CLC Research Methods & Evidence-Based Practice 2 — Questions and Answers
Question 1: In a case-control study examining a rare laboratory error type, researchers compare 50 cases with 200 controls. What is the primary measure of association used?
- Relative risk
- Odds ratio (Correct answer)
- Attributable risk
- Number needed to treat
Correct answer: Odds ratio
Odds ratio is the appropriate measure in case-control studies because incidence cannot be calculated directly from such designs.
Question 2: A laboratory implements a new QC algorithm. To assess its diagnostic performance, staff calculate sensitivity and specificity. Which combination best minimizes patient harm from missed critical values?
- High specificity, low sensitivity
- High sensitivity, low specificity (Correct answer)
- Equal sensitivity and specificity
- Low sensitivity, low specificity
Correct answer: High sensitivity, low specificity
High sensitivity minimizes false negatives, ensuring critical values are rarely missed, which is paramount for patient safety.
Question 3: A systematic review pools five studies on turnaround time interventions. Cochran's Q test yields p=0.02. What does this indicate?
- Studies are homogeneous and pooling is appropriate
- Significant heterogeneity exists among studies (Correct answer)
- The combined effect size is statistically significant
- Publication bias is present
Correct answer: Significant heterogeneity exists among studies
A significant Cochran's Q (p<0.10) indicates substantial statistical heterogeneity, warranting caution in pooling results.
Question 4: Which of the following best describes the principle of 'intention-to-treat' analysis in a randomized controlled trial of a new laboratory workflow?
- Analyzing only participants who completed the protocol
- Analyzing participants based on the group they were assigned to, regardless of adherence (Correct answer)
- Excluding outliers before analysis
- Analyzing only the primary outcome measure
Correct answer: Analyzing participants based on the group they were assigned to, regardless of adherence
Intention-to-treat analysis preserves randomization and avoids bias by analyzing all participants in their originally assigned groups.
Question 5: A laboratory director reviews a study with an I² value of 78%. What is the most appropriate interpretation?
- Low heterogeneity; fixed-effects model is preferred
- Moderate heterogeneity; either model is acceptable
- High heterogeneity; a random-effects model or subgroup analysis is warranted (Correct answer)
- The study results are invalid and should be discarded
Correct answer: High heterogeneity; a random-effects model or subgroup analysis is warranted
I² above 75% indicates high heterogeneity, suggesting a random-effects model or investigation of sources of variability is needed.
Question 6: When evaluating a new laboratory test, a receiver operating characteristic (ROC) curve is constructed. An AUC of 0.95 indicates:
- The test performs no better than chance
- Excellent discriminative ability between diseased and non-diseased populations (Correct answer)
- The optimal cutoff has been identified
- The test has 95% sensitivity
Correct answer: Excellent discriminative ability between diseased and non-diseased populations
AUC of 0.95 indicates excellent discrimination, meaning the test correctly ranks a diseased individual above a non-diseased individual 95% of the time.
Question 7: A quality improvement team uses a Pareto chart to prioritize laboratory errors. Which principle underlies this tool?
- All error types contribute equally to overall defect rates
- Roughly 80% of problems stem from 20% of causes (Correct answer)
- Random sampling identifies the most impactful errors
- Statistical process control determines error priority
Correct answer: Roughly 80% of problems stem from 20% of causes
The Pareto principle holds that ~80% of effects come from ~20% of causes, helping teams focus on the most impactful error sources.
CLC Research Methods & Evidence-Based Practice 2
In a case-control study examining a rare laboratory error type, researchers compare 50 cases with 200 controls.
What is the primary measure of association used?