This is the most important skill in evidence-based physiotherapy. Here's the full process — from reading an article to applying it with a real patient:
Reading an Article → Understanding Evidence → Applying in Practice
PART 1: How to Read a Research Article
The Right Reading Order (Don't start with the Introduction)
| Step | Section | Why |
|---|
| 1 | Title + Abstract | Decide if it's relevant in 2 minutes |
| 2 | Methods | Is the study design trustworthy? |
| 3 | Results | What exactly did they find? |
| 4 | Discussion | How do authors interpret it? |
| 5 | Introduction | Background context (read last) |
How to Read Each Section
Abstract
Ask yourself:
- What condition/population was studied?
- What intervention was tested?
- What outcome was measured?
- What was the main finding?
This is the PICO framework — Patient, Intervention, Comparison, Outcome.
Methods — Most Important Section
This tells you if you can trust the results. Ask:
| Question | Why it matters |
|---|
| Was there a control group? | Without it, results may be placebo |
| Was it randomized? | Reduces selection bias |
| Was it blinded? | Single/double blind reduces bias |
| How large was the sample size? | Small samples = unreliable results |
| How long was follow-up? | Short studies miss long-term effects |
| Who were the participants? | Do they match your patients? |
Use the PEDro scale — 11 questions that score study quality from 0–10. Any score ≥6 is considered good quality.
Results — Numbers That Matter
Do NOT just look at whether p < 0.05. Look at:
| Statistic | What to look for |
|---|
| Effect size | How big was the improvement? (Cohen's d, SMD) |
| Mean difference | Did the treatment group improve more than control? |
| MCID | Did the change exceed the Minimum Clinically Important Difference? |
| 95% Confidence Interval | Wide CI = less precise, less reliable |
| NNT (Number Needed to Treat) | How many patients need treatment for 1 to benefit? |
Key point: A result can be statistically significant (p<0.05) but clinically meaningless if the effect size is tiny or doesn't exceed MCID.
Example for stroke:
- Berg Balance Scale MCID = 6 points
- If a study shows 3-point improvement (p=0.04) → statistically significant but NOT clinically meaningful
PART 2: Levels of Evidence — Know What to Trust
HIGHEST
▲
│ Systematic Review + Meta-Analysis
│ Randomized Controlled Trial (RCT)
│ Cohort Study
│ Case-Control Study
│ Case Series / Case Report
▼
LOWEST
In practice:
- Systematic Review = Read this first. It summarizes ALL RCTs on a topic.
- RCT = Best single study design. Look for PEDro score ≥6.
- Case report = Interesting but don't change your practice based on it alone.
PART 3: Critical Appraisal — 5 Questions for Any Article
Before applying anything to a patient, ask:
- Is it valid? — Was the study design rigorous? (Check PEDro score)
- What are the results? — Effect size, MCID, confidence intervals
- Are the results important? — Clinically meaningful, not just statistically significant
- Can I apply this to my patient? — Does the study population match your patient?
- Are all outcomes considered? — Did they measure what matters to the patient (function, QoL, not just a score)?
PART 4: Bridging Evidence to Practice — The Clinical Translation Step
This is where most physiotherapists get stuck. Here's how to do it:
Step 1 — Extract the Key Clinical Message
After reading, write one sentence:
"In patients with [condition], [intervention X] delivered [dose/frequency] improved [outcome] by [amount] compared to [control]."
Example:
"In sub-acute stroke patients, 45 minutes of task-oriented gait training 5×/week improved 10MWT speed by 0.18 m/s compared to conventional therapy."
Step 2 — Match to Your Patient (PICO Check)
Ask:
- P — Is my patient similar to the study population? (age, severity, chronicity, comorbidities)
- I — Can I deliver this intervention in my setting? (equipment, time, skill)
- C — Is the comparison relevant to what I currently do?
- O — Is this outcome meaningful to MY patient's goals?
If the answer is mostly yes → apply it. If not → modify or seek a more applicable study.
Step 3 — Translate Dose & Parameters Precisely
Do not just say "I'll try gait training." Extract the exact protocol:
| Parameter | Extract from the Methods section |
|---|
| Frequency | How many sessions per week? |
| Intensity | What level of effort/load? |
| Time | How long per session? |
| Type | Exactly what was done? |
| Progression | How was it progressed? |
This is the FITT principle — Frequency, Intensity, Time, Type.
Step 4 — Choose an Outcome Measure to Track Progress
Match the outcome measure from the study to your patient:
| Condition | Outcome Measure |
|---|
| Stroke — gait | 10-Meter Walk Test, 6MWT |
| Stroke — balance | Berg Balance Scale |
| Low back pain | PSFS, Oswestry, NPRS |
| Shoulder | DASH, ASES |
| Knee | KOOS, IKDC |
| General function | Barthel Index, FIM |
Measure at baseline, mid-treatment, and discharge. Did you exceed the MCID?
Step 5 — Reflect After Applying
After treating the patient using the evidence-based protocol:
- Did the patient respond as the study predicted?
- If not — why? (patient factors, delivery issues, wrong population match?)
- Write a one-paragraph reflection in your clinical journal
This builds your clinical expertise — the third pillar of EBP alongside research evidence and patient values.
PART 5: The 3 Pillars of EBP — All Must Work Together
BEST
RESEARCH
EVIDENCE
/\
/ \
/ \
/ \
CLINICAL ── PATIENT
EXPERTISE VALUES
& GOALS
Evidence alone is not enough. You must integrate:
- Best research — What does the literature say works?
- Clinical expertise — What does your experience tell you about this specific patient?
- Patient values — What does the patient want, fear, and prioritize?
PART 6: Simple Weekly Workflow
Monday: Find 1 article on your toughest current case (PEDro or PubMed)
Tuesday: Read Abstract + Methods + Results (20 min)
Wednesday: Apply the protocol to that patient
Friday: Note the outcome — did it match the evidence?
Monthly: Bring one case + one article to a journal club or peer discussion
Quick Reference Card
| Step | Action |
|---|
| Find article | PEDro → search condition → filter by systematic review |
| Trust it? | Check PEDro score (≥6 = good) |
| What did it find? | Effect size + MCID (not just p-value) |
| Apply it? | PICO match to your patient |
| Exact dose | Extract FITT from Methods |
| Track progress | Use validated outcome measure |
| Reflect | Did patient respond? Why or why not? |
The mindset shift:
Reading an article is not the end — it's the start of a clinical experiment with your patient. Every patient you apply evidence to is a test of whether that evidence holds in your specific context. Reflection closes the loop.