Related Experiment Videos
Causal inference in infectious diseases
1Department of Biostatistics, Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA.
Abstract:
Since the 1970s, Rubin has promoted a model for causal inference based on the potential outcomes if individuals received each of the treatments under study. Commonly, the assumption is made that the outcome in one individual is independent of the treatment assignment and outcome in other individuals. In infectious diseases, however, whether one person become infected is quite often dependent on the infection outcome in other individuals, a situation known as dependent happenings. Here, we review the model proposed by Rubin for the example of infectious disease. Consequences of the violation of the stability assumption include the need for an expanded representation of outcomes, and the existence of different kinds of effects, such as direct and indirect effects. Effects of interest include changes in susceptibility as well as changes in infectiousness. We define the transmission probability formally as an average causal parameter of effect in a population by conditioning on exposure to infection. Unconditional indirect and total effects are difficult to define formally using this model for causal inference. The assignment mechanism can influence the sampling mechanism when it determines who is exposed to infection, raising problems that require further inquiry. We conclude by contrasting the role of differential exposure to infection in direct and indirect effects.
Insights
This study reviews Rubin's causal inference model for infectious diseases, highlighting how dependent outcomes (like infections spreading) violate standard assumptions and require new methods to analyze direct and indirect effects.
Area of Science:
- Epidemiology
- Causal Inference
- Biostatistics
Background:
- Rubin's causal inference model, widely used since the 1970s, assumes independence of outcomes between individuals.
- Infectious disease dynamics often exhibit 'dependent happenings,' where one person's infection status depends on others.
- This violates the stability assumption crucial for standard causal inference models.
Purpose of the Study:
- To review Rubin's causal inference model in the context of infectious diseases.
- To explore the consequences of violating the stability assumption in infectious disease studies.
- To define and analyze causal effects, including direct and indirect effects, in the presence of dependent outcomes.
Main Methods:
- Review of Rubin's potential outcomes framework for causal inference.
- Adaptation of the model to account for 'dependent happenings' in infectious disease transmission.
- Formal definition of transmission probability as an average causal parameter.
- Analysis of challenges in defining unconditional indirect and total effects.
Main Results:
- Violation of the stability assumption necessitates an expanded outcome representation.
- Direct and indirect effects, including changes in susceptibility and infectiousness, become key considerations.
- Defining unconditional indirect and total effects within this framework presents formal challenges.
- The assignment mechanism's influence on exposure complicates causal effect estimation.
Conclusions:
- The standard Rubin model requires adaptation for infectious disease causal inference due to dependent outcomes.
- New approaches are needed to formally define and estimate indirect and total effects in these settings.
- Differential exposure to infection plays a critical role in understanding direct versus indirect effects.