MPT Research Methods and Evidence-Based Practice 2 — Questions and Answers
Question 1: A clinical test has a sensitivity of 90%. This means:
- 90% of people who test negative actually have the disease
- 90% of people who test positive actually have the disease
- 90% of people who have the disease will test positive (Correct answer)
- 90% of people without the disease will test negative
Correct answer: 90% of people who have the disease will test positive
Sensitivity is the true positive rate — the proportion of people with the condition who correctly test positive; high sensitivity means few false negatives.
Question 2: The Number Needed to Treat (NNT) is best calculated as:
- The ratio of treatment benefit to potential harm
- The reciprocal of the absolute risk reduction (Correct answer)
- The difference in outcome rates between treatment and control groups
- Total participants divided by the number who benefited
Correct answer: The reciprocal of the absolute risk reduction
NNT = 1/Absolute Risk Reduction and represents how many patients must be treated for one additional patient to benefit compared to the control condition.
Question 3: The minimal clinically important difference (MCID) is best defined as:
- The smallest measurement error acceptable for a reliable outcome tool
- The statistical significance threshold for reporting outcome changes
- The smallest change in an outcome measure that patients perceive as meaningful (Correct answer)
- The average improvement expected with a standard course of treatment
Correct answer: The smallest change in an outcome measure that patients perceive as meaningful
MCID represents the threshold of change that patients consider beneficial, bridging the gap between statistical significance and clinical relevance.
Question 4: A 95% confidence interval for a mean difference that does not include zero indicates:
- The result is clinically significant
- Statistical significance at the p<0.05 level (Correct answer)
- The sample size was sufficiently large
- The study had high internal validity
Correct answer: Statistical significance at the p<0.05 level
A 95% CI not crossing zero corresponds to p<0.05, indicating the true difference is unlikely to be zero and the result is statistically significant.
Question 5: The Intraclass Correlation Coefficient (ICC) is most appropriately used to assess:
- The validity of a diagnostic screening test
- The effect size magnitude of a therapeutic intervention
- The reliability of continuous measurement data (Correct answer)
- The correlation between two different construct variables
Correct answer: The reliability of continuous measurement data
ICC quantifies reliability (test-retest or inter-rater) for continuous data, with values approaching 1.0 indicating excellent reliability.
Question 6: In a prospective cohort study, participants are:
- Randomly assigned to treatment or control conditions
- Selected because they already have the outcome of interest
- Identified by exposure status and followed forward in time to observe outcomes (Correct answer)
- Matched based on similar demographic and clinical characteristics
Correct answer: Identified by exposure status and followed forward in time to observe outcomes
In prospective cohort studies, participants are enrolled based on exposure status before the outcome occurs and then followed forward in time to observe outcome incidence.
Question 7: The kappa statistic is used to measure:
- Correlation between two continuous variables
- Agreement between categorical measurements beyond what is expected by chance (Correct answer)
- The strength and direction of a treatment effect
- Reliability of continuous outcome measurements
Correct answer: Agreement between categorical measurements beyond what is expected by chance
Cohen's kappa measures inter-rater agreement for categorical data, correcting for agreement that would be expected by chance alone.
MPT Research Methods and Evidence-Based Practice 2
A clinical test has a sensitivity of 90%.
This means: