I see patterns

I see patterns, quick flashback to the Sixth Sense.


“Nonspecific LBP accounts for the great majority of cases of LBP and is defined as LBP for which there is no identifiable cause (e.g, injury or disease). As a result, treatment recommendations commonly involve a one-size-fits-all approach.”

This is reality. When someone has back pain, it is a guess and a poor one at that as to what is the cause of the back pain. Herniated discs? Sure. Arthritis? Sure, why not. Spinal stenosis? Must be. Cancer? Naw, this one we could rule in or out with imaging. The sinister (read really bad) stuff can be picked up through imaging and is assumed to be the cause of pain. What else is out there? Lumbago…WTF is this about? My favorite is back pain. For real, this is how it works. The patient goes to the doctor with a complaint of back pain and after the end of the session, the doctor says…You have back pain. Here’s your script for back pain. See me in a few weeks.

The problem when we can’t identify different causes of back pain, then all back pain is treated via a “shake and bake” or cookie cutter approach. Is Suzy’s back pain the same as Johnny’s, probably not since the symptoms aren’t even in the same location, but it is still coming from the back so it must be treated the same way. There’s a reason that we as the industry of healthcare have failed in treating back pain…we can’t even define it.


“The current treatment classification system (ie, a small group [5%-10%] of patients with identified specific pathology versus the large group [90% -95%] with nonspecific LBP) is clearly not working well.”

Have you seen the numbers?! Not working well is an understatement. Here are some scary stats. The 5-10% that physicians can diagnose are those sinister (read really bad) problems.

“Subgrouping patients in LBP does not need to be complex or difficult”

Everyone subgroups patients. Tony Delitto has stated in an article (It’s late and I don’t want to go find it so trust me…I’m a professional) that everyone classifies patients, but the classifcation system may be very rudimentary. For instance, if someone comes in with a history of back pain and has failed at therapy elsewhere, we would say that this person may fail again. This is a way of classifying, albeit not a good one, but one way. There are methods of classifying back pain (don’t see this as diagnosing) based on signs and symptoms and response to movement or other interventions. This is a slightly more sophisticated way. There are methods that have withstood the rigor of research and demonstrate moderate reliability in the assessment of back pain.


“A good example in the LBP field is the STarT Back trial that used a simple prognostic tool (9 questions only) to match patients to treatment packages appropriate for them.”

I was fortunate enough to hear Nadine Foster, one of the authors of the original study, speak at a spine conference in 2013. The questionnaire can help clinicians, especially the primary care coordinator (Physician Assistant, Primary care physician, orthopedist, Advance Nurse Practitioner) determine if the patient may improve without treatment or if PT could be beneficial. The final category that a patient could be classified into is the inclusion of physical therapy with the addition of a psychosocial approach to pain.


“Clinicians are usually favorable to the idea of individualized treatments for nonspecific LBP.”

If all back pain were created equal, then I’d be in favor for all treatments being equal. When a patient comes in looking crooked with 9/10 pain, then that patient should not receive the same treatment as someone that has 1/10 pain and is looking to return to sports. Different presentations call for different solutions. There is an excellent book out there for patients and insurance companies called: Rapidly Reversible Low Back Pain by an orthopedic surgeon. He follows the thought and ideas of Robin McKenzie.

“Put simply, if there is a subgroup that does well, it must be balanced by a subgroup that does poorly.”

This research is out there, but because it doesn’t meet the stringent standards of most research studies, it is frowned upon. The problem with the study is that the authors of the study aren’t blinded to the treatments and patient classification. This means that the authors could be biased in one way or another. Aside from this, the study is a legitimate study assessing varying treatment for low back pain. There was one group that did very well and one group that did poorly. One group was in the middle of the two, but leaned more towards poor than well. Check out the study from Audrey Long

“Two aspects of human nature that could explain this situation (treatment effect) are that we tend to see patterns where none exist (patternicity) and that we presume we have more control over events than we truly do (illusion of control).”

This is great stuff. I actually printed off the articles so that I could read them later. I’d love to believe that this isn’t me…but wouldn’t everyone. I’d love to believe that I actually see dead people…I mean patterns and no, not the patterns that people create when they see a shadow and believe it’s a ghost. It does intrigue me though to learn more about pattern recognition.

“…we must conclude that in general, the current research initiatives and achievement in this field are far from optimal and not yet ready to be implemented in clinical practice.”

I wish I could agree with this, but then we are treating all patients the same. If we can’t give individualized instruction to each patient, then it doesn’t matter who the patient sees for their problem. It doesn’t matter that one person’s back pain started 2 years ago and hasn’t subsided or that another’s started this week and is expected to improve with time. Both patient’s would get the same treatment approach if we can’t classify.




Wait…PT’s perform manipulations?!

Wait…PT’s perform manipulations?!


“Without the ability to match patients to specific interventions, clinicians are left without evidence or guidance for their decision-making”


This couldn’t be truer. If we believe that all patients with back pain are the same, then we will give all patients the same treatment. If not all patients respond to the same treatment, then we can say that not all back pain is the same. We have to be able to classify which patients are most likely to respond to a specific treatment; otherwise we are just throwing spaghetti at a wall and hoping that some sticks. When a patient walks into the clinic, I am forming hypotheses as soon as I see the patient get out of the chair in the waiting room. By watching a patient move from the chair to walking and from walking to sitting, we can start to assess pain response (facial expressions) and movement quality (upright versus bent forward or sideways in addition to stride length of the legs and how much rotation is happening with arm swing). We can also have a short chat with the patient to determine how the patient describes their symptoms. Some patients are okay with waiting until we get to the private area before telling their story and others just want to start unloading their story before I have pen to paper to write things down. These are all of the actions that I take into consideration before we even get to the room to assess the patient.


“Identifying methods for classifying patients with LBP has been identified as an important research priority”


Why do most things matter…MONEY! We as a country lose almost $100 billion per year on back pain. This is a lot of money. If we were to put in a dollar every second to pay for this, it would take 31 years to equal $1 billion! As you can see, LBP is an ailment that we have to figure out in order to keep healthcare solvent.


“The purpose of this study was to develop a clinical prediction rule for identifying patients with LBP likely to respond favorably to a specific manipulation technique.”


This is a derivation study, the first step of trying to come up with a clinical prediction rule. One must understand CPR’s prior to reading and implementing the research. Here is a quick link that has to do with the types of CPR (clinical prediction rules). Also, there is much controversy surrounding CPR’s from people such as Dr. Chad Cook, who I highly respect. I don’t know if I would go as far as he does in saying that Clinical Prediction Rules are dead, but they do have to be read thoroughly and criticized. They also have to be validated and placed into an environment in which they can be utilized in order to have an environmental impact. This has been done with diagnostic CPR’s such as the ankle or knee rules.


Me personally, I don’t believe that we should give up on a quest to determine which intervention works best for a specific set of the population. We can provide value to our customers by providing the best evidence based treatments that we have available. To kill off a method of prescribing treatment limits a therapist’s ability to confidently provide treatment.

“…patients with LBP at two outpatient facilities: Brooke Army Medical Center and Wilford Hall Air Force Medical Center…between the ages of 18 and 60 years…baseline Oswestry disability score had to be at least 30%”


This study is highly specific to a military crowd, with an average age of younger than 40. Now if this is not the patient that is being treated in my clinic, it is difficult for me to make the correlation from one population to another. The only thing that we can say for certain about the results of this study is that is pertains to the population that was involved in the study. The baseline Oswestry disability score (for more on the Oswestry see this link.


“After the manipulation, the therapist noted whether a cavitation was heard or felt by the therapist or patient.”


The cavitation is the audible pop that people think of when getting a fast manipulation. This is similar to popping your knuckles. This pop is not needed for a manipulation to occur, as the movement and speed instead of by the noise that occurs define the manipulation.


“A maximum of two attempts per side was permitted.”


This doesn’t make sense to me to perform multiple manipulations directed at the same region. The authors noted that if no cavitation was produced that another manipulation would be performed up to four maximum manipulations. We just covered that an audible pop is not needed, so I am unsure why two were allowed for the patient. Let’s just assume that a patient gets better from the manipulation, was it one manipulation or two manipulations that improved the patient? Is it possible that a patient could get better with one, but then get worse with the second…even though a cavitation is heard? There are too many variables that start to play into this study.   This is the landmark study for giving the prediction recommendations for spinal manipulation in PT. Which brings us to the next point.


“Two additional treatment components were included: 1) instruction in a supine pelvic tilt range of motion exercise…and 2)instruction to maintain usual activity level within the limits of pain.”
Now we have 3 possible variables introduced into this science experiment. Any scientist would look at this and say that there are too many independent variables, which can affect the outcome. The first is obviously the manipulation. The second is the pelvic tilt. The third is time.


“The mean OSW (Oswestry Disability Index) score at baseline was 42.4+/-11.7, and at study conclusion was 25.1 +/- 13.9.”


This means that the scores initially ranged from 31-53 and the final scores ranged from 11-39. A change of 10 can be considered significant, so there was a significant change overall for the better.


“Thirty-two patients (45%) were classified as treatment successes, and 39 (55%) were nonsuccesses”


A majority of patients didn’t respond to the intervention(s), but it was close to a coin flip. This indicates that if we manipulated everyone that came through the door, we would have a success (about 50% improvement in ability) in about half of the patients. This isn’t a bad ratio if it is only done in one visit.


“…duration of symptoms < 16 days, at least one hip with >35 degrees of internal rotation, hypomobility with lumbar spring testing, FABQ work subscale score <19, and no symptoms distal to the knee…were used to form the clinical prediction rule.”


Here it is! All students are expected to memorize this by the time that they graduate from PT school. All PT’s (at least those that work on backs) are expected to know these criteria for manipulation. There are of course some that will state that CPR’s aren’t very effective in practice, but this rule seems to have stood the test of time over many studies.


“…a subject with four or more variables present at baseline increases his or her probability of success with manipulation from 45% to 95%. If the criteria were changed to three of more variables present, the probability of success was only increased to 68%”


WHAAAT!? If someone has 4 of the 5 guidelines from above, the success was 95%! This is yuge. I’ll take those odds of success to the tables any day of the week. Now with this said, I have manipulated very few patients. Those that I have manipulated had immediate positive results and the pain was abolished…didn’t return upon follow-up over the course of 2 weeks. I may not be manipulating as many patients as I could, but I also give the patient the opportunity to independently manage and abolish before attempting to perform a manipulation. It’s a theory from another spine management system.


“In the present study, only one manipulation technique was used, and it is unknown whether other techniques would provide similar results.”


This is also very important to state. There was little research regarding manipulations in the physical therapy research. It must be said that not all manipulations are created equal and that performing a different technique may not have the same result. It may be better or worse. We can only extrapolate this study’s results to those that would match the type of patient treated in this study and the manipulation performed in this study.






Flynn T, Fritz J, Whitman J et al. A Clinical Prediction Rule for Classifying Patients with Low Back Pain Who Demonstrate Short-Term Improvement With Spinal Manipulation. Spine. 2002;27(24):2835-2843.

Call a spade a spade

  1. “Although numerous propositions have been put forward in the literature about how we might usefully subclassify low back pain (LBP), we must first consider the potential utility and futility of such aspirations and ask, “Will it change the outcomes of patients?”


This first statement in the paper is great. All therapist classify patients using either a sophisticated method (which will be spoken of in this paper) or a method that lacks sophistication (a patient’s education level, income level, etc). The big question that we have to ask is “does any of it really matter”.


  1. “Within this arena, there are two schools of though-nominalist and essentialist. Nominalists define a disease by its symptom profile (CLBP = back pain of duration > 3 months). Essentialists state that each specific disease has an underlying pathophysiology, implying treatment of the disease requires treatment of the pathology”


This is fun for me to read. I never though of it this way, but I guess that I would be a nominalist in most cases. Rarely do I believe that the underlying pathophysiology must be treated in order to resolve symptoms. Let me give you an example. For patients that have degenerative disc disease (this is a very common diagnosis in the clinic and most will have this over the course of the lifetime) there is nothing that I will do to regenerate the disc, but I may be able to teach the patient how to either shut off the pain or manage the pain. This would be the nominalist in me. The essentialist in me has another example, which is also a real example in the clinic. There was a patient coming to therapy for treatment of his shoulder. In the process of treating the shoulder he developed back pain (not while in the clinic with me). Anyway, he neglected to tell me about the back pain, but later in the course of care (all within a couple of weeks) went to an urologist for urinary issues. He never told his urologist about the back pain and was advised to use a catheter to urinate! Anyway, he told me about his catheter issues and I was curious. I asked if he was experiencing any back pain or leg pain and sure enough he was. I called a surgeon that I trust and the patient was in surgery within a day. He had an issue that required surgical correction of a pathological issue. In this case, I am an essentialist. Now that I think about it, I am not sure if one can root for only one team.


  1. “These classifications can broadly be divided into three groups: (1) those that consider clinical descriptors, (2) those that describe prognosis, and (3) those that consider response to treatment.”


