CSC Research & Evidence-Based Practice 5 — Questions and Answers
Question 1: A researcher conducts 20 hypothesis tests in a cardiac outcomes study and finds 2 significant results at p < 0.05. What statistical problem is most likely?
- Type II error inflation
- Multiple comparisons increasing Type I error rate (Correct answer)
- Inadequate statistical power
- Undercoverage sampling bias
Correct answer: Multiple comparisons increasing Type I error rate
Performing multiple comparisons inflates the Type I error rate; by chance alone, 1 in 20 tests would be expected to reach p < 0.05.
Question 2: A cardiac rehabilitation program is evaluated using a pre-post design without a control group. What is the main limitation of this design?
- It cannot measure continuous outcomes
- It is unable to establish temporal sequence
- It cannot control for regression to the mean and natural disease progression (Correct answer)
- It requires blinding of participants
Correct answer: It cannot control for regression to the mean and natural disease progression
Without a control group, improvements may reflect natural disease progression or regression to the mean rather than the intervention's effect.
Question 3: Which agency publishes clinical practice guidelines specifically graded with the GRADE system for cardiovascular care?
- American Heart Association (AHA)
- World Health Organization (WHO) via GRADE Working Group integration (Correct answer)
- FDA Center for Drug Evaluation
- Joint Commission on Accreditation
Correct answer: World Health Organization (WHO) via GRADE Working Group integration
The GRADE (Grading of Recommendations Assessment, Development and Evaluation) system was developed by the GRADE Working Group and is used by WHO and many cardiovascular guideline bodies.
Question 4: A clinical trial enrolls 500 patients per arm to detect a 20% relative risk reduction in cardiac mortality with 80% power at α=0.05. If the actual effect is smaller than expected, what happens to power?
- Power increases because more events occur
- Power decreases, increasing the likelihood of a Type II error (Correct answer)
- Power is unaffected by effect size
- Power increases due to the larger-than-needed sample
Correct answer: Power decreases, increasing the likelihood of a Type II error
If the true effect is smaller than the assumed effect size, the study is underpowered and more likely to miss a real difference (Type II error).
Question 5: In an EBP model, after a PICO question is formulated for anticoagulation in cancer-associated thrombosis, the next step is to:
- Apply the best available evidence immediately
- Search and acquire the best available evidence (Correct answer)
- Evaluate patient outcomes after the intervention
- Convene a multidisciplinary committee
Correct answer: Search and acquire the best available evidence
After formulating the PICO question (Ask), the next EBP step is to Acquire the best available evidence through a systematic literature search.
Question 6: A study reports a number needed to treat (NNT) of 50 for a new antiplatelet agent to prevent one MI over 5 years. How should this be interpreted?
- 50% of treated patients will benefit from the drug
- 50 patients must be treated for 5 years to prevent one MI (Correct answer)
- The drug prevents MI in 1 out of every 5 patients treated
- The drug has a 50% relative risk reduction
Correct answer: 50 patients must be treated for 5 years to prevent one MI
NNT of 50 means 50 patients must receive the treatment for the specified duration to prevent one additional MI compared to control.
Question 7: In a blinded RCT of a cardiac device, blinding of outcome assessors is most critical to prevent which type of bias?
- Performance bias
- Attrition bias
- Detection bias (Correct answer)
- Reporting bias
Correct answer: Detection bias
Detection bias occurs when outcome assessors are aware of treatment allocation and unconsciously adjudicate outcomes differently; blinding prevents this.
CSC Research & Evidence-Based Practice 5
A researcher conducts 20 hypothesis tests in a cardiac outcomes study and finds 2 significant results at p < 0.05.
What statistical problem is most likely?