Evaluating Scientific Evidence Flashcards

A systematic review includes only studies published in English. This restriction introduces which type of bias?

Answer: Language bias

Excluding non-English studies introduces language bias because findings from other regions may differ systematically, skewing the pooled conclusions.

A cohort study follows 5,000 smokers and 5,000 non-smokers for 20 years to compare lung cancer incidence. This design is best classified as:

Answer: Prospective cohort

A prospective cohort study follows participants forward in time from exposure to outcome, making this the correct classification.

Researchers report a confidence interval of 95% CI [0.85–1.15] for a risk ratio. What does this interval suggest?

Answer: The data are consistent with no meaningful difference between groups

A 95% CI that includes 1.0 for a risk ratio indicates the result is not statistically significant, and a null effect (no difference) is plausible.

In evidence-based practice, which source of evidence should take precedence when making a clinical decision?

Answer: Current best research evidence integrated with clinical expertise and patient values

Evidence-based practice explicitly integrates the best available research evidence with clinician expertise and individual patient preferences.

A drug company funds a trial comparing its new medication to a competitor's. The trial shows superiority. Which concern most threatens the validity of this conclusion?

Answer: Industry funding may introduce reporting or design bias favoring the sponsor

Industry-funded research is associated with a higher likelihood of favorable outcomes for the sponsor due to selective reporting or methodological choices.

A researcher performs 20 separate hypothesis tests on one dataset and finds 1 significant result (p < 0.05). The most appropriate interpretation is:

Answer: One significant result is expected by chance alone when conducting 20 tests

With a 5% false-positive rate, conducting 20 tests predicts approximately one spurious significant result by chance, a problem known as multiple comparisons.

A health study uses self-reported dietary data collected via questionnaires administered years after the study period. This introduces which type of bias?

Answer: Recall bias

Recall bias occurs when participants inaccurately remember past exposures, and the risk increases when there is a long interval between the exposure and the data collection.