I am credentialed in Mechanical Diagnosis and Therapy, formerly known as the McKenzie Method. In this respect I am a little biased and it is important that you guys know that I am biased towards one method before reading the rest of the article. MDT would be a patient response approach.


Other systems, such as the Treatment Based Classification System (TBCS), which wasn’t even considered in this article it looks like, is a system that is based on clinical prediction rules. This means that if you come in and say some key words and test positive on some key tests that it would dictate a specific category of treatment, which is completely different from a patient that speaks of different key words and test negative on key tests.


  1. “We identified 28 classification systems of CLBP (chronic low back pain)…systems that described subclasses based on pathoanatomy, pahtophysiology, or clinical signs and symptoms without attempting to predict outcome or direct treatment were labeled as ‘diagnostic’…systems attempting to predict outcome irrespective of treatment were termed ‘prognostic’…systems that suggested treatments for different subclasses were termed ‘treatment based’…16 diagnostic, seven prognostic, and five treatment-based classification systems for CLBP.”


Typically, when I am writing a blog post I go to the back of my library (actually a trunk in the crawlspace) and grab an article that I read years ago. (I know…I am a nerd because I keep research articles that I read years ago). Anyway, re-reading the highlights of this article is like reading the article for the first time. I forgot that there were this many classification systems out there. Typically only a few are spoken of in the clinic and these are: the movement impairment system, Quebec Task Force, Mechanical Diagnosis and Therapy, Treatment Based Classification System and the Canadian Back Institute Classification System.


I will have to read the highlights of the article again in order to figure out which system fits into which category.


  1. “The first description of a treatment based system was by McKenzie, who classified patients into three main syndromes based on physical signs, symptom behavior, and their relations to end-range lumbar test movements”


Is it wrong that I was pounding my chest when I was typing the above sentence? This reminds me that I will have to write a blog on the history of MDT. One can see the history of MDT in the book Against the Tide.


  1. “Riddle and Rothstein assessed 49 physical therapists with varying clinical experience, in their ability to classify 363 patients according to the McKenzie system. Their ability to agree at the subsyndrome level was poor…Agreement among examiners was only marginally improved for classification into the three main syndromes…Agreement among examiners was better in three studies that assessed physical therapists who completed a certification in the McKenzie method with percent agreement ranging from 74% to 91% for subsyndromes and 93% to 100% for main syndromes.”


This tells us a few things. First is that those certified in using a method are actually good at using the method and those that aren’t certified aren’t as good at using a method. I think that this thought process would hold true for many aspects of different professions. I actually had a discussion on FB about this topic and I don’t think that it is the magic of the certification that increases agreement, but the hours upon hours of studying that went into preparation for the test that increases therapist’s competency of using a method. When a therapist is certified though, the agreement is close to perfect.


  1. “Movement System Impairment classification…proposed by Van Dillen et al and includes five categories based on signs and symptoms elicited with direction-specific tests in the direction of lumbar flexion, extension, rotation, rotation with flexion, or rotation with extension…shown to be reliable in three different studies”


I’ll have to read more about this system because at a glance it sounds eerily similar to McKenzie’s method. Both appear to have a “directional preference” based treatment and avoidance (I’ll assume only temporarily) of the aggravating factors.


  1. “Canadian Back Institute Classification system…recognition of syndromes or patterns of pain with no direct reference to pathoanatomy…the classification was based on the location of dominant pain, whether the pain was constant or intermittent, and which movements or postures exacerbated or alleviated the symptoms…shown to be reliable in one study.”


Again…these systems are starting to sound familiar and similar to each other. Figure out the symptom location, what makes them worse, what makes them better, is it mechanical or chemical and then name it for what it is. This appears to be the same in the three classification systems.


  1. “Movement and Motor Control Impairment (MCI) classification system by O’Sullivan proposed treatment based on subgroups of patients with CLBP categorized by five distinct patterns based on a specific direction of MCI…flive categories included flexion pattern (loss of motor control into trunk flexion resulting in excessive abnormal flexion strain), flexion/lateral shifting pattern (MCI around the lumbar spine with a tendency to flex and laterally shift at the symptomatic segment), active extension pattern (MCI around the lumbar spine with a tendency to hold the lumbar spine actively into extension), passive extension pattern (loss of lumbar motor control around the lumbar spine with a tendency to passively overextend at the symptomatic segment), and multidirectional pattern (MCI around the lumbar spine in multiple directions)…The percent agreement was 70%.”


It seems like this system is all about a loss of control at the lumbar spine. The agreement of classification isn’t bad at 70%. I struggle with this system because it does not appear to be a patient response based system. I’ll have to read more into this system. The first thing that I think of is “how do we know if we are doing the right thing and how long do we have to wait in order to determine if we are moving in the right category.


  1. “An RCT assessed the classification system by McKenzie by randomizing 260 patients into two groups: Group A was treated with the McKenzie method, and group B was treated with intensive dynamic strengthening training…tendency toward a difference in reduction of disability using the Low Back Pain Rating Scale in favor of the McKenzie group at the 2-month follow-up assessment, but no differences at the end of treatment (4 months) and at the 8-month follow-up evaluation.”


Some would look at this and say that MDT was no better than strengthening at 4 and 8 months. Others would look at it and say that MDT was better than strengthening at 2 months. If you were a patient, which would you rather have? Would you rather be better at 4 months or two months…knowing that you would be at the same place in 8-months? This study doesn’t seem too realistic in that once a patient is improved with MDT, then the treatment would transition towards a functional strengthening phase.


  1. “…overall strength of evidence …is High for the McKenzie and Movement Impairment Classification systems, especially when examiners have been extensively trained; Insufficient for the Canadian Back Institute Classification; and Moderate for the MCI Classification”


This sentence sums it up. MDT has moderate evidence to support that it is highly reliable. The Canadian Back Institute Classification system has low evidence to show that it is insufficiently reliable.


If I were a therapist going to learn a new method, I would have to start with MDT based on the volume of studies demonstrating reliability.


  1. “Once it is established that patients can be classified reliably, it then must be demonstrated that by directing a specific treatment at the subgroup, one can expect an improvement in treatment outcomes.”


This means that once we know what we are seeing…can we fix what we see? What is the purpose of classifying a patient into a group if the treatment for that group is ineffective?


  1. “This suggests that the ideal classification system should minimize the number of subgroups to ensure that the user can become confident (and competent) it its use with little training.”


