Week 11 Flashcards
What is interaction?
Interaction is present if the effect of an exposure on an outcome depends on the presence, or level, of a third factor happens
What is another name for interaction?
effect modification
What is the difference between biological and statistical interaction?
Biological interaction is situated in causal theory (such as the potential outcomes or sufficient cause model), and is closely related to the mechanisms underlying a causal relationship. Statistical interaction can be tested using the data.
What does it mean to say that “interaction is scale dependent”?
Interaction should be assessed on both the additive scale (absolute measures) and multiplicative scale (relative measures). It could exist on one scale, but not the other scale, on both scales, or on neither scale.
How does interaction differ from confounding and bias?
Confounding is a distortion of the association between an exposure and outcome that occurs when the groups differ with respect to other factors that influence the outcome and exists in nature. Bias is introduced into the study by the investigator. Confounding and bias should be adjusted or controlled for in order to estimate a causal effect. Interaction exists when the strength of an association between the exposure and outcome differs depending on a third factor. Interaction should not be controlled for, rather it should be reported.
A study conducted in Brazil finds that wearing limb-covering clothing (long sleeve shirts and pants rather than shorts) is protective against contracting Zika virus (carried by mosquitos). Give an example of a possible third variable that might modify that association.
works inside vs works outside
What is synergistic (positive interaction)?
If interaction exists and the presence of the effect modifier strengthens the effect of the exposure of interest.
What is antagonistic interactions?
If interaction exists and the presence of the effect modifier diminished or eliminates the effect of the exposure of interest.
If effect modification/ interaction exists, will you report the crude association between SSB consumption and obesity or the stratum-specific associations? Why?
Stratum-specific, because the effect of SSB consumption on obesity varies across strata of the genetic variant. Interaction describes the effect of an exposure under different possible conditions, enriching our understanding of the exposure-outcome relationship. We therefore want to report it.
What are strategies to evaluate interaction?
heterogeneity of effects, comparing observed vs. expected joint effects
We want to recognize interaction when it exists and control for it. True or false?
False
What is an example of interaction for the association between lead exposure and learning disabilities in children?
The association is stronger among children whose parents smoke
A researcher finds that the risk of lung cancer among people who smoke is lower among the subgroup of people who eat a diet with a lot of coniferous vegetables. What is this an example of?
antagonistic (negative) interaction
There is a statistical test that can be used during data analysis for interaction. True or false?
TRUE
When does multiplicative or additive interaction occur?
When the joint effects are more than what we expected from the independent effects.
