Evidence-Based
Evaluation
Use current best evidence to support evaluation, differential diagnosis, and prognostic decisions in physical therapy practice.
What Is Evidence-Based Practice?
Evidence-based practice (EBP) is the integration of three pillars to make sound clinical decisions. No single source dominates: each informs and constrains the others.
The Three Pillars Together
Why Evidence Matters in Evaluation
Evaluation is the clinician's judgment based on examination data. Using evidence here improves diagnostic accuracy, sets realistic prognoses, and protects the patient from low-value or harmful interventions.
Key Terms: Sensitivity & Specificity
Core Rule: Likelihood Ratios
Likelihood ratios (LRs) tell you how much a test result shifts the probability of a condition. They combine sensitivity and specificity into one usable number.
A large \( \text{LR}+ \) (greater than 10) strongly increases probability of disease. A small \( \text{LR}- \) (less than 0.1) strongly decreases it.
How LRs Shift Probability
Distinction: Diagnostic vs Prognostic Evidence
The Evidence Hierarchy
How to Apply Evidence Step by Step
Example 1: Ruling Out a Fracture
A patient presents with ankle pain. The Ottawa Ankle Rules have sensitivity \( \text{Sn} \approx 0.98 \). A patient meets none of the criteria (a negative result).
This evidence-based screen reduces unnecessary imaging while keeping miss rates very low.
Example 2: Calculating Post-Test Odds
A cluster test for cervical radiculopathy has \( \text{LR}+ = 6.1 \). The pre-test probability is estimated at \( 30\% \), giving pre-test odds of about \( 0.43 \).
A positive cluster more than doubles the likelihood of radiculopathy, strengthening the diagnostic decision.
Example 3: Prognostic Decision-Making
A clinical prediction rule for low back pain identifies patients likely to respond to manipulation. Your patient meets 4 of 5 criteria.
Common Mistakes
Summary: Test Metrics Reference
| Best For | Rule of Thumb | |
|---|---|---|
| High Sensitivity | Ruling out | SnNout |
| High Specificity | Ruling in | SpPin |
| \( \text{LR}+ \) > 10 | Confirming | Large prob. rise |
| \( \text{LR}- \) < 0.1 | Excluding | Large prob. drop |
Question 1 of 5
A test with very high sensitivity returns a negative result. What can you conclude?
Question 2 of 5
Which value indicates a test result that strongly increases the probability of disease?
Question 3 of 5
Which combination represents true evidence-based practice?
Question 4 of 5
Which study design generally sits at the top of the evidence hierarchy for therapy questions?
Question 5 of 5
A clinical prediction rule forecasts likely recovery. This is primarily an example of which evidence type?