ABSTRACT
Resumo Este ensaio trata da questão da causalidade em epidemiologia a partir da década de 1970, cujo marco inicial aqui adotado foi a publicação de The causal thinking in health sciences, por M. Susser, até os dias de hoje, buscando elencar os vários movimentos filosóficos, teóricos e metodológicos que ao longo destes 50 anos buscaram refletir sobre o problema da causalidade na disciplina, tendo em vista o predomínio das pesquisas observacionais no campo. Partindo da contribuição seminal de Susser, foram discutidos vários movimentos, bem como as críticas a eles, tais como a proposta da adoção de lógica popperiana na década de 1980, a crítica aos modelos multicausais e a teoria ecossocial proposta por N. Krieger na década de 1990, as críticas à epidemiologia social também da década de 1990, a influência de J. Pearl e a adoção dos gráficos acíclicos direcionados como nova metodologia na questão da causalidade. A chamada revolução metodológica no início deste século e as críticas de filósofos e epidemiologistas a esta abordagem reducionista também foram revisadas, bem como as alternativas propostas nos últimos 10 anos, incluindo a perspectiva inferencialista, a triangulação de métodos e a defesa da epidemiologia social e de seus modelos de determinação.
Abstract This essay deals with the issue of causality in epidemiology from the 1970s onwards, whose starting point adopted here was the publication of The Causal Thinking in Health Sciences by M. Susser, up to the present day, seeking to list the various philosophical, theoretical and methods that throughout these 50 years have sought to reflect on the problem of causality in the discipline, in view of the predominance of observational research in the field. Starting from Susser's seminal contribution, several movements were discussed as well as their criticisms, such as the proposal to adopt Popperian logic on the 1980s, the criticism of multicausal models and the ecosocial theory proposed by N. Krieger in the 1990s, criticism of social epidemiology also in the 1990s, the influence of J.Pearl and the adoption of directed acyclic graphs as a new tool in the issue of causality. The so-called methodological revolution at the beginning of this century and the criticism of philosophers and epidemiologists to this reductionist approach were also reviewed, as well as the alternatives proposed in the last 10 years, including the inferentialist perspective, the triangulation of methods and the defense of social epidemiology and their determination models.
ABSTRACT
Detection bias is an information bias.It was first proposed by Horwitz from the study investigating the association of the administration of estrogen after menopause with the occurrence of endometrial cancer, which still prevails in most epidemiological studies.We use the Directed Acyclic Graph to analyze the effect of a given exposure on a specific outcome with the association estimates between the measured exposure and outcome.Detection bias occurs when there is additional open paths irrelevant to the target path of interest which is originated from measured exposure to measured outcome.We further analyzed how the detection bias was formed under different study designs, including cohort study, randomized clinical trial and case-control study in order to further investigate its potential influence on the effect/association estimation.
ABSTRACT
Detection bias is an information bias.It was first proposed by Horwitz from the study investigating the association of the administration of estrogen after menopause with the occurrence of endometrial cancer, which still prevails in most epidemiological studies.We use the Directed Acyclic Graph to analyze the effect of a given exposure on a specific outcome with the association estimates between the measured exposure and outcome.Detection bias occurs when there is additional open paths irrelevant to the target path of interest which is originated from measured exposure to measured outcome.We further analyzed how the detection bias was formed under different study designs, including cohort study, randomized clinical trial and case-control study in order to further investigate its potential influence on the effect/association estimation.
ABSTRACT
Evidence-based medicine (EBM) is a kind of clinic practice where clinicians use the best and the latest available evidence to diagnose and treat patients, and both evidence providers and users need to identify and control different kinds of biases in medical research.Directed acyclic graphsis is a tool to explore the causal relationship.The possible biases in the study can be revealed in a simple graphical language.The use of directed acyclic graphs could avoid the occurrence of bias and improve the quality of medical research and better guide clinical practice.
ABSTRACT
Confounding affects the causal relation among the population.Depending on whether the confounders are known,measurable or measured,they can be divided into four categories.Based on Directed Acyclic Graphs,the strategies for confounding control can be classified as (1) the broken-confounding-path method,which can be further divided into single and dual broken paths,corresponding to exposure complete intervention,restriction and stratification,(2) and the reserved-confounding-path method,which can be further divided into incomplete exposure intervention (in instrumental variable design and non-perfect random control test),mediator method and matching method.Among them,random control test,instrumental variable design or Mendelian randomized design,mediator method can meet the requirements for controlling all four types of confounders,while the restriction,stratification and matching methods are only applicable to known,measurable and measured confounders.Identifying the mechanisms of confounding control is a prerequisite for obtaining correct causal effect estimates,which will be helpful in research design.
ABSTRACT
One of the commonly accepted merits of cohort studies (CSs) refers to the exposure precedes outcome superior to other observational designs.We use Directed Acyclic Graphs to construct a causal graph among research populations under CSs.We notice that the substitution of research population in place of a susceptible one can be used for effect estimation.Its correctness depends on the outcome-free status of the substituted population and the performance of both screening and diagnosis regarding the outcomes under study at baseline.The temporal precedence of exposure over outcome occurs theoretically,despite the opposite happens in realities.Correct effect estimate is affected by both the suitability of population substitution and the validities of outcome identification and exclusion.
