Are our PT Programs Good Enough?

Educating PT students is a tall order. The broad scope of PT demands a focus spread across Cardiorespiratory, Neurological, and Orthopedic practice domains. Not to mention some forays into foundational topics such as anatomy, pathology, exercise science, and development.

It’s a big task.

A big task, however, is not an excuse for complacency. Universities are our bastions of Science and should be champions of Evidence Based Practice. PT students should reasonably expect to graduate with strong clinical reasoning skills, and for that to be possible, they need a firm grasp of the literature and an understanding of what they can stand on. Which, in the MSK domain, means a solid familiarity with rehabilitation sciences.

How many students know what sensitivity and specificity mean and what the psychometric properties are of the tests that they commonly use? How many know what the mechanisms of manual therapy are? What about pain sciences? Do they know what a clinical prediction rule is? Test item clusters? Are they familiar with the concept of biopsychosocial?

My observation is that these crucial building blocks of clinical reasoning are often absent, at least in my corner of Canada. Even post-grad courses, in some cases, leave this stuff out. It’s a shame that students instead direct their energies toward short wave diathermy, ultrasound, cross frictions, and how to do stability testing of the Lumbar and Thoracic Spine.

Show me one shred of evidence for the validity of stability testing the thoracic spine and I’ll drink a bottle of snake oil.

With that out of my system, let’s zero in on the apropos issue of spinal manipulation as an entry-level skill—it is currently considered to be an ‘advanced practice skill’. In fact, it’s actively warned against in many entry-level Physio schools in Canada. So today, let’s examine this issue through the lens of science.

First, let’s acknowledge the efficacy and effectiveness of spinal manipulation particularly for low back complaints. It’s a technique that is well supported by randomized clinical trials and its presence in clinical practice guidelines. We see a lot of support also for thoracic thrust techniques, particularly for populations with shoulder impingement and neck pain (ref and ref). As a technique spinal thrust manipulations have proven to not only improve outcomes but also lead to cost savings. Ref

And it’s not exactly a once in a while kind of treatment. Studies show that 35-48% of patients referred to physiotherapy for low back pain respond to spinal manipulation. And low back pain is close to 50% of what we see in MSK private practice. This is a 'bread and butter' skill that should be used every day! As a caveat, and to avoid any stickling, I'm not saying that manipulations are uniquely special. We know that 'interesting' mobs (non-neutral range) can have a similar effect in many cases (ref). What I'm getting at is that manipulations are effective, quick, and a great tool for MSK PTs.

With this last statement, I know I'm bumping into a larger professional issue: despite all the evidence that manipulations are clinically- and cost-effective, studies and surveys show that it is vastly under-utilized within the profession. This is especially concerning when we observe what PTs are doing instead. Jette and Delitto (ref) looked at 1279 episodes of LBP care. They found that manipulations were used 3.7% of the time in ‘initial stage care’. The alarming finding is that heat was used 79.1% of the time; electrical stimulation was used 39.1% of the time, and flexibility exercises were used 75% of the time in this same period. These latter interventions all share one thing: none of them are supported by evidence.

Surely universities won’t teach stuff that isn’t supported by the evidence, or at least they’ll teach it with full disclosure along side the things that are supported. Right...? Wrong!

Li and Bombardier (ref) add another dimension to this apparent preference for unsupported treatments. They investigated the treatment beliefs of Canadian Physiotherapists. They asked 569 physical therapists if they thought manipulations were effective for LBP. Only 30% of the PTs responded yes! Other interventions (that lacked supporting evidence) did better: ice (82%), heat (66%), electrical stimulation (53%), and mechanical traction (36%) . This demonstrates an obvious lack of knowledge of current evidence by our profession at large.

The most concerning issue is that we may be passing on outdated and unsupported practices onto new PTs. Surely universities won’t teach stuff that isn’t supported by the evidence, or at least they’ll teach it with full disclosure along side the things that are supported. Right?


Unlike 93% of entry-level Universities in the USA, Canadian PT programs typically don’t teach any spinal thrust manipulations. Our Canadian bastions of Science instead dedicate class time to thermo and electro-therapeutic placebo machines like IFC and short wave diathermy.  

Why is this important? Well, entry-level programs shape our foundation for future clinical reasoning and decision-making. This is the time when habits get established. My worry is that we’re leaving out important skills and information that are necessary to deliver evidence-based practice and effective clinical reasoning and replacing it with unsupported reasoning models and therapeutic techniques.