Holy smoly do I disagree. We just said that the subgroups must lead to a specific treatment that performs better than other forms of treatment. If we minimize the number of subgroups, then we are minimizing the impact of subgrouping. For instance, if we state that there is only one subgroup, then what is the likelihood of the treatment for that one subgroup helping all of the patients? We already know that it’s pretty low…this is how we got into this mess to begin with. In the past, all low back pain was treated very similarly, with horrible effects. Now, if there is only one subgroup, we can be assured that most people would fit into this subgroup. Therefore, the therapists would be highly reliable in choosing the group in which to place the patient. THIS DOESN’T MEAN THAT IT WILL ACTUALLY BE EFFECTIVE TREATMENT!

Back pain is very costly in the US. We need to do a better job of minimizing the disability from LBP and educating the patients regarding back pain natural course and how to live and manage this ailment. There have been other systems created since this article was published in 2011 and we will see how these systems fair over time.

Excerpts from:

Fairbank J, Gwilym SE, France JC, et al. The Role of Classification of Chronic Low Back Pain. Spine. 2011;36(215):519-542.


link to article

Lumbar stenosis



  1. Lumbar spinal stenois (LSS)…defined by any narrowing of the spinal canal and/or nerve root canals…In patients with severe LSS, a space reduction of 67% has been found in the spinal canal.”


Spinal stenosis is the narrowing of the holes of the spine. The spine has 3 holes in it in the lumbar region. Each hole carries a nerve. It could either be the nerve of the spinal cord down the middle, and larger, hole. It could be the nerve roots out of the holes on the side of the spine. Each hole needs to be big enough so that it doesn’t irritate the nerve that it allows to pass through the hole. Picture a water pipe. If you put too much stuff in the pipe it will clog up. Sometimes there are tissues that can make their way into the holes of the spine to clog the holes. When the hole is clogged, the nerves don’t have as much room to do their job (transmitting signals to and from the brain). Now take that same pipe and come back and look at it over decades. There will be sludge and stuff built up around the pipe. This is essentially creating a smaller diameter on the inside of the pipe. This smaller diameter due to sludge is also creating a smaller hole. This could happen in the spine with severe arthritis or degenerative disc issues in which the hole gets smaller. A visual is much better so maybe this will help. image for spinal stenosis


  1. “…estimated the incidence of LSS in Denmark to 272 per one million inhabitants per year”


In other words, it is not very common in Denmark.


  1. “…it is important to discriminate between LSS and disc generated pain since these conditions have different prognoses and the range of evidence based treatments are different, as well.”


The treatment between the two issues, discogenic back pain and stenotic back pain, is very different. A thorough evaluation can start to correlate symptoms with either discogenic pain or non-discogenic pain. Many patients believe that an MRI will be the answer to why they have pain, but unfortunately this isn’t so.


  1. “a valid and reliable clinical assessment protocol for identifying LSS would be valuable in terms of choosing relevant treatment and informing the patient about the prognosis as early as possible.”


This article was written in 2009. The medical profession has existed for eons. There is still not a valid way to assess a patient in order to determine spinal stenosis. There are biologically plausible ways, meaning that when I assess you, I can make an educated guess from some of the findings in the history and physical, but it is not a valid (proven) way of coming to a conclusion.


  1. “The high sensitivity and specificity of MRI suggests this is a good test for ruling in and out the disease.”


The MRI does a great job of telling us what is abnormal, but it doesn’t do a great job of telling us if the abnormal finding is causing symptoms. As seen in the link above, there are abnormal findings in a population without symptoms. We have to take the imaging findings and see if they make sense after performing a physical exam.


  1. “…history will provide strong clues to the presence of spinal stenosis…more than 65 years of age…prolonged history of low back pain and intermittent radiating symptoms having developed gradually…limited walking capacity…Movements or positions involving flexion e.g. sitting or stooping, will often abolish symptoms…total loss of lumbar extension range is usually found, while flexion most often is well preserved.”


The typical patient with lumbar spinal stenosis will notice that the ability to walk has gradually reduced over time and there is a need to sit due to back or leg pain. Sitting will typically turn down or off the symptoms rapidly. This patient will have limited motion into extension (think of looking over your head to see the stars or bending backwards while standing).


  1. “…stenosis from zygapophyseal joint hypertrophy, ligament thickening or other degenerative changes, it cannot be expected that physical exercise or manual treatment will create a lasting change in the degree of space reduction in the spinal canal or intervertebral foramina”


In the presence of physical changes to the bones, ligaments or loss of disc height, there is nothing that a PT can do to change these back to the way that they were previously. These have been described as wrinkles on the inside. If we look at your face we can start to see how much age you have based on the wrinkles in the face. This is also done on the inside in that some “degenerative” changes are normal. Wrinkles are normal; they are not symptoms of anything sinister. The same can be said for physical changes on the inside. They don’t have to be pain generators. It takes a physical exam to determine how your symptoms respond and whether or not this matches the images on an MRI or X-ray. Even then, we can’t say that movement won’t help, only that we won’t change the physical “inside wrinkles”.


  1. “The main purpose of this pilot study is to evaluate the validity and intertester reliability of an algorithm of physical examination tests, in relation to identifying symptomatic lumbar spinal stenosis.”


This is good. A pilot study is like a pilot for a t.v. show. This is done to see if additional episodes should be done. This study will conclude if additional studies on this topic should be done.   What it hopes to find is a reliable (consistent) way of determining validity (actually seeing what the test hopes to see) in testing for lumbar spinal stenosis. A test that is both reliable and valid should be able to test for spinal stenosis regardless of who is performing the test and who is measuring the test.


  1. “Two patients were classified as “LSS” and five patients “Not LSS”, meaning a 29% prevalence of “LSS” Intertester agreement for overall diagnostic conclusion was 100%”


There are so few patients that this study will likely not yield any results that are actionable. The interesting thing is that the examiners agreed 100% of the time. This is not common in the medical field to have 100% agreement on near anything.


  1. “…the algorithm in its present form can not be used as a screening test to rule out LSS, although it may be able to diagnose the condition.”


There were so few people in the study that it is hard for any clinician to put it to use in the clinic. It may be able to diagnose the condition in that it demonstrated a specificity of 1.0, which is really good.



Excerpts taken from:


Lengsoe L, Lyhne S, Melbye M. An algorithm for clinical identification of spinal stenosis-a pilot study of validity and intertester reliability. International J of MDT. 2009;4(2):21-28.


Can’t find the abstract to the study, but it is listed under the author’s CV–a-pilot-study-of-validity-and-intertester-reliability(6d714ee0-d910-11de-9e3b-000ea68e967b).html





Lateral shift deformity

Crooked patients

1. “A lumbar lateral shift (LLS) is defines as a lateral displacement of the trunk in relation to the pelvis…repeatedly associated with discogenic pathology…McKenzie reported that 90% respond rapidly to manual correction.”


