The Art and Science of Physical Therapy

The 2011 Oxford Debate at the APTA Annual Conference in Washington DC pitted clinical decision rules - algorithms - against clincial intuition. You can see some of the debaters coments here.

I wrote, at the time, that this was a false choice. Algorithms and intuition can BOTH be used by the physical therapist to improve patient outcomes and speed the clinical workflow.

It's not an either-or decision. The question is WHEN to use the algorithm and WHEN to use your intuition.

A recently published systematic review by Henschke to diagnose spinal fractures provides an example of HOW physical therapists can use algorithms and intuition together.


I included Henschke's original decision rule in my new book, Bulletproof Expert Systems: Clinical Decision Support for Physical Therapists in the Outpatient Setting due out January 15th, 2012.

Henschke's new rule was also recently posted to PhysioPedia in a page titled Subjective Exam - Diagnostic Strength by Brian Duffy, Carleen Jogodka, Jeff Ryg, James White of the Evidence in Motion Fellowship program.

Henschke's New Rule to Diagnose Spinal Fractures

Henschke's rule to diagnose spinal fracture in a low risk setting provides physical therapists a unique opportunity to use their clinical intution.

Clinical Decision Rules are usually intended to provide probabilities confirming a diagnosis or predicting an outcome so the physical therapist can make clinical decisions with confidence.

Henschke's rule screens patients for vertebral fractures without the use of expensive and overly sensitive diagnostic imaging. This rule may be employed in two different settings: low risk primary care offices or high risk emergency rooms. The setting determines the pre-test probability, or prevalence.

Here is Henschke's new rule:
  • History of major trauma
  • Pain and tenderness
  • Age < 50 years
  • Female
  • Corticosteroid use
The base rate of vertebral fractures in a population of 1,172 patients accessing primary care for treatment of lower back pain in Sydney, Australia was 0.5%. Primary care in Australia is defined as offices of physicians, physical therapists and chiropractors.

Low Risk Decision Rule
Screening Finding PresentLikelihood of the Diagnosis
1 present1% chance of a spinal fracture
2 present7% chance of a spinal fracture
3 or more present52% chance of a spinal fracture

The base rate of vertebral fractures in patients accessing the emergency room and specialty physicians’ offices for treatment of lower back pain in Sydney, Australia was 3.0%.

High Risk Decision Rule
Screening Finding PresentLikelihood of the Diagnosis
1 present5% chance of a spinal fracture
2 present32% chance of a spinal fracture
3 or more present87% chance of a spinal fracture

The predictive power of the decision rule varies with the setting in which the clinician sees the patient – high risk patients seen in specialty clinics had a higher prevalence of spinal fracture. The new rule is the same in both settings.

The physical therapist's intuition is especially important in the LOW RISK situation when three or more of the subjective variables were present. In this situation, the clinical decision rule returns a probability of 52% favoring the diagnosis of vertebral fracture. The rule, in this situation, barely performs better than chance.

A physical therapist flipping a coin could do just as well in predicting a spinal fracture (~50%).

In this situation, the physical therapist should rely on their clinical intuition. Intuiton might include additional data points from the physical therapy evaluation, including:
  • the patient's history
  • subjective pain complaints
  • physical examination
  • special tests 
  • other pathology screening exams.


Also, a medication list, cognitive status and input from family members could add useful data points that might increase or decrease the probability of a fracture.

This example is meant to demonstrate WHY physical therapist intuition is still important, combined with first-pass screening algorithms that supplement human memory for low-frequency events. In these situations, the use of clinical intuition and experience supplements the algorithmic decision rule.

Henschke's rule to diagnose spinal fracture is a useful algorithm for screening high risk patients in the emergency room.

In the low risk setting, such as an ambulatory PT clinic, the rule requires that the physical therapist remain alert to subtle cues that might affect the diagnosis.

Henschke's rule demonstrates clearly how your diagnosis requires both the art and the science of physical therapy.
Share on :
The Art and Science of Physical Therapy
The Art and Science of Physical Therapy
Reviewed by Merlyn Rosell
Published :
Rating : 4.5