Turner and colleagues support me here. They found that the main rationale that therapists had for decisions in practice was based on what they were taught in their initial training.

This means that the optimal time to influence clinical decision making and evidence-based practice habits is during the initial professional degree program. With this in mind, the Manipulation Education Manual for Physical Therapists Professional Degree Programs (MEM) was developed to support the ongoing efforts in PT education. The manual facilitates appropriate, evidence-based instruction in thrust manipulation.

Maybe they haven’t heard about this yet in Canada.

Let’s change tracks and address a few common arguments against teaching entry level students manipulations:

  1. Students are too novice to learn this stuff
  2. Students can’t be specific enough
  3. Thrust manipulations are just unsafe


These studies (ref, ref, ref) demonstrate that there is no difference between novice and experienced clinicians with regard to biomechanical parameters and technique for spinal thrust manipulations. Indeed, in some of these trials, the novices showed a higher interrater reliability for manual therapy examination and diagnosis than the experienced clinicians.

Furthermore, Whitman and colleagues (ref) found that a therapist's experience with manual therapy did not affect patient outcomes with manipulation. Remarkably, this study showed a trend for greater functional improvements in patients manipulated by therapists with less experience. The take home here is that novice clinicians were given only 15 minutes to learn the technique, and they got slightly better outcomes! So no excuses about there not being enough time to teach this stuff.

But wait, there's no way that novices can identify all the biomechanical faults. Can they? Well no, they can't. But there's a rub: experts can't either. The tests proposed to identify most biomechanical lesions are of questionable reliability and validity. Have a look at these studies, which represent a fraction of the literature challenging these patho-anatomical tests: (ref, ref, and ref). Also check out my previous blogs on why Shoulder Special Tests are Nothing Special and my exposé on a more pragmatic approach: No Shells, Just Pearls.

If you're still a believer that we can achieve segmental specificity, have a read of these:  (Refref, and ref). "Our ability to actually produce cavitation in a targeted joint with non-maitland techniques has failed scientific validation." 

And even if we were able to be specific, there is no support that we need to achieve directional specificity without manual therapy interventions; the effects of MT are nonspecific! “There is reason to question the very existence of a joint-specific, mono-segmental lesion used to guide the application of specific thrust interventions as we are still unable to reliably determine its presence and characteristics.” (Ref). There has been a lot of research about the mechanisms of manual therapy, and the biomechanical model has proven again and again to be inaccurate. (Ref).

Let’s let that horse rest, and talk about the risk of these thrust manipulation techniques.

First, I'll flat out say that there is no evidence of increased risk of harm to patients manipulated by a novice compared with an experienced PT. In fact, this study shows that student’s taught within an evidence-based framework use manipulations more appropriately than experienced clinicians. These well-taught students are more compliant with best practice guidelines and are, as such, arguably lower risk than experienced clinicians.  

When looking at the thoracic spine, one article investigated the risk of rib fractures with A-P directed thrust manipulations (ref). They concluded that the compressive forces used for a thrust manipulation was 22.7% of the compression required to have a 10% risk of a minor severity rib injury. The risk was non-existent especially when contraindications and precautions are there to guide.

It's important to put risk in context with other competing interventions, i.e., if we used NSAIDs or hot packs. 

Shekels and colleagues found that the relative risk of a serious adverse event from a Spinal manipulation to the lumbar spine is 1/100,000,000, other studies have it much higher 1/10,000,000. Compare this to vigorous exercise, which resulted in 1/1,500,000 deaths. So we're 70-7x more likely to kill someone with vigorous exercise than have a serious complication with lumbar manipulations. The incidence of serious adverse events due to hot packs is much higher still. 

John Childs led an interesting study that demonstrated that patients with non-specific LBP who were not manipulated were placed at higher risk for pain and disability. He concluded that it is riskier to not manipulate than to manipulate. I like how he turned the subject on its head; this is an important way to consider the issue.

My hope is that we get some light shining on how our entry level programs are doing things. I respect and value the task of educating students, but I'm convinced that complacency has no place at graduate level university programs. Let's eradicate unsupported paradigms, useless assessment / treatment techniques, and falsified clinical reasoning strategies, and replace it all with evidence, critical thinking, and supported techniques.

'Evidence Based Practice' is the CPA’s first value, it’s past time that everyone follows that lead.