The inheritance of liability to diseases with variable age of onset, with particular reference to diabetes mellitus
The inheritance of liability to diseases with variable age of onset, with particular reference to diabetes mellitus
复制标题
不同发病年龄的疾病的遗传倾向，特别是糖尿病
DOI：
--
复制
发表时间：
1967
影响因子：
1.9
通讯作者：
D. Falconer
中科院分区：
文献类型：
--
作者：
D. Falconer
在以前的一篇论文（法尔科纳，1965）中，我提出了一种多因子遗传疾病的遗传分析方法，这种方法可以估计疾病易感性的遗传力。本文将首先指出一个错误的假设，使该方法不可靠时，适用于双胞胎。然而，本文的主要目的是表明，一个限制条件，在前面的文件是不必要的。在删除这一条件后，将以比以前对发病年龄可变和发病率随年龄增加的疾病进行的分析更充分的方式进行分析。然后将分析应用于已发表的糖尿病数据。前一篇文章中所描述的方法所依据的基本原理非常简单，如下所示。所有的原因，遗传和环境，使一个人或多或少可能发展的疾病可以结合成一个单一的措施称为个人的“责任”。人口中个人的负债构成一个连续变量。受影响和正常之间的明显不连续性来自于某种程度的责任的“门槛”。负债高于门槛值的个人受到影响，负债低于门槛值的个人则不受影响。在实践中，患病和正常之间的区别必须基于诊断。因此，影响一个人被诊断出的疾病发展阶段的因素是他的责任的一部分，并导致个人之间的责任差异。个人的责任无法衡量，但人口或群体的平均责任可以从该人口或群体的疾病发生率中评估。为此，必须假定，负债原则上可以在一个使其分布正常的尺度上以单位表示。平均负债的计量单位就是这个尺度上的标准差。该分析提供了一个估计亲属之间的相关性方面的责任，这导致，与某些假设，以估计遗传的责任在人口中。遗传力是归因于加性遗传方差的责任方差的比例。这是人类数据所能达到的最接近基因决定责任的程度。分析所需的数据是人群中疾病的发病率和从人群中抽取的受影响先证者亲属的发病率。在前一篇文章中使用的术语“发病率”的含义是在流行病学中将被称为“患病率”，为了保持一致性，我将在本文中继续使用“发病率”。我所说的“发病率”是指受影响个体的频率;即在人群中或亲属中发现受影响的活体个体的比例。特定年龄组的发病率是指该年龄组中患有该疾病的人的比例，而不论发病年龄。
In a previous paper (Falconer, 1965) I proposed a method of genetic analysis for diseases with multifactorial inheritance which leads to an estimate of the heritability of liability to the disease. The present paper will first point out an erroneous assumption that makes the method unreliable when applied to twins. The main purpose of the paper, however, is to show that one restrictive condition made in the previous paper is unnecessary. With this condition removed, the analysis will be developed in a more adequate manner than was done previously for diseases with variable age of onset and incidence increasing with age. The analysis will then be applied to published data on diabetes mellitus. The premisses on which the method described in the previous paper is based are, very briefly, as follows. All the causes, both genetic and environmental, that make an individual more or less likely to develop the disease can be combined into a single measure called the individual’s ‘liability’. The liabilities of individuals in a population form a continuous variable. The apparent discontinuity between affected and normal arises from a ‘threshold’ at some level of liability. Individuals with liabilities above the threshold are affected and individuals with liabilities below it are not. In practice the distinction between affected and normal must be based on diagnosis. Therefore the factors that influence the stage in the development of a disease at which an individual is diagnosed are part of his liability and contribute to the variation of the liability among individuals. The liability of an individual cannot be measured, but the mean liability of the population or group can be evaluated from the incidence of the disease in that population or group. For this it must be assumed that liability can in principle be expressed in units on a scale that renders its distribution normal. The units of measurement of the mean liability are then standard deviations on this scale. The analysis provides an estimate of the correlation between relatives in respect of liability, and this leads, with certain assumptions, to an estimate of the heritability of liability in the population. The heritability is the proportion of the variance of liability that is ascribable to additive genetic variance. This is the nearest one can get with the human data to the degree of genetic determination of liability. Thedata required for the analysis are the incidence of the disease in the population and the incidence in relatives of affected propositi drawn from the population. The term ‘incidence’ was used in the previous paper with the meaning that in epidemiology would be termed ‘prevalence’, and for the sake of consistency I shall continue to use ‘incidence’ in this paper. By ‘incidence’ I mean the frequency of affected individuals; i.e. the proportion of living individuals found to be affected in the population, or among the relatives. The incidence in a particular age-group is the proportion of people in that age-group who have the disease, irrespective of the age of onset.