Match each clinical research scenario or observation with the specific epidemiological concept, methodological phenomenon, or bias it best illustrates.
- An epidemiological analysis initially suggests that individuals who consume green tea daily have a drastically lower risk of developing cognitive decline. Once investigators stratify the data by baseline exercise frequency and income level, the protective link completely vanishes in all subgroups.Confounding
- A multicenter trial for a new antithrombotic medication yields an overall efficacy that is statistically identical to the current standard of care. When investigators separate the participants by CYP450 genotype, they discover the drug profoundly prevents strokes in one genetic variant while increasing bleeding without benefit in the other.Effect modification
- To study the relationship between severe chronic obstructive pulmonary disease and gastroesophageal reflux, investigators assemble a cohort of cases and controls selected entirely from the intensive care units of a tertiary referral center, inadvertently manufacturing a false association.Berkson paradox
- While testing a new artificial intelligence tool designed to detect pulmonary nodules on chest radiographs, clinicians only order the definitive, highly sensitive diagnostic CT scans for patients whom the AI algorithm has already flagged as 'high probability'.Verification (workup) bias
Cevap
The scenarios correctly match to Confounding (association disappears upon stratification), Effect modification (varying effects across strata), Berkson paradox (hospital-based selection bias), and Verification bias (differential application of the gold standard).
Each scenario perfectly exemplifies a specific epidemiological concept. Confounding is revealed when stratification nullifies an association. Effect modification is revealed when stratification uncovers significantly different effects between groups. Berkson paradox occurs due to selective sampling from hospitalized patients. Verification bias occurs when the index test dictates who receives the gold standard test.
Adım Adım Çözüm
Anahtar Kavram
Methodological flaws and phenomena in study design, specifically differentiating selection biases, measurement biases, and effect modifiers.