Evidence-Based Evaluation - National Physical Therapy Examination NPTE

1. Identify the formula for number needed to treat (NNT) from absolute risk reduction (ARR).

   $NNT=\frac{1}{ARR}$. This reciprocal indicates how many patients must receive treatment to prevent one additional adverse event.

2. Which effect measure is most appropriate for time-to-event prognosis studies?

   Hazard ratio. It compares the instantaneous risk of an event between groups in survival analysis, accounting for time.

3. What is the formula for negative predictive value (NPV) using a $2\times^2$ table?

   $NPV=\frac{TN}{TN+FN}$. This measures the probability that a negative test result truly indicates the absence of the condition.

4. Which magnitude of $LR^+$ is generally considered strong evidence to rule in a diagnosis?

   $LR^+\ge 10$. A value of 10 or greater substantially increases post-test probability, providing strong diagnostic confirmation.

5. What is the formula for specificity using a $2\times^2$ diagnostic table?

   $\text{specificity}=\frac{TN}{TN+FP}$. This proportion identifies true negatives among all individuals without the condition in a $2\times^2$ contingency table.

6. Which bias is most reduced by allocation concealment in an RCT?

   Selection bias. Allocation concealment prevents investigators from influencing group assignments, reducing non-comparable groups.

7. Which diagnostic metric is least affected by disease prevalence: predictive values or likelihood ratios?

   Likelihood ratios. Unlike predictive values, which vary with prevalence, likelihood ratios remain stable across populations.

8. Identify the formula for absolute risk reduction (ARR) given event rates $CER$ and $EER$.

   $ARR=CER-EER$. This subtracts the experimental event rate from the control event rate to quantify absolute benefit.

9. What study design is highest level of evidence for overall treatment effectiveness?

   Systematic review with meta-analysis of RCTs. This design synthesizes multiple RCTs, providing the highest evidence level by pooling data for robust effectiveness estimates.

10. What is the formula for sensitivity using a $2\times^2$ diagnostic table?

    $\text{sensitivity}=\frac{TP}{TP+FN}$. This proportion identifies true positives among all individuals with the condition in a $2\times^2$ contingency table.

11. What is the formula for negative likelihood ratio in diagnostic testing?

    $LR^- = \frac{1-\text{sensitivity}}{\text{specificity}}$. This determines how many times more likely a negative result occurs in those without the condition versus with.

12. What is the best evidence design for determining whether a PT intervention causes benefit?

    Randomized controlled trial (RCT). RCTs establish causality in interventions by randomly allocating participants to minimize bias and confounding factors.

13. What is the formula for positive likelihood ratio in diagnostic testing?

    $LR^+ = \frac{\text{sensitivity}}{1-\text{specificity}}$. This calculates how many times more likely a positive result occurs in those with the condition versus without.

14. Which test property is prioritized when the goal is to rule IN a condition with a positive result?

    High specificity (SpPin). High specificity ensures few false positives, making a positive result effective for confirming the condition (SpPin mnemonic).

15. What is the formula for pre-test odds from a pre-test probability $p$?

    $\text{odds}=\frac{p}{1-p}$. This converts probability to odds, facilitating multiplication with likelihood ratios for Bayesian updating.

16. What is the formula for updating odds after a diagnostic test using a likelihood ratio?

    $\text{post-test odds}=\text{pre-test odds}\times LR$. This Bayesian approach updates pre-test odds by multiplying with the test's likelihood ratio to obtain post-test odds.

17. What is the formula for positive predictive value (PPV) using a $2\times^2$ table?

    $PPV=\frac{TP}{TP+FP}$. This measures the probability that a positive test result truly indicates the presence of the condition.

18. What is the primary purpose of a clinical prediction rule in PT differential diagnosis?

    Estimate diagnosis or prognosis to guide decisions. Clinical prediction rules combine variables to estimate probabilities, aiding in differential diagnosis and treatment planning.

19. Which magnitude of $LR^-$ is generally considered strong evidence to rule out a diagnosis?

    $LR^-\le 0.1$. A value of 0.1 or less substantially decreases post-test probability, providing strong evidence against the diagnosis.

20. Which analysis strategy preserves randomization by keeping participants in assigned groups?

    Intention-to-treat analysis. It analyzes all participants as originally randomized, maintaining group equivalence despite dropouts or non-compliance.

21. What statistic best expresses the relative likelihood of disease after a positive test result?

    Positive likelihood ratio ($LR^+$). It quantifies how much a positive test increases the probability of disease compared to no disease.

22. What is the formula to convert post-test odds to post-test probability?

    $p=\frac{\text{odds}}{1+\text{odds}}$. This back-converts odds to probability after applying a likelihood ratio to pre-test odds.

23. What does a narrow $95\%$ confidence interval around an effect estimate indicate?

    Greater precision (less random error). A narrow interval suggests the estimate is reliable with minimal variability due to sampling error.

24. Which bias is most reduced by assessor blinding when outcomes are measured?

    Detection (measurement) bias. Blinding assessors minimizes subjective influence on outcome measurements, ensuring objective data collection.

25. Which test property is prioritized when the goal is to rule OUT a condition with a negative result?

    High sensitivity (SnNout). High sensitivity ensures few false negatives, making a negative result effective for excluding the condition (SnNout mnemonic).

Study these flashcards