DPT Research & Evidence-Based Practice 3 — Questions and Answers
Question 1: A researcher reports sensitivity of 90% and specificity of 60% for a clinical test. Which statement is correct?
- The test is better at ruling in disease than ruling it out
- The test is better at ruling out disease than ruling it in (Correct answer)
- The test has high positive predictive value in all populations
- Sensitivity and specificity are population-dependent statistics
Correct answer: The test is better at ruling out disease than ruling it in
High sensitivity (90%) means the test rarely misses true positives, making it effective for ruling out disease (SnOUT mnemonic).
Question 2: Which type of reliability coefficient is most appropriate for evaluating agreement between two therapists rating the same ordinal outcome?
- Pearson r
- Intraclass correlation coefficient (ICC)
- Cohen's kappa (Correct answer)
- Cronbach's alpha
Correct answer: Cohen's kappa
Cohen's kappa is the preferred statistic for inter-rater reliability when data are categorical or ordinal.
Question 3: A PT reviews a clinical prediction rule (CPR) that has been validated but not yet impact-analyzed. At which phase of CPR development is this rule?
- Derivation phase
- Validation phase (Correct answer)
- Impact analysis phase
- Implementation phase
Correct answer: Validation phase
A CPR that has been validated but not yet tested for its impact on clinical decision-making and outcomes remains in the validation phase.
Question 4: A study reports a Number Needed to Treat (NNT) of 5 for a new exercise protocol. What does this mean?
- 5 patients must be treated for 1 to experience harm
- 5 patients must be treated for 1 additional patient to benefit (Correct answer)
- The treatment works 5 times better than placebo
- 5% of patients will not respond to treatment
Correct answer: 5 patients must be treated for 1 additional patient to benefit
NNT represents the number of patients who must receive a treatment for one additional patient to benefit compared to the control condition.
Question 5: Which bias occurs when participants who drop out of a study differ systematically from those who complete it?
- Recall bias
- Attrition bias (Correct answer)
- Detection bias
- Performance bias
Correct answer: Attrition bias
Attrition bias occurs when the characteristics of dropouts differ from completers, potentially distorting study results.
Question 6: A researcher designs a study where neither participants nor outcome assessors know group allocation. This describes which type of blinding?
- Single-blind
- Double-blind (Correct answer)
- Triple-blind
- Open-label
Correct answer: Double-blind
Double-blinding means both participants and the outcome assessors are unaware of group assignments, reducing performance and detection bias.
Question 7: The GRADE framework is used in systematic reviews to rate the quality of evidence. Which factor can UPGRADE the quality rating?
- Risk of bias in included studies
- Large magnitude of effect (Correct answer)
- Inconsistency across studies
- Indirectness of evidence
Correct answer: Large magnitude of effect
A very large magnitude of effect is one of three factors (along with dose-response and confounding that would underestimate effect) that can upgrade evidence quality in GRADE.
DPT Research & Evidence-Based Practice 3
A researcher reports sensitivity of 90% and specificity of 60% for a clinical test.
Which statement is correct?