WHAT IS A MORTALITY RATE?
WHAT IS A MORTALITY RATE?
复制标题
什么是死亡率？
DOI：
--
复制
发表时间：
1950
期刊：
影响因子：
--
通讯作者：
H. M. Luykx
中科院分区：
文献类型：
--
作者：
H. M. Luykx
在医学问题中，人类被认为是临床医生的领域。这是充分证明了这一事实，在疾病的研究中，对病人的第一个义务是使他们康复。临床医生接受的训练是把病人的个体需求放在首位，不能指望他们只局限于冷冰冰的科学判断。出于这个原因，他必须在实验中严重依赖统计学家或生物统计学家。大多数统计人员都受过科学或实验方法的训练。最初，这可能不被认为适用于人类，除非在某些不准确的方式。然而，随着生物统计学的现代发展，未知和不明确的因素也在数学处理中占据了一席之地，结果是统计方法可以越来越广泛地应用于生物科学，包括现在基本上是人类科学的医学。然而，有证据表明，数理统计学家在一个鼓励他们尽一切努力的领域做出宝贵贡献时，往往会忽视人类物质的一些基本局限性。当一个人与临床受试者一起工作时，在控制和定义实验观察方面显然是困难的;因此，一个人只限于最客观的现象类型。在这些因素中，死亡的事实在许多方面是最有用的，因为它很容易被生物学定义，也很容易被描述为可识别的属性，如死亡原因、死亡年龄等，因此，一个被称为死亡率的因素被广泛使用，通常被定义为死亡人数与存活人口的比率。(That这一定义首先是误导性的，我们将在下面讨论。）数学家，提出了一个比例，很容易探索的关系，分子和分母的理论，误差，抽样统计，分布函数等，不够重视，但是，更基本的属性的死亡率，而不是数学属性，但临床属性，因为它是。也许我们这次可以指出一些关于这个问题的问题，这些问题往往没有得到充分的理解。
In medical problems, the human population is considered to be the province of the clinician. This is amply justified by the fact that, in the study of disease, the first obligation to the diseased is to make them well. Being trained to consider the requirements of the individual patient as paramount, the clinician cannot be expected to confine himself exclusively to cold scientific judgment. For this reason, he must rely heavily on the statistician or biometrician in his experimentation. Most statisticians are trained in the scientific or experimental approach. Initially, this might not have been considered to be applicable to the human population, except in certain inexact ways. With the modern development of biometry, however, unknown and ill-defined factors also have assumed their places in mathematical treatment, with the result that statistical methods can be applied more and more extensively in biological sciences, including now that essentially human science, medicine. There is evidence, however, that mathematical statisticians, in making their invaluable contributions in a field where their every effort is to be encouraged, have a tendency to overlook some of the fundamental limitations of human material. When one works with clinical subjects, there is obvious difficulty in controlling and defining experimental observations; hence, one is limited to only the most objective type of phenomena. Of these, the fact of death is in many ways the most useful, as it lends itself readily to biological definition and to description in terms of recognizable attributes such as cause of death, age a t death, and so on. Thus, there has resulted the wide usage of a factor called the mortality rate, commonly defined as the ratio of deaths to the living population. (That this definition is misleading in the first place will be touched upon presently.) The mathematician, being presented with a proportion, has easily explored the relationships of numerator and denominator in the light of the theory of errors, sampling statistics, distribution functions, etc. Insufficient attention has been paid, however, to the more fundamental attributes of a mortality rate, not the mathematical properties, but the clinical properties, as it were. Perhaps we can point out, a t this time, some questions concerning the subject which too often are not adequately understood.