93

H.—l

1932. You have told us that if you had been asked when she came into the Hospital how long she was likely to stay there, you would hare answered ton days, but that you are not always correct in your diagnosis. Do you think that a mistake in a calculation of that kind will at all account for the enormous difference in the time you thought Mrs. T would be in the Hospital and the time she has actually been there ?—Certainly not. 1933. The Chairman.] Do you think it is attributable to natural causes '?—No. 1934. Mr. Solomon.] In trying the other day to account for the unusually high death-rate in Dunedin Hospital, it was suggested that it might be accounted for by the large number of cancer cases that occurred in the Dunedin Hospital. Have you made a comparative table of these cases in the Dunedin and Auckland Hospitals ?—I have. Mr. Chapman has attempted to show that the proportion of deaths in the Dunedin Hospital may be due partly to the heavy mortality from cancer, in Auckland there being four deaths in thirty-two cases, while in Dunedin we have eleven deaths in twenty-seven cases. Now, every medical man knows that there are a large number of cases of diseases of the digestive organs where the exact diagnosis is extremely difficult, and one man might enter a case as being due to a death from ulcer of the stomach, or some such condition, whereas another would diagnose the case as one of cancer. Now, for comparison, we will add together the deaths from the diseases of the digestive organs and deaths from cancer in Auckland and Dunedin:— Auckland. Dunedin. Cancer ... ... ... 32, with 4 deaths ... 27 with 11 deaths. Digestive system ... ... 6.1, with 12 deaths ... 83 ~ 3 ~ 93 „ 16 deaths ... 110 „ 19 deaths. = 1 death in 513 cases. = 1 death in 515 cases. If Mr. Chapman likes to telegraph to Auckland to the effect that the medical men in Dunedin consider the Auckland practitioners fail to diagnose r their cancer cases correctly, he will find that they will put a different interpretation on the value of his statistical calculations. My conclusion as to the usefulness of these statistics, when they are taken to pieces, is that they are practically valueless. 1935. Have you any authority for that ? —-Yes. Parkes, in Iris work on hygiene, in chapter 20, says : " The elements of statistical inquiries are individual facts, or so-called numerical units, which having to be put together or classed must have precise, definite, and constant characters. For example, if a number of cases of a certain disease are to be assembled in one group with a definite signification, it is indispensible that each of these cases should be what it purports to be, an unit not only of a definite character, but of the same character as the other units. In other words, an accurate diagnosis of disease is essential, or statistical analysis can only produce error. If the numerical units are not precise and comparable it is better not to use them. A great responsibility rests on those who send in inaccurate statistical tables of disease, for it must be remembered that the statist does not attempt to determine if his units are correct. He simply accepts them, and it is only if the results he brings out are different from prior results that he begins to suspect inaccuracy." In reference to this mortality question, the only part to which I attribute any real importance, or from which any deductions of value can be drawn, is the death proportions to the cases of admission. In the return of the results of operations in the Dunedin Hospital, it will be found, I think, that a large proportion of deaths are entered under the terms of " shock or collapse " following operation. Now to these cases I consider the gravest suspicion attaches, several of them occurring days after an operation —after an interval of reaction and febrile disturbance. Ericbsen, in speaking of mortality after amputation, points out how frequent it is for cases to be entered under these heads while the real cause of death will be found due to septic causes. Wherever a death occurs more than twenty-four hours after operation, especially where there has been a rise in temperature, septic influences will probably come into play. 1936. Do you hold any office in connection with the Medical School of the University '?—Yes, I am Lecturer on Midwifery and Gynecology. 1937. Mr, Solomon :] (To the Chairman:) I understand you wish to ask Dr. Batchelor a question. The Chairman : I think that was all I wanted to ask you. I did not mean to open up any further matter. Witness ; All I wish to say, before 1 leave the box, is that I am sorry if I have been rather warm at times ; but anybody who feels as strongly on the matter as I do could hardly help feeling stirred up and using warm expressions. Dr. H. Lindo Fekguson sworn and examined. 1938. Mr. Solo?non.] What is your name?— Henry Lindo Ferguson. 1939. And your full professional title?— Fellow of the Eoyal College of Surgeons (Ireland); Licenciate of the King's and Queen's College (Ireland). 1940. You are a registered practitioner in New Zealand also ?—Yes. ] 941. How long have you been in New Zealand ?—Since the latter part of 1883—October, 1883, I think I came here. 1942. And you have beerua member of the surgical staff?—l have never been on the staff pure and simple, but I have been specially on the staff for diseases of the eye since December. 1883. 1943. Shortly after your arrival? —Yes, very^shortly. 1944. And you have remained in that position until now? —Yes.