledge and understanding of certain fundamental statistical procedures is invaluable. It may be not amiss here to include a few notes which will enable the reader to realize the order'of magnitude of. the errors for which he must be prepared to make allowances for, and to understand the limitations as well as the uses, of the usual statistical criteria of significance. For example, in a consideration of a form of therapy such as routine digitalization of patients with lobar pneu- monia, one may express his results as xW: death rate in c number of control cases, and WZ: death rate in t number of treated patients. The question that now arises is the difference between xfk and yfk significance. The approach is as follows: the ratio of a particular difference to the standard deviation enables us to calculate, on certain assumptions, how often this difference would appear in the two groups under comparison if they were identical, that is statistically homogenous, in all relavent characteristics, and was the result solely of chance in the statistical sense. The method of determining the stand- ard deviation may be found in many treatises dealing with statistical methods of study such as in R. A. Fisher's C29 book. Thus if our observed difference is equal to about 2X3 of its own standard deviation it will likely as not that this difference might appear, even though our experimental procedure has been without influence on the results. However, if the difference is IW times as great as the standard deviation, the odds against it being due to errors of random sampling or chance becomes about 5 to l. When the ratio increases Two hundred