7.5 Evidence-Based Practice for PTAs
Key Takeaways
- The research evidence hierarchy ranks systematic reviews/meta-analyses highest, followed by randomized controlled trials (RCTs), cohort studies, case-control studies, case series, and expert opinion.
- A p-value below 0.05 indicates statistical significance — the result is unlikely to be due to chance alone — but does not measure effect size.
- The minimal clinically important difference (MCID) is the smallest change a patient perceives as meaningful, distinct from statistical significance.
- SnNOut: a sensitive test that is Negative rules a condition OUT; SpPIn: a specific test that is Positive rules a condition IN.
- A PTA applies evidence by faithfully carrying out the PT-selected, evidence-based protocol within the plan of care — not by independently changing the protocol based on a personal literature review.
The Evidence Hierarchy
Evidence-based practice (EBP) integrates three things: the best available research, clinical expertise, and patient values and circumstances. Study designs are ranked by how well they control bias, and the NPTE-PTA expects you to know the order.
| Rank | Study Design | Strength |
|---|---|---|
| 1 | Systematic review / meta-analysis | Strongest — pools and synthesizes many studies |
| 2 | Randomized controlled trial (RCT) | Strong — randomization limits selection bias |
| 3 | Cohort study | Moderate — follows groups forward over time |
| 4 | Case-control study | Moderate-low — looks backward from an outcome |
| 5 | Case series / case report | Weak — no comparison group |
| 6 | Expert opinion / narrative review | Weakest — no systematic data |
When two studies conflict, the design higher on this hierarchy generally carries more weight. A single high-quality RCT can outweigh several case reports.
Basic Statistics for the NPTE-PTA
| Term | Meaning |
|---|---|
| p-value | Probability the observed result occurred by chance; p < 0.05 is the conventional threshold for statistical significance |
| MCID (minimal clinically important difference) | Smallest change a patient perceives as meaningful or worthwhile |
| NNT (number needed to treat) | Number of patients who must be treated for one to benefit; a lower NNT is better |
| Sensitivity | Ability of a test to correctly identify patients who HAVE the condition (true positives) |
| Specificity | Ability of a test to correctly identify patients who do NOT have the condition (true negatives) |
| PPV (positive predictive value) | Probability a patient with a positive test truly has the condition |
| NPV (negative predictive value) | Probability a patient with a negative test truly does not have the condition |
| Reliability | Consistency/reproducibility of a measure (e.g., test-retest, inter-rater) |
| Validity | Whether a test measures what it claims to measure |
Statistical vs. clinical significance: a study can show a statistically significant change (p < 0.05) that is still smaller than the MCID — the change is real but too small for the patient to notice. The reverse is also possible: a large, clinically meaningful change in a small underpowered study may not reach p < 0.05. Both dimensions matter.
Sensitivity vs. specificity memory hooks: SnNOut — a highly Sensitive test, when Negative, helps rule the condition OUT. SpPIn — a highly Specific test, when Positive, helps rule the condition IN.
How a PTA Applies Evidence
This is the exam's key distinction, and it recurs in many ethics and supervision items as well. A PTA uses evidence within the plan of care (POC), not to override it:
- The PT evaluates the patient and selects an evidence-based intervention and protocol as part of the POC.
- The PTA implements that protocol faithfully — correct dosage, parameters, and progression — collects objective data on the patient's response, and reports findings back to the PT.
- A PTA who reads a journal article suggesting a different approach should discuss it with the supervising PT. The PTA does NOT unilaterally substitute a new protocol, because changing the plan requires evaluative judgment reserved for the PT.
- The PTA strengthens evidence-based care by recognizing when reported progress is genuinely meaningful (exceeds the MCID), by flagging when a patient is not responding as expected, and by explaining to patients why the prescribed program is appropriate — all without stepping outside the PTA scope of practice.
Think of EBP for a PTA as quality assurance on delivery and data, not as authority to redesign the plan.
Outcome Measures and Reliability vs. Validity
PTAs administer standardized outcome measures within the POC and must understand their properties:
- Reliability is consistency. Test-retest reliability is consistency across time; inter-rater reliability is consistency between different testers; intra-rater is consistency for the same tester. The intraclass correlation coefficient (ICC) quantifies it — closer to 1.0 is better.
- Validity is accuracy — whether the tool measures what it claims. A measure can be highly reliable (consistent) yet invalid (consistently measuring the wrong thing), so reliability is necessary but not sufficient for validity.
- Responsiveness is the ability to detect change over time, which is why the MCID matters when interpreting a patient's progress.
Common PTA-relevant measures include the Berg Balance Scale, Timed Up and Go (TUG), 6-Minute Walk Test, Numeric Pain Rating Scale, and goniometry for range of motion. Each has published MCID and minimal detectable change (MDC) values; a change must exceed the MDC to be confident it is real and not measurement error, and exceed the MCID to be meaningful to the patient.
Interpreting a 2x2 Table
Diagnostic-accuracy items sometimes show a contingency table. Anchor the four cells:
| Disease Present | Disease Absent | |
|---|---|---|
| Test Positive | True Positive (TP) | False Positive (FP) |
| Test Negative | False Negative (FN) | True Negative (TN) |
- Sensitivity = TP / (TP + FN) — of those who truly have the condition, the fraction the test catches.
- Specificity = TN / (TN + FP) — of those without the condition, the fraction the test correctly clears.
- PPV = TP / (TP + FP) and NPV = TN / (TN + FN) — these depend on disease prevalence, while sensitivity and specificity are properties of the test itself.
Worked Scenario and Traps
A PT selects a balance-training protocol supported by a systematic review. The PTA delivers it, and after two weeks the patient's Berg score improves by 3 points, but the published MCID for that population is 5 points. Correct interpretation: the change has not yet exceeded the MCID, so although the trend is positive, the PTA should continue the prescribed protocol and report the objective progress to the PT rather than declare the goal met.
Traps to avoid: (1) treating a low p-value as proof of a large or meaningful effect — significance is not effect size; (2) overruling a higher-evidence study with a single case report; (3) a PTA changing parameters because of a journal article — that is a PT decision; (4) confusing reliability with validity. The PTA's evidence role is faithful delivery, accurate measurement, and clear communication — never independent redesign of the plan of care.
A PTA reads a recent randomized controlled trial (RCT) suggesting a different exercise protocol may work better than the one the supervising PT selected. What is the appropriate action?
A study reports that a treatment produced a change in outcome scores with a p-value of 0.03, but the change was smaller than the validated minimal clinically important difference (MCID). How should this be interpreted?
A screening test for a serious condition has very high sensitivity. A patient's result is negative. What does this best support?
Match each biostatistics term to its definition.
Match each item on the left with the correct item on the right