PRC Research Methods & Evidence-Based Practice 3 — Questions and Answers
Question 1: A PRI researcher calculates a Pearson correlation of r = 0.62 between thoracic rotation ROM and a functional outcome score. Which statement best interprets this finding?
- Thoracic rotation causes improved functional outcomes
- Approximately 38% of variance in functional outcomes is explained by thoracic rotation ROM
- The relationship explains 62% of shared variance between variables
- A moderate-to-strong linear relationship exists with 38.4% shared variance (Correct answer)
Correct answer: A moderate-to-strong linear relationship exists with 38.4% shared variance
The coefficient of determination (r²= 0.384) indicates that 38.4% of variance in one variable is accounted for by the other, while r = 0.62 reflects a moderate-to-strong linear relationship.
Question 2: Which research design would provide the strongest evidence for establishing that a specific PRI inhibition technique causes neuromuscular changes in the iliopsoas?
- Prospective cohort study with matched controls
- Randomized controlled trial with an active control condition (Correct answer)
- Case series documenting pre-post EMG changes in 10 patients
- Cross-sectional survey of PRC clinicians' technique preferences
Correct answer: Randomized controlled trial with an active control condition
An RCT with randomization and an active control condition provides the strongest evidence for causation by controlling for confounding and placebo effects.
Question 3: A PRI assessment tool has sensitivity of 85% and specificity of 70% for identifying a right BC pattern. If a patient tests positive, the POST-test probability of truly having a right BC pattern depends on:
- Sensitivity and specificity values only
- Specificity and the false negative rate
- Pre-test probability (prevalence) combined with the likelihood ratio (Correct answer)
- The standard error of measurement of the test
Correct answer: Pre-test probability (prevalence) combined with the likelihood ratio
Post-test probability is determined using Bayes' theorem, combining the pre-test probability (prevalence) with the positive likelihood ratio (sensitivity/[1-specificity]).
Question 4: In a PRI outcomes study, researchers control for age, BMI, and pain duration using multiple regression. The purpose of including these covariates is to:
- Increase sample size requirements for adequate power
- Isolate the effect of the PRI intervention by accounting for variables that may confound the relationship (Correct answer)
- Improve the reliability of the outcome measurement tools
- Ensure blinding of outcome assessors to group assignment
Correct answer: Isolate the effect of the PRI intervention by accounting for variables that may confound the relationship
Covariates in regression models statistically control for variables known to influence the outcome, allowing a cleaner estimate of the intervention effect.
Question 5: A PRI clinical prediction rule (CPR) is described as having a positive likelihood ratio (LR+) of 8.5 for predicting response to a repositioning intervention. This LR+ indicates:
- The intervention is 8.5 times more effective than placebo
- A positive test result moderately increases post-test probability of response
- A positive test result generates a large and clinically meaningful shift in probability (Correct answer)
- 85% of patients who test positive will respond to treatment
Correct answer: A positive test result generates a large and clinically meaningful shift in probability
An LR+ of 8.5 generates a large shift in post-test probability (LR+ > 5 is considered large and clinically meaningful per Jaeschke et al.).
Question 6: Which statistical test is most appropriate when comparing PRI outcomes (measured on an ordinal scale) between three independent treatment groups with non-normally distributed data?
- One-way ANOVA with Tukey post-hoc test
- Kruskal-Wallis test (Correct answer)
- Repeated measures ANOVA
- Mann-Whitney U test
Correct answer: Kruskal-Wallis test
The Kruskal-Wallis test is the non-parametric equivalent of one-way ANOVA, appropriate for comparing three or more independent groups with ordinal or non-normal data.
Question 7: A PRC clinician critiques a published case report on PRI treatment for PFPS. Which statement correctly identifies a key limitation of this evidence level?
- Case reports cannot describe mechanistic hypotheses about treatment effects
- Findings cannot be generalized to the broader population due to lack of a control condition and single-patient focus (Correct answer)
- Case reports are unacceptable in peer-reviewed rehabilitation journals
- A case report can never include quantitative outcome measures
Correct answer: Findings cannot be generalized to the broader population due to lack of a control condition and single-patient focus
Case reports describe individual patients without controls, making generalization impossible and causal inference unwarranted, though they are valuable for hypothesis generation.
PRC Research Methods & Evidence-Based Practice 3
A PRI researcher calculates a Pearson correlation of r = 0.62 between thoracic rotation ROM and a functional outcome score.
Which statement best interprets this finding?