In school we learn the theoretical aspect of the shift, but when you see your first patient that is shifted the though process immediately goes to a mixture of “oh shit and piss on yourself excitement”. The shift can be extremely painful and students, if not treating this in a clinical, may not be prepared for a patient in a true 10/10 pain status. After so many years in practice, it is just another puzzle to solve now. The excitement has gone away and lucky for the patients, so has the “oh shit” response.


Patients come into the clinic “crooked”. Scott Herbowy once said it is like looking around the corner to see if the dog is hiding.


2. “…prevalence of LLS is difficult to establish, but estimates range from 5.6 to 80% of patient with low back pain (LBP).


This statistic is so far away from informative, that it shows that it is present in any where from 5-80 out of 100 patients with back pain. I don’t see it in 80% of the patients, but 5% may be more applicable to my population in the clinic.


3. “Lumbar spinal fusion, perhaps the most invasive of these (surgical) procedures, is increasingly common in the United States. However, its effectiveness is questionable…”


If you are going to have a fusion, go so someone that is either certified or diplomaed in MDT first. Some things can’t be undone, and this is one of those things. Make sure that there are no other options of getting relief prior to undergoing something that may not be effective and can not be undone.


4. This article is a case study of a patient that has a lateral shift deformity in the presence of an “X-stop” device, which is typically used to prevent lumbar extension in the case of spinal stenosis. The patient centralized with side gliding mobilizations and was issued side gliding against the wall in order to close the affected side. The patient responded well to this motion within the initial 4 visits and the final 4 visits were used to improve functional performance without the return of the lateral shift. The X-stop makes this case interesting because typically patients that are post-surgical are excluded from most research.


5. “The rapid centralization of symptoms observed in this patient is similar to that reported in previous case reports describing a lateral shift correction. Centralization or peripheralization during repeated movement testing has been positively correlated with pain provocation during lumbar discography.


Centralization phenomenon is something that trained clinicians are looking for during examination of the spine. When noted, the results are typically great, but if the peripheralizes (opposite of centralization), then the patient’s results are typically poor, at least if it happens with all movements tested.


First point to make from this is that if you have back pain, seek out a trained therapist in order to address your symptoms. Always start conservative before going invasive for pain based symptoms. If you have progressive weakness or have a loss of bowel and/or bladder function go the doctor immediately, but aside from this stay conservative first.


Second, people get crooked. If the crooked is not associated with pain, it may be that the person has always been crooked. Not all crooked people need therapy.


Excerpts taken from:


Peterson S, Hodges C. Lumbar lateral shift in a patient with interspinous device implantation: a case report. JMMT. 2016;24(4):215-222.

Core stabilization compared to McKenzie method treatment


  1. “The condition has been identified as the leading contributor to ‘years of life lived with disability’ in the world including the United States.”


Big surprise, we are talking about back pain again. I see a majority of my schedule as back pain for the previous 8 years. There is no loss of people with back pain. This is an epidemic. The only reason it is not treated in such high regard has cancer, AIDS, Zika, and others is because it’s not deadly and does not cause major deformities. Because back pain is so common, it’s treated with little urgency such as the common cold.

  1. “In Australia, LBP is estimating to reduce gross domestic product by $3.2 billion annually and is the leading cause of early medical retirement for older working people.”

Think about that! You go to school and you load up on student loan debt. After school, you get a job paying much less than you think you’re worth. Then you get sidelined by low back pain and are forced to retire well before you’re ready. It doesn’t have to be this way! Not all low back pain is the same, and when you figure out what type of back pain you have it becomes a lot easier to prevent recurrent issues of back pain.

  1. “Directional preference classification is characterized by a reduction in distal pain and/or observation of the centralization phenomenon with the application of repeated or sustained end-range loading strategies to the spine that remain better after assessment. Centralization is defined as a progressive change in pain from a more distal location to a more proximal location that remains better after applying repeated or sustained end-range movement to the spine… hallmark characteristic of the McKenzie derangement classification.”

There is no doubt that a directional preference correlates with great outcomes. There is no doubt that centralization correlates with great outcomes. The thing that needs to happen is that therapists need to be trained to see these during the initial evaluation. A majority of patients demonstrate a classification utilizing the McKenzie method, based on the research of Stephen May. The derangement classification is the largest classification syndrome based off of Stephen May’s previous research, but there are other syndromes. Typically, it’s the derangement syndrome that the research attempts to study. I see very few articles on the other two syndromes in the mainstream research journals.

  1. “There is some evidence that improvement in size and recruitment of the muscles of the spine, including the transverse abdominis, is associated with improved function in the short-term when patients with low back pain receive motor control exercises compared to general exercise or spinal manipulation. However, increases in transvere abdominis and lumbar multifidus thickness using real time ultrasound have also been observed immediately and one week following spinal manipulation in people with low back pain, suggesting that increases in transverse abdominal recruitment may not be specific to motor control exercises.”

OK, a muscles ability to contract is not dependent on its size. A muscle’s ability to contract is based off of that muscle’s ability to receive the nervous system input from the central nervous system. Should there be something that allows for better neural activity, we expect to see an increase in muscle contraction and possibly an increase in muscle size. This is important because we may not have to train a muscle in the traditional sense in order to making muscle contract better.

  1. “The McKenzie method was prescribed according to the principles described by McKenzie and May… Delivered by two therapists who had obtained the level of credentialed therapist from the McKenzie Institute International… Mechanical therapy, including patient and therapist generated forces utilizing repeated or sustained and range loading strategies in loaded or unloaded postures, according to the patient’s directional preference..that guided by symptom response. The aim was to reduce, centralize, and abolish peripheral symptoms… Once symptoms centralize, any movement loss was then treated with repeated and range movements in the direction of movement loss… Received a copy of treat your own back to supplement treatment and self-management.”

The patients included in the study were all patients of the derangement syndrome. When assessing a patient utilizing the McKenzie method, we are attempting to classify the patient into one of three syndromes. This has a high reliability when performed by therapists that are highly trained. The hallmarks of the derangement syndrome is centralization, this occurs when symptoms move from a segment far away from the spine towards the spine. The symptoms in the furthest position from the spine have to decrease or abolish. This is accompanied by the directional preference. A directional preference is as stated, when we move you in a specific direction…your body prefers that. Your body tells us it prefers that direction by centralizing symptoms, improving range of motion, improve strength, or improving other neurological tests such as reflexes and dural tension testing. One can also have a directional preference in the absence of centralization, as extremities also demonstrate directional preferences.

