Learning Check | Module 3: Bias in Epidemiological Studies (Part 3) - Selection Bias

In 1990, a case-control study was conducted to investigate the positive association between artificial sweetener use and bladder cancer. Controls were selected from a hospital sample of patients diagnosed with obesity-related conditions. Obesity-related conditions have been positively associated with artificial sweetener use. How would the use of these patients as controls affect the estimate of the association between artificial sweetener use and bladder cancer?

The estimate of association would accurately reflect the true association regardless of the association between artificial sweetener use and obesity-related conditions
The estimate of association would tend to underestimate the true association
The estimate of association would tend to overestimate the true association

True or False: Selection bias can only arise in observational studies.

True
False

Investigators conducted an observational study of past vitamin E supplementation and infertility. The researchers evaluated the relationship between taking a daily vitamin E supplement over the past year and abnormal sperm count among males. The participants were all recruited from a health clinic. Health-conscious men who took vitamin supplements were more likely to visit the health clinic compared to men who did not take vitamin supplements. Additionally, men having difficulty conceiving were also more likely to attend the health clinic.

Which of the DAGs below represents the study the investigators conducted? (assume the assumptions of well-defined interventions and positivity are met)

DAG 1
DAG 2
DAG 3

Learning Check | Module 3: Bias in Epidemiological Studies (Part 3) – Selection Bias

Quiz Summary

Information

Results

Results

Categories

1. Question

2. Question

3. Question