Examkin

Question

Difficulty: HardBias and Confounding in Research

An epidemiological review board is evaluating four recent clinical studies for methodological validity. Each study exhibits a distinct analytical or structural flaw that distorts its findings. Review the methodology descriptions below and pair each study with the specific epidemiological phenomenon that best characterizes its flaw.

  • Study 1: A retrospective analysis compares surgical resection rates between patients receiving a novel third-line biologic and those on standard first-line therapy for severe Crohn disease. The third-line biologic is exclusively prescribed to patients who have failed multiple prior interventions. Results show a significantly higher resection rate in the novel biologic group.Confounding by indication
  • Study 2: A randomized controlled trial investigates a new oral anticoagulant. The treatment protocol mandates monthly hepatic ultrasounds for the intervention arm, whereas the control arm receives standard clinical follow-up. The trial reports a significantly increased incidence of asymptomatic benign hepatic hemangiomas in the treatment arm.Surveillance (detection) bias
  • Study 3: A case-control study investigating the association between a rare genetic marker and respiratory failure recruits both cases and controls exclusively from an inpatient intensive care unit. Because both the exposure and outcome independently increase the likelihood of ICU admission, the resulting odds ratio is falsely elevated.Berkson paradox (admission rate bias)
  • Study 4: A newly implemented community screening program for prostate cancer identifies a high volume of new cases. Pathological analysis reveals that the screened cohort is diagnosed almost entirely with indolent, slowly progressive tumors, unlike the unscreened population which typically presents with aggressive variants upon symptom onset.Length-time bias

Answer

Study 1 demonstrates Confounding by indication, Study 2 demonstrates Surveillance (detection) bias, Study 3 demonstrates Berkson paradox (admission rate bias), and Study 4 demonstrates Length-time bias.
Each study description perfectly aligns with a classic epidemiological phenomenon. Study 1 involves treatment allocation based on severity (Confounding by indication). Study 2 involves unequal diagnostic scrutiny (Surveillance bias). Study 3 involves studying intersecting admission probabilities in a hospital setting (Berkson paradox). Study 4 involves screening capturing slower-growing disease variants (Length-time bias).

Step-by-Step Solution

1
Analyze Study 1's methodology.
Identified that the treatment assignment depends on disease severity (failure of prior interventions).
When a provider's indication for prescribing a drug is correlated with the outcome of interest, it creates confounding by indication.
2
Analyze Study 2's methodology.
Identified unequal monitoring intervals between study arms.
Unequal monitoring leads to unequal detection of incidentalomas, known as surveillance or detection bias.
3
Analyze Study 3's methodology.
Noted the restriction of the sample to an ICU subpopulation.
When exposure and outcome independently affect hospital admission, selecting only hospitalized patients distorts the true association, defining the Berkson paradox.
4
Analyze Study 4's methodology.
Identified the disproportionate capture of slowly progressive disease during screening.
Diseases with longer asymptomatic phases are statistically more likely to be caught by periodic screening, which characterizes length-time bias.

Key Concept

Identifying specific types of bias and confounding requires analyzing whether the distortion arises from treatment allocation (confounding by indication), intensity of monitoring (surveillance bias), population selection parameters (Berkson paradox), or the natural biological history of the disease intersecting with screening intervals (length-time bias).
Estimated Time:2m 0s
Rate this question

Topics

Bias and Confounding in Research

More questions from this topic