  1. “Initially, promotion of independent contraction of the deep stabilizing muscles, such as the TrA and multifidus, was facilitated by pelvic floor contraction…Objectively, skill mastery of TrA recruitment was measured by palpation and visual assessment for a reduction of overactivity of the superficial trunk muscles…practice daily…attend the physical therapy clinic twice a week for the first 4 weeks and once per week for the remaining 4 weeks”

This is beat into students during PT school…understanding the impact of performing TrA contractions on low back pain. The problem with this theory is that the research is scant on cause and effect. We know that patients with low back pain have smaller multifidi and TrA muscles, but we can’t say “chicken or the egg” yet. We also can’t say if the back pain caused the smaller muscle or if the muscle was smaller and then it caused back pain. More research needs to take place. The topic of centralization and directional preference was briefly touched upon while I was in PT school and the topic of TrA was hammered into us. Now it appears that centralization and directional preference are being taught more in PT schools based on the students that I get as a clinical instructor.

  1. “Participants allocated to the McKenzie method group attended an average of 5.4 +- 2.5 treatment sessions over an average of 38.6+-18.8 treatment days, while participants in the motor control group attended an average of 6.5+-2.7 treatment sessions over 47.3+-22.7 treatment days”

This doesn’t look like a huge difference, but this indicates that those being treated by a MDT credentialed therapist, one less session was required. Think about this again. Each session is performed at a cost to insurance companies (read Medicare) of about $100. At this point, each patient would save $100 to insurance companies when seen by a credentialed MDT therapist. This, over the long term, has dramatic effects on the total cost of spending in the US.

  1. “…no statistically significant effect for treatment group for muscle thickness…at an 8-week follow-up in a population of people reporting chronic LBP classified with a directional preference. Global perceived improvement was the only secondary outcome that demonstrated a significant between-group difference, which favored the McKenzie method”

Let me say this slowly. Using a directional preference based exercise provides the same result as actually training a specific muscle in terms of muscle size! This is huge! We all are taught that to make a muscle bigger (hypertrophy) requires up to 6 weeks of performing an exercise in order to specifically improve a muscles size. This indicates that a muscle’s size can increase without any direct exercises to improve a muscle’s size.

The final piece of this is that those treated with MDT based principles actually felt better than those receiving motor control exercises (read this as core stabilization).

You walk into any clinic in America (aside from those that are doing MDT) and you will see bridges, bird-dogs, pull your belly into your spine exercises, and of course the traditional hot pack and e-stim. These types of treatments may not be the best. Ask your therapist how your back pain is classified. If they can’t give you a straight, honest, and well reasoned answer…FIND A NEW THERAPIST!

  1. I am bolding this, because it is important to read straight from the article. There will be no explanation needed.

Results from our study suggest that in patients with a directional preference, receiving exercises matched to their directional preference is likely to produce a greater sense of improvement than receiving motor control exercises.”

Excerpts taken from:

Halliday MH, Pappas E, Hancock MJ, et al. A Randomized Controlled Trial Comparing the McKenzie Method to Motor Control Exercises in People with Chronic Low Back Pain and a Directional Preference. J Orthop Sports Phys Ther.2016;46(7):514-522.

centralization and the correlation to discogenic pain

Critical Appraisal for a Reference-Standard Validity Study


P: For patients with chronic low back pain, with varying levels of distress,

I: can the centralization phenomenon

C: as compared to discography results

O: provide diagnostic power for discogenic pain



Vincent Gutierrez, PT, MPT, cert. MDT



Ovidsp with keyword terms “low back pain and centralization and specificity and sensitivity”.   44 citations were found between the years 2004 and 2014.


Date of Search: January 21,2014

Re-evaluation date: January 25, 2014



Laslett M, Oberg B, Aprill C, McDonald B. Centralization as a predictor of provocation discography results in chronic low back pain, and the influence of disability and distress on diagnostic power. Spine Journal 2005;5:370-380.



This validation study has two purposes. The first is to investigate the predictive value of the centralization phenomenon (CP) in relation to provocation discography, which is the only reference standard available for discogenic pain. The second is to investigate the role of distress and disability with regards to the predictive value of the centralization phenomenon in relation to provocation discography.


The inclusion criteria were patients with persistent low back pain (LBP), with or without lower extremity (LE) symptoms, whom were referred to a private radiology practice. Patients were excluded for the following reasons: a normal magnetic resonance imaging (MRI) assessment, severe degeneration associated with spondylolisthesis, and if the discography was contraindicated or a referral ruled out discography. The patients that were included were assessed consecutively.


Prior to the evaluation by a physical therapist, the patient completed a visual analog scale (VAS) for pain and the Roland-Morris Disability Questionnaire (RMDQ). The Zung Depression Index (ZDI), Modified Somatic Perception Questionnaire (MSPQ) and the Distress Risk Assessment Method (DRAM) were also completed prior to the physical therapy (PT) evaluation. The evaluation was performed prior to the discography and the physician performing the discography was blinded to the therapist’s results. The therapist was blinded to the results of the subjective outcome measures.


The physical evaluation consisted of a McKenzie evaluation. The exam required 30-60 minutes and also included sacro-iliac joint (SIJ) provocation tests. Centralization or peripheralization was noted and at this point the examination was terminated.


Discography was performed using standard technique and the patient was required to report pain in at least one disc, without pain at an adjacent disc in order to receive a positive test result.


One hundred eighteen patients participated in the PT evaluation and discography. One hundred seven patients were included in the initial analysis. Of the 107 patients, 69 received a full PT evaluation, 21 received a partial evaluation and 17 did not receive an evaluation. Of the above, the physical therapist offered an opinion regarding CP for 83 patients.



The authors utilized the only reference standard studied, provocation discography, in order to determine if CP is predictive of discogenic pain. The physician was blinded to the physical therapists’ evaluation and the physical therapist was blinded to the patients subjective outcome measures. Not all patients received both the PT evaluation and discography.


The confidence interval was 95%. For non-distressed patients, the following statistical measures were calculated: sensitivity of 37%, specificity of 100%, positive likelihood ratio (LR+) and negative likelihood ratio (LR-) were incalculable due to a specificity of 100%. For distressed patients, the following statistical measures were calculated: sensitivity of 45%, specificity of 89%, LR+ of 4.1, and LR- of 0.61. For not severely disabled patients, the following statistical measures were calculated: sensitivity of 35%, specificity of 100%, LR+ and LR- are incalculable due to 100% specificity. For severely disabled persons, the following statistical measures were calculated: sensitivity of 46%, specificity of 80%, LR+ of 3.2 and LR- of 0.63



Performing a McKenzie evaluation in order to determine the presence of CP is a good test for determining a positive discography, especially in patients without severe disability or distress. The presence of CP improves the pre-test probability to post-test probability of positive discography from 39% to greater than 75% in patients with severe disability or distress. CP is a strong predictor of positive discography in patients without severe distress or disability.

