Physiotherapy special tests for the CPTE

Musculoskeletal conditions make up half of the Canadian Physiotherapy Examination (CPTE): the CPTE Blueprint 2025 gives the musculoskeletal Area of Practice 50 ± 5% of the exam. Assessing those conditions often calls for special tests, and the exam's assessment milestones ask for judgment rather than a list. Milestone A.03 expects you to "justify test selection," and A.04, interpreting findings, asks you "to generate hypotheses and identify both typical and unexpected results." This guide explains how to read a test's accuracy, then sets out the tests the clinical practice guidelines name for each region, from the neck to the ankle, with the sensitivity and specificity they report. Every figure comes from the guideline or review cited, as it stood on October 15, 2026.

The short answer

• Few tests diagnose on their own. A test with high sensitivity is most useful when it is negative, to rule a condition out; one with high specificity is most useful when it is positive, to rule it in.
• Accuracy varies by study. For femoroacetabular impingement, the hip guideline reports the FADIR test's sensitivity as anywhere from 0.08 to 1.00 across 19 studies.
• Combinations change the trade-off. In the meniscal guideline's composite score, more positive findings raise specificity and lower sensitivity.
• Some tests are decision rules, not diagnoses: the Ottawa ankle rules decide whether an X-ray is needed.
• Screen before you test. A special test explores a musculoskeletal hypothesis; it does not rule out serious pathology.

The QualExam lessons on assessing ligament and joint injuries, the cervical spine and the thoracic and lumbar spine cover each region's tests in full, with worked cases, and the course's assessment questions, in the app, name the document behind each answer.

How to read a test's accuracy

Two numbers describe a test against a reference standard, such as imaging or surgery. Sensitivity is the share of people with the condition whom the test picks up. Specificity, as one review puts it, "refers to the ability to effectively identify all those without the disease." The guidelines' shorthand follows from them. In the words of one set of international recommendations, an "Absolute SpPin" is "a diagnostic finding whose specificity is so high that a positive result rules‐in the diagnosis," and an "Absolute SnNout" is one whose "sensitivity is so high that a negative result rules out the diagnosis."

Likelihood ratios turn those numbers into a change in probability, and the Ottawa ankle rules show how. A 2003 systematic review found "The pooled negative likelihood ratios for the ankle and mid-foot were 0.08," and that applying these ratios "to a 15% prevalence of fracture gave a less than 1.4% probability of actual fracture." A negative result moved the probability of a fracture from 15% to under 1.4%. The same test means less when the condition is already unlikely, and more when it is likely, so start from the history: the hypotheses you bring to a test decide what its result can tell you.

Neck

The neck pain guideline's evidence on cervical radiculopathy is the classic example of tests that rule in and tests that rule out: "A positive Spurling test (sensitivity, 0.50; specificity, 0.86-0.93), traction/neck distraction test (sensitivity, 0.44; specificity, 0.90-0.97), and Valsalva test (sensitivity, 0.22; specificity, 0.94) may suggest cervical radiculopathy, while a negative neurodynamic test (sensitivity, 0.17-0.78; specificity, 0.72-0.83) may rule it out."

The Spurling and distraction tests are specific, so a positive result is informative and a negative one is not; the upper limb neurodynamic test is the reverse. The guideline groups them: for neck pain with radiating pain, the "positive test cluster includes upper-limb nerve mobility, Spurling's test, cervical distraction, cervical ROM."

Low back

The 2012 low back pain guideline uses neural tension tests to classify, not to diagnose. For acute low back pain with radiating pain, "Symptoms are reproduced or aggravated with initial to mid-range spinal mobility, lower limb tension/straight leg raising, and/or slump tests," and nerve root signs may be present. A reproduced leg symptom on the straight leg raise or the slump test points to the radiating pain classification; deficits in sensation, strength or reflexes point to nerve root involvement.

Shoulder

The evidence for shoulder tests is weaker. A 2024 meta-analysis of tests for rotator cuff tears and subacromial impingement syndrome, covering 20 studies and 3,438 patients, concluded: "We present a large body of low-quality evidence for the diagnostic accuracies of physical examination tests for the identification of both RCT and SIS." The studies it included examined the Neer, Hawkins and Jobe tests, the painful arc, the drop arm sign, the lift-off, bear hug and Yergason's tests, and the external and internal rotation lag signs.

Two results stand out. For a rotator cuff tear, "the External Rotation Lag Sign at 90 Degrees having the highest magnitude of significance (DOR, 12.70; 95% CI, 3.68 – 43.86; P < .0001)," and the authors found new data on the external and internal rotation lag signs "for ruling in tears of the rotator cuff." For subacromial impingement, Yergason's test had the highest diagnostic odds ratio, 4.71. A diagnostic odds ratio summarizes how well a test separates people with and without a condition in one number, and its wide confidence intervals here are a reminder to combine the tests with the history and a broader examination.

Elbow

For lateral elbow tendinopathy, the guideline notes: "There are 3 common special tests (Maudsley's, Cozen's, and Mill's stretch) that are used to arrive at a clinical diagnosis of LET." It reports a sensitivity of 84% for Cozen's test in one study and 91% (95% CI: 81%, 96%) in another, so a negative test makes the condition less likely.

