Examkin

Question

Difficulty: MediumBias, Confounding, and Validity

Public health researchers are investigating the cardiovascular safety of a widely used electronic vaping device. In an initial, unadjusted analysis of an observational cohort containing 15,000 adults, the researchers calculate a relative risk of 2.2 (95% CI: 1.6 - 3.0) for the development of incident coronary artery disease among daily users of the device compared to non-users.

Suspecting that other behavioral factors might be influencing the results, the research team performs a secondary analysis stratified by a history of traditional combustible cigarette smoking. In the stratified analysis, the relative risk for coronary artery disease among vaping device users is 1.06 (95% CI: 0.8 - 1.3) in those with a history of smoking, and 1.04 (95% CI: 0.7 - 1.4) in those without a history of smoking.

Based on these findings, which of the following best characterizes the role of traditional cigarette smoking in this study?

  1. It is associated with both the exposure and the outcome but does not lie on the causal pathway.Answer
  2. B
    It biologically alters the magnitude of the exposure's effect, demonstrating true stratum-specific differences.
  3. C
    It indicates that the investigators should have utilized an analysis of variance rather than relative risk calculations.
  4. D
    It shifts the predictive values of the diagnostic studies by altering the baseline prevalence of the disease.
  5. E
    It represents a clinical distractor that leads to the overinterpretation of transient, mild laboratory abnormalities.

Answer

Traditional cigarette smoking acts as a confounding variable because it is associated with both the exposure (vaping) and the outcome (coronary artery disease) without being in the causal pathway.
The correct answer identifies traditional cigarette smoking as a confounding variable. A confounder is an extraneous factor that correlates with both the independent variable (vaping) and the dependent variable (coronary artery disease) but does not lie on the causal pathway between them. In this scenario, individuals who vape are more likely to have a history of traditional smoking, and traditional smoking is an independent cause of coronary artery disease. When the researchers performed a stratified analysis to isolate the groups, the relative risk of coronary artery disease among vaping users dropped to near 1.0 in both strata (confidence intervals 0.8 - 1.3 and 0.7 - 1.4). Because the initial significant association disappeared once the data were separated by smoking status, it is clear that smoking was responsible for the spurious unadjusted relative risk of 2.2.

Step-by-Step Solution

1
Analyze the unadjusted study results.
The initial relative risk is 2.2, suggesting a significant association between vaping and coronary artery disease.
Establishing the baseline finding is necessary before interpreting the effect of stratification.
2
Evaluate the results after stratifying by the suspected variable (smoking).
The relative risks in both strata (1.06 and 1.04) are approximately 1.0, and the confidence intervals cross 1.0.
Stratification is a primary method used to distinguish between confounding and effect modification.
3
Interpret the change in risk.
Because the association disappears upon stratification, the initial relationship was spurious and driven by smoking status.
This behavior perfectly matches the definition of a confounder, not an effect modifier.

Key Concept

In epidemiological studies, confounding occurs when a variable is associated with both the exposure and the outcome but is not on the causal pathway, leading to a spurious association that disappears upon stratification.
Rate this question

Topics

Bias, Confounding, and Validity

More questions from this topic