Not all back pain has a definitive cause

“Findings such as disk height loss and disc bulges are coming in individuals without low back pain.”

Disc bulges, degenerative joint disease, spinal stenosis, do you all a result of living in this world. We have gravity acting a force on us almost 16 hours a day. Anytime that there is a problem, we want to blame something or somebody. Low back pain is an enigma at times. we can draw correlations, we can come up with risk factors, we can even tell you how to treat it sometimes, but what we can’t do is tell you exactly what causes your back pain.

“Surprisingly, disc protrusions were associated with a lower risk of subsequent back pain. Nerve root contact and central stenosis had the largest hazard ratios on baseline imaging findings, and they were associated with incident back pain in the expected direction but not statistically significant. Self identified Depression was the strongest predictor of subsequent back pain, with a greeter hazard ratio than any imaging findings.”

What should be taken from the above statistics is that mental health plays a role in pain. There are a lot of new studies that are associating catastrophizing and external locus of control with increased pain levels. Work by Nadine Foster demonstrates screen for patients who will have a difficult time improving with therapy alone. New were books, such as the one by Annie O’Connor and Melissa Kolski (two people with whom I’ve studied at our RIC), goes into great detail regarding pain science. Big picture, we can not neglect the patient’s emotional well-being when attempting to treat the patients physical complaints.

“Our results indicate that depression is a strong predictor of who will subsequently reports low back pain then baseline imaging findings.Subjects with self reported depression at baseline were 2.3 times is likely to have back pain compared with those who do not report depression.”

There is obviously a psycho social component to low back pain. The question is… Chicken or the egg. Is a person more likely to be depressed because they have back pain that is not improving? Or is that person more likely to have back pain because they are depressed? I don’t think that there are cause and affect articles in the literature at this point, but there is definitely a high correlation between patients who are depressed and patient who continue to report low back pain.

“In our analysis of baseline data, we concluded that central stenosis, nerve root contact, and disc extrusion were the most important imaging findings related to prior low back pain. Our current analysis indicates that central stenosis, disc extrusion, and route contact may also be risk factors for future low back pain.”

In other words, if you have a major deformity you will probably have pain. This doesn’t mean that you will definitely have pain, it just increases your risk of experiencing symptoms.

The moral of the story is that we cannot deny the brain. The brain has the ability to see pain, and some patients are more susceptible to seeing this pain. Don’t get me wrong, a thorough mechanical evaluation should be performed when a patient has pain, but when this patient is not inclined to respond to mechanical therapy, the patient should be referred to someone that can better handle this patient’s pain.Sometimes, that person will be a behavioral therapist, a psychotherapist, or a clinical psychologist. Physical therapists are not always the go to in order to treat a patient’s pain.
Excerpts from:

Jarvik JG, Haegerty PJ, Boyko EJ. Three-Year Incidence of Low Back Pain in an Initially Asymptomatic Cohort. Spine. 2005;30(13):1541-1548.

CAT of the Oswestry and Roland Morris

A Critical Appraisal of the responsiveness of a patient specific outcome measure compared with the Oswestry Disability Index v2.1 and Roland and Morris Disability Questionnaire


P: For patients with at least a six week history of back and/or leg pain

I: is the patient specific outcome measure

C: as compared to the Oswestry Disability Index and Roland and Morris Disability Questionnaire

O: as responsive



Vincent Gutierrez, PT, MPT, cert. MDT



Ovidsp with title term of “Oswestry”. The results were limited to full text.   73 citations were found with no limit to year published.


Date of Search: February 22,2014



Frost H, Lamb S, Stewart-Brown S. Responsiveness of a Patient Specific Outcome Measure Compared With the Oswestry Disability Index v2.1 and Roland and Morris Disability Questionnaire for Patients With Subacute and Chronic Low Back Pain. Spine 2008;33(22):2450-2457.




The purpose of this study is to assess the responsiveness of the Patient Specific Activity Questionnaire compared to the Oswestry Disability Index v2.1 and the Roland and Morris Disability Questionnaire.


The inclusion criteria were subjects at least 18 years old, with at least a 6-week history of low back pain. Subjects with or without leg pain and/or neurological signs were also included. Subjects were excluded for the following: serious pathologies, gynecological problems, ankylosing spondylitis, tumors, infection, past spinal operations, pregnancy, serious spinal pathology, unable or unwilling to complete questionnaires independently, received physical therapy in the previous month or were referred for intensive functional restoration.


The subjects were randomized to either a physiotherapy advice group or an advice with additional physiotherapy intervention group, which was not described. The subjects completed the following measures both prior to intervention and after 12 months: the Patient Specific Activity Questionnaire (PSAQ), the Oswestry Disability Index v2.1 (ODI), the Roland and Morris Disability Questionnaire (RMDQ), and the Global Transition Rating Scale (GTRS).


Subjects were divided into three groups, based on the GTRS, after 12 months: better, same or worse. The change score was calculated by subtracting the baseline scores from the follow-up scores for the ODI, RMDQ and the PSAQ. A relationship was established between the GTRS and each of the latter three outcome measures using 1 way analysis of variance (ANOVA).


Of the 286 subjects randomized initially, 201 completed the follow-up questionnaire. The PSAQ is responsive among subjects noting improvement, but is less sensitive to those reporting no improvement, when compared to the ODI and RMDQ.



The validity of the study is questionable, as there are flaws in the design. The authors do not account for the patient’s lost to follow-up, which may alter the results of the study. The authors fail to note the differences between the two groups at baseline, with regards to demographics. The authors note that a P value of <0.01 was utilized to differentiate among the outcome measures, which provides for a more limited confidence interval to detect true change.


The authors note the area under an operating receiver operator curve as follows: ODI as 0.752, PSAQ as 0.751 and RMDQ as 0.689. This indicates the order of responsiveness to change respectively.


Because the study results indicate that the tested outcome measure (PSAQ) is less sensitive to those patient’s with little change in status pre-post intervention, it would be prudent to utilize an outcome measure than can assess both small and large changes in status, such as the ODI and RMDQ.



