Examkin

Question

Difficulty: MediumBias and Confounding in Research

Public health officials review data from a large population registry to investigate the relationship between residential proximity to major highways and the development of childhood asthma. An initial unadjusted analysis reveals a relative risk of 2.1 (95% CI: 1.6-2.8) for asthma in children living within 100 meters of a highway compared to those living farther away. The researchers subsequently conduct a stratified analysis based on household income levels. The relative risk for asthma among children in low-income households is 1.1 (95% CI: 0.8-1.4), and the relative risk among children in high-income households is 1.2 (95% CI: 0.9-1.5). Which of the following epidemiologic phenomena best explains the discrepancy between the initial and stratified findings?

  1. ConfoundingAnswer
  2. B
    Effect modification
  3. C
    Selection bias
  4. D
    Recall bias
  5. E
    Detection bias

Answer

Confounding
The correct answer identifies that the initial association between highway proximity and asthma was a spurious result caused by a third variable: household income. In epidemiological studies, when a crude measure of association (like the initial RR of 2.1) differs significantly from the stratified measures of association (the RRs of 1.1 and 1.2), but the stratified measures are similar to each other, the third variable is acting as a confounder. Low income is likely associated with both living near highways (cheaper housing) and higher asthma rates (due to other environmental or systemic factors), creating the false impression that the highway itself was the primary risk factor.

Step-by-Step Solution

1
Analyze the crude, unadjusted statistical findings.
The initial relative risk is 2.1 (95% CI: 1.6-2.8), indicating a statistically significant association between highway proximity and asthma.
To establish the baseline apparent relationship between the exposure and outcome.
2
Evaluate the statistical findings after stratification by the third variable (household income).
The relative risks drop to 1.1 and 1.2, both of which cross 1.0 in their confidence intervals and are therefore not statistically significant.
To determine how adjusting for the third variable impacts the observed association.
3
Compare the crude and stratified measures of association.
The stratified relative risks are similar to each other but distinctly different from the crude relative risk, confirming that household income was responsible for the initial apparent association.
This pattern is the classic epidemiological hallmark used to distinguish confounding from effect modification.

Key Concept

Identifying confounding by comparing crude and stratified measures of association.
Rate this question

Topics

Bias and Confounding in Research

More questions from this topic