ABSTRACT
The overall details of causality frames in the objective world remain obscure,which poses difficulty for causality research.Based on the temporality of cause and effect,the objective world is divided into three time zones and two time points,in which the causal relationships of the variables are parsed by using Directed Acyclic Graphs (DAGs).Causal DAGs of the world (or causal web) is composed of two parts.One is basic or core to the whole DAGs,formed by the combination of any one variable originating from each time unit mentioned above.Cause effect is affected by the confounding only.The other is an internal DAGs within each time unit representing a parent-child or ancestor-descendant relationship,which exhibits a structure similar to the confounding.This paper summarizes the construction of causality frames for objective world research (causal DAGs),and clarify a structural basis for the control of the confounding in effect estimate.
ABSTRACT
In the studies of modem epidemiology,exposure in a short term cannot fully elaborate the mechanism of the development of diseases or health-related events.Thus,lights have been shed on to life course epidemiology,which studies the exposures in early life time and their effects related to the development of chronic diseases.When exploring the mechanism leading from one exposure to an outcome and its effects through other factors,due to the existence of time-variant effects,conventional statistic methods could not meet the needs of etiological analysis in life course epidemiology.This paper summarizes the dynamic path analysis model,including the model structure and significance,and its application in life course epidemiology.Meanwhile,the procedure of data processing and etiology analyzing were introduced.In conclusion,dynamic path analysis is a useful tool which can be used to better elucidate the mechanisms that underlie the etiology of chronic diseases.
ABSTRACT
Confounding affects the causal relation among the population.Depending on whether the confounders are known,measurable or measured,they can be divided into four categories.Based on Directed Acyclic Graphs,the strategies for confounding control can be classified as (1) the broken-confounding-path method,which can be further divided into single and dual broken paths,corresponding to exposure complete intervention,restriction and stratification,(2) and the reserved-confounding-path method,which can be further divided into incomplete exposure intervention (in instrumental variable design and non-perfect random control test),mediator method and matching method.Among them,random control test,instrumental variable design or Mendelian randomized design,mediator method can meet the requirements for controlling all four types of confounders,while the restriction,stratification and matching methods are only applicable to known,measurable and measured confounders.Identifying the mechanisms of confounding control is a prerequisite for obtaining correct causal effect estimates,which will be helpful in research design.
ABSTRACT
One of the commonly accepted merits of cohort studies (CSs) refers to the exposure precedes outcome superior to other observational designs.We use Directed Acyclic Graphs to construct a causal graph among research populations under CSs.We notice that the substitution of research population in place of a susceptible one can be used for effect estimation.Its correctness depends on the outcome-free status of the substituted population and the performance of both screening and diagnosis regarding the outcomes under study at baseline.The temporal precedence of exposure over outcome occurs theoretically,despite the opposite happens in realities.Correct effect estimate is affected by both the suitability of population substitution and the validities of outcome identification and exclusion.
ABSTRACT
The overall details of causality frames in the objective world remain obscure,which poses difficulty for causality research.Based on the temporality of cause and effect,the objective world is divided into three time zones and two time points,in which the causal relationships of the variables are parsed by using Directed Acyclic Graphs (DAGs).Causal DAGs of the world (or causal web) is composed of two parts.One is basic or core to the whole DAGs,formed by the combination of any one variable originating from each time unit mentioned above.Cause effect is affected by the confounding only.The other is an internal DAGs within each time unit representing a parent-child or ancestor-descendant relationship,which exhibits a structure similar to the confounding.This paper summarizes the construction of causality frames for objective world research (causal DAGs),and clarify a structural basis for the control of the confounding in effect estimate.
ABSTRACT
In the studies of modem epidemiology,exposure in a short term cannot fully elaborate the mechanism of the development of diseases or health-related events.Thus,lights have been shed on to life course epidemiology,which studies the exposures in early life time and their effects related to the development of chronic diseases.When exploring the mechanism leading from one exposure to an outcome and its effects through other factors,due to the existence of time-variant effects,conventional statistic methods could not meet the needs of etiological analysis in life course epidemiology.This paper summarizes the dynamic path analysis model,including the model structure and significance,and its application in life course epidemiology.Meanwhile,the procedure of data processing and etiology analyzing were introduced.In conclusion,dynamic path analysis is a useful tool which can be used to better elucidate the mechanisms that underlie the etiology of chronic diseases.
ABSTRACT
Nearly all scientific studies explore causality,which will be met by directed acyclic graphs (DAGs).This paper systematically introduces graphic language,basic and interference rules of DAGs,and their applications into identifying research questions,understanding and undertaking research designs,guiding data analysis,classifying biases,etc.DAGs play key roles in causality studies.
ABSTRACT
Nearly all scientific studies explore causality,which will be met by directed acyclic graphs (DAGs).This paper systematically introduces graphic language,basic and interference rules of DAGs,and their applications into identifying research questions,understanding and undertaking research designs,guiding data analysis,classifying biases,etc.DAGs play key roles in causality studies.