With patients presenting to the clinic with complaints of at least a six week history of low back pain, with or without leg symptoms, the ODI would be the most effective to assess change.



key to unlocking pain

The Audrey Long article is commonly cited as one of the landmark articles for those of us that treat spines utilizing directional preference and centralization.  As well it should be! The results are unbelievable.  This researcher…I take that back…clinician performing research in the clinic published an article that, up until this time, was only speculation.  How could patients not get better using evidence based practice?  For a long time, the evidence was based on expert opinion and not really research.  I enjoy reading the publications on spine and it is interesting to read the changes in the Clinical Practice Guidelines for Low Back, published in JOSPT over the years.  Prior to the most recent publication, flexion based exercises were the rage and directional preference was only moderately supported.  Thanks to clinicians such as Audrey Long, this type of treatment has gained more support in the practice guidelines.  When I was in PT school (I sound like an old man, and year to year this is true, but not yet) this was a very small talking point in our curriculum.  Students now come out of school with a better awareness, though not a true understanding, of the concept of directional preference.


A Critical Appraisal of Directional Preference Exercises Compared to Two Other Exercise Paradigms


P: For patients with low back pain, with or without leg pain, demonstrating a directional preference

I: is treatment with a directional preference

C: as compared to treatment in the opposite direction of the directional preference or an evidence based approach

O: more beneficial when compared with subjective outcome measures



Vincent Gutierrez, PT, MPT, cert. MDT


Search: with the keyword terms “directional preference and low back pain”. Nine results were found and the article with the highest score was chosen.


Date of Search: February 15,2014



Long A, Donelson R, Fung T. Does it Matter Which Exercise? A Randomized Control Trial of Exercise for Low Back Pain. Spine 2004;29(23):2593-2602.




Eleven clinics, from five separate countries, participated in the study. Consecutive patients presenting for treatment of LBP, with or without leg pain, were asked to participate in the study. The inclusion criteria is as follows: consecutive patients with low back pain, with or without one neurological sign, age 18-65 years and demonstrating a directional preference. The exclusion criteria is as follows: cauda equina, two or more neurological signs, spinal fractures, post surgical, off work for one year or more due to low back pain, medical causes, uncontrolled medical conditions, pregnancy, inability to read English (except for those from Germany), patients with prior knowledge of, or specific physician referral for, the Mckenzie method, or no directional preference (DP) elicited.

Therapists credentialed or diplomaed in the McKenzie method performed the assessments. The directional preference was noted as either extension, flexion or lateral, but the subjects were shielded from the significance of directional preference. The subjects were then randomized to one of three groups: matched directional preference exercises, opposite directional preference (ODP) exercises, or evidence-based care (EBC). There were no baseline differences among the groups regarding demographics

The subjects attended at least three and no more than six sessions over the course of two weeks. Those in the DP group received exercises that matched the DP and were instructed to avoid all activities that increase intensity or radiation of symptoms. Those in the ODP group received exercises that were opposite to the DP, and the EBC group performed mid-range exerices and stretches for the hips and thighs. The final two groups also were instructed to return to remain active.


The outcome measures utilized in the study are as follows: back and leg pain intensity ratings using an 11 point visual analogue scale, the Roland Morris Disability Questionnaire (RMDQ), and medication use.


Five hundred three subjects were assessed and 230 demonstrated a DP as follows: 83% extension, 7% flexion and 10% lateral. Twenty-nine dropped out of the study at two weeks, and the remaining 201 were eligible for analysis. There were 36 withdrawals, which indicated that the subjects worsened or had no change in symptoms and were transitioned to alternative care. None of the DP group withdrew, but 16 in the ODP and 20 in the EBC withdrew. All outcome measures improved in the three groups over the course of two weeks, with the DP group demonstrating significant improvement compared to the ODP and EBC.



The authors satisfied eight of the ten questions regarding the Quality Appraisal Checklist. The subjects’ group design was not blinded to those enrolling the subjects and this was a comparison study of varying interventions, which indicates that a true control group was absent.


This study will have a direct impact for clinical therapists. Because this study compared three different interventions, opposed to identifying the efficacy of a single intervention compared to a control group, it mimicked clinical practice. The authors compared evidence based care with a directional preference treatment paradigm, which would be similar to a question asked in clinical practice.



Directional preference exercises are superior to performing exercises opposite to the directional preference or “evidence based care”. Patients that demonstrate a directional preference and are treated accordingly perform significantly better in outcomes measured in this study. There appears to be no harm in treating a patient with directional preference exercises, but the same does not hold true for performing exercises opposite of the directional preference or “evidence based care”.

The following is breakdown of the systems involved in “assessing” research articles.


  1. Were the subjects randomly assigned into groups?

Yes. The subjects were randomly assigned to one of three groups.


  1. Was each subject’s group assignment concealed from the people enrolling individuals in the study?

No. Because the study is a multi-centered study in outpatient practice, it was not acceptable to the authors to have patients drop out of the study due to changing therapists.


  1. Did groups have similar characteristics at the start of the study?

Yes. The authors note that there were “no differences among the three treatment groups in any baseline demographic characteristics or outcome measures.”


  1. Were subjects masked or blinded to their group assignment?

Yes. Although subjects couldn’t be blinded to the treatment, they were unaware of the specific grouping (i.e. matched vs unmatched vs EBP)


  1. Were clinicians and/or outcome assessors blinded to the subjects group assignment?

No and yes. Although the clinicians were not blinded, which is common in practice, the assessors of the outcome measures were blinded. The outcome measures were subjective measurements in order to minimize therapist bias.


  1. Did the investigators manage all of the groups in the same way except for the experimental investigation?

No. Because this is an intervention study in patient’s seeking treatment, it was impossible to withhold treatment to establish a true control group. The types of treatment for the opposite direction and EBP groups were vaguely described.


  1. Did the investigators apply the study protocol and collect follow-up data on all subjects over a time frame long enough for the outcomes of interest to occur?

No. There was a 12% dropout rate, which was anticipated by the authors when determining the number of patients needed to maintain a power of .90. Thirty-eight patients withdrew from the study, as opposed to dropped out, due to no improvement or worsening of symptoms in the EBP and unmatched group. Two weeks was long enough in order to assess change.


  1. Did subject attrition occur over the course of the study?

Yes. Twenty-nine subjects dropped out of the study, with 12 of the 29 dropping out due to no change or worsening of symptoms. Although these participants did not “drop out”, indicating that follow-up information was unattainable, 36 subjects withdrew early due to no change or worsening of symptoms. No subjects from the matched group withdrew. Those that withdrew completed the outcome measures prior to 2 weeks in order to seek alternative care.