Turtles all the way down: do biological mechanisms for epidemiological observations always matter?
Turtles all the way down: do biological mechanisms for epidemiological observations always matter?
复制标题
一直以下乌龟：流行病学观察的生物学机制总是很重要吗？
DOI：
10.1007/s10654-021-00807-8
复制
发表时间：
2021-09
影响因子：
13.6
通讯作者：
van Roosmalen J
中科院分区：
文献类型：
--
作者：
Buis DTP;van Roosmalen J
生物学机制在流行病学研究报告中无处不在。许多研究人员广泛阐述了他们的研究结果背后的生物学潜在机制，假设一个人的存在与他们的工作质量有关。这一有趣的现象与著名流行病学家奥斯汀·布拉德福德·希尔的工作相呼应，他创造了希尔在流行病学研究中判断因果关系的标准。用希尔自己的话来说，它们是所有这些标准中的标准，“我们应该在哭泣因果关系之前研究关联”。它们包括强度、一致性、特异性、时间性、生物梯度、可解释性、连贯性、实验性和类比性。希尔的所有标准几乎从未被系统地用于判断关联是否代表因果效应。然而，希尔的因果关系标准仍然经常被用来加强或削弱流行病学研究中的因果关系。Höfler将这一标准表述为“观察到的关联可以通过实质性物质（例如生物学）解释来合理解释”[2]。“机制”一词的确切定义是哲学争论的主题。在这种观点中，我们将行为的机制定义为Glennan [3]提出的“通过许多部分的相互作用产生该行为的复杂系统”。为了论证的目的，我们对发生在个体生物体内的“生物机制”与发生在群体中的“社会”机制进行了某种程度上的武断区分[4]。在阅读流行病学报告时，人们不禁感到流行病学家高度重视这些生物学机制。为什么会这样，它总是合理的吗？
Biological mechanisms are ubiquitous in reports of epidemiological research. Many researchers extensively elaborate on a biological potential mechanism underlying their findings under the assumption that the presence of one is somehow related to the quality of their work. This interesting phenomenon echoes the work of Austin Bradford Hill, the famous epidemiologist who coined the Hill’s criteria for judgement of causality in epidemiological research. In Hill’s own words they are criteria from all of which “we should study association before we cry causation”[1]. They include strength, consistency, specificity, temporality, biological gradient, plausibility, coherence, experiment and analogy. All Hill’s criteria are almost never systematically used to judge whether associations represent causal effects. Hill’s plausibility criterion, however, is still frequently used to strengthen or weaken causal claims in epidemiological studies. Höfler phrased this criterion as “the observed association can be plausibly explained by substantive matter (eg biological) explanations”[2]. The exact definition of the term “mechanism” is topic of ongoing philosophical debate. In this viewpoint we define a mechanism of a behavior as “a complex system which produces that behavior by the interaction of a number of parts” as proposed by Glennan [3]. For the sake of the argument, we make a somewhat arbitrary distinction between “biological mechanisms” taking place within individual organisms, in contrast with “social” mechanisms that occur to populations [4]. While reading epidemiological reports, one cannot help feeling that epidemiologists highly value these biological mechanisms. Why is that and is it always justified?
登录
查看更多内容
影响因子：
12.7
作者：
Diamond, C;Buskin, S
通讯作者：
Buskin, S
影响因子：
13.6
作者：
Dekkers, Olaf M.;Mulder, Jesse M.
通讯作者：
Mulder, Jesse M.
影响因子：
--
作者：
Kotaska, AR
通讯作者：
Kotaska, AR
影响因子：
56.9
作者：
MAYR, E
通讯作者：
MAYR, E
影响因子：
6.3
作者：
Hern치n, MA
通讯作者：
Hern치n, MA