A 48 year old man with alcohol use disorder is admitted for pancreatitis. Researchers study only hospitalized patients and find that among admitted patients, obesity appears negatively associated with alcohol use disorder. In the source population, however, both obesity and alcohol use independently increase hospitalization risk. Which of the following best explains the spurious inverse association seen in the hospitalized sample?
- A)Lead time bias
- B)Confounding
- C)Collider biasGABARITO
- D)Observer expectancy bias
- E)Random error
Explicação
Collider bias occurs when investigators condition on a variable that is caused by both the exposure and the outcome or by two exposures being compared. Here, hospitalization is influenced by both obesity and alcohol related illness, so restricting the analysis... Ver explicação completa e trilha adaptativa →