Wrist and hand

The 2026 carpal tunnel syndrome guideline recommends three tests and one combination: "Clinicians should use the Phalen test, Tinel sign, and the carpal compression (Durkan) test to help diagnose those with suspected CTS," and "Clinicians should use the Carpal Tunnel Syndrome-6 (CTS-6) to help diagnose those with suspected CTS." How a test is done changes what it shows. In one study of a timed Phalen test, "a positive Phalen test in ≤10 seconds produced high specificity (96.8%)," but it "demonstrated low sensitivity (13.9%)."

Hip

The nonarthritic hip pain guideline is blunt about the variation: "The diagnostic accuracy for 9 special tests for FAIS varied greatly across studies," with the FADIR test's sensitivity ranging from 0.08 to 1.00 and its specificity from 0.03 to 1.0 across 19 studies, and the FABER test's specificity from 0.18 to 1.00 across 8. So the guideline uses the tests with the history and imaging: it classifies nonarthritic hip joint pain from pain "reproduced with the hip flexion, adduction, internal rotation (FADIR) test or the hip flexion, abduction, external rotation (FABER) test, along with consistent imaging findings." And it notes a role for the FADIR: "Because of the high sensitivity but lower specificity, it has been suggested that the FADIR be used as a screening tool for FAIS."

Knee

The knee guidelines give the clearest numbers. For an anterior cruciate ligament tear, the knee ligament guideline lists a positive Lachman test, with a soft end feel or increased anterior tibial translation, at a sensitivity of 85% and a specificity of 94%, and a positive pivot shift test at a sensitivity of 24% and a specificity of 98%. The Lachman test is useful either way; a positive pivot shift is strong evidence of a tear, but a negative one rules out very little.

For a meniscal tear, the meniscal guideline's summary gives:

It then combines five findings (a history of catching or locking, pain with forced hyperextension, pain with maximum passive knee flexion, joint-line tenderness, and McMurray's manoeuvre) into a Meniscal Pathology Composite Score. With more than three positive findings, sensitivity is 30.8% and specificity 90.2%; with more than one, 76.6% and 43.1%. Requiring more positives makes a positive score more convincing and a negative one less reassuring. The same guideline also reports a study in which the Thessaly test had a sensitivity of 0.66 and a specificity of 0.39 against MRI, which is why no single figure should be learned as fixed.

Ankle and foot

• Lateral ankle sprain: "Clinicians should use special tests, including the reverse anterolateral drawer test and anterolateral talar palpation in addition to the traditional anterior drawer test, and a thorough history and physical examination to aid in the diagnosis of a LAS."
• A possible fracture: the ankle guideline says the Ottawa ankle rules and the Bernese ankle rules "should be used to determine whether a radiograph is required to rule out a fracture of the ankle and/or foot." The 2003 review concluded that the Ottawa rules have "a sensitivity of almost 100% and a modest specificity, and its use should reduce the number of unnecessary radiographs by 30-40%."
• Midportion Achilles tendinopathy: "In addition to the arc sign and Royal London Hospital test clinicians can use a subjective report of pain located 2 to 6 cm proximal to the Achilles tendon insertion that began gradually and pain with palpation of the midportion of the tendon to diagnose midportion Achilles tendinopathy."
• Heel pain and plantar fasciitis: the guideline lists a positive windlass test among the findings that support the diagnosis, with negative tarsal tunnel tests.

Using special tests well

• Start from the history. Choose the tests that can confirm or exclude your leading hypotheses, rather than every test for the region.
• Know what each result can do. A negative highly sensitive test, such as the Lachman or the Ottawa rules, lowers the probability of the condition; a positive highly specific one, such as the pivot shift or a timed Phalen test, raises it.
• Combine findings. Clusters and composite scores, like the cervical radiculopathy cluster and the meniscal composite, change sensitivity and specificity together.
• Remember the reference standard and the population. Each figure was measured against imaging or surgery in a particular group of people.
• Screen for serious pathology first. Our guide to physiotherapy red flags covers the findings that need referral before any special test, and our guide to physiotherapy outcome measures covers the measures that track change once you treat.

Questions people ask

Which test is most accurate for an anterior cruciate ligament tear? In the knee ligament guideline, the Lachman test has the best balance, with a sensitivity of 85% and a specificity of 94%. The pivot shift is more specific, at 98%, but far less sensitive, at 24%, so it confirms a tear better than it excludes one.

Do special tests replace imaging? No. The figures above were measured against a reference standard such as MRI, and the guidelines combine tests with the history and, for the hip, with imaging. For a possible ankle fracture, the Ottawa ankle rules decide whether imaging is needed at all.

How many tests should I do in a case? As many as answer your hypotheses, and no more. In an oral case, saying why you chose a test, and what a positive or negative result would change, earns more than a long list.

Are the numbers the same in every study? No. The hip guideline's ranges for the FADIR test, and the meniscal guideline's two very different results for the Thessaly test, show how much accuracy depends on the population, the way the test is done and the reference standard.