A new study on the Alexander Technique and knee pain was published last month in the journal BMC Musculoskeletal Disorders. 21 subjects with knee osteoarthritis were each given 20 Alexander Technique (AT) lessons. After their lessons, they not only reported a 50% reduction in pain, but showed significantly less co-contraction in their leg muscles during walking. The entire study is available online to read here.
When I first read the study, I was struck by its size. 21 subjects (the study also used 20 healthy individuals as a control) didn’t seem to be that many. I’ve grown used to reading the larger randomized control trials, like the ATEAM back pain study published in the British Journal of Medicine in 2008. The ATEAM study involved 579 subjects. My assumption was that the larger the study, the more robust the findings. What could a study of 21 people really tell us?
The design of a study, however, depends on its purpose. The ATEAM back pain study—or last year’s ATLAS Annals of Internal Medicine study of whether AT or acupuncture are effective in treating chronic neck pain—is an example of a clinical trial, and such trials need to be large for a number of reasons. Subjects are divided into multiple groups. For example, the ATEAM back pain study randomly assigned subjects to groups that took 6 AT lessons, 24 AT lessons, 6 massages, or a control. Such large trials also deal with conditions that tend to have nebulous causes. The idiopathic back pain diagnosis studied in the ATEAM trial is just the technical way of saying, “your back hurts and we don’t know why.” The measurements in clinical trials—like self-report—are often subjective, and the clinical effects of an intervention are usually small. So a clinical trial needs to be large to show statistical significance.
Such large studies, though, can shift clinical practice—what doctors’ prescribe when they’ve diagnosed a patient with idiopathic back pain or chronic neck pain. As such, they also tend to generate headlines in the popular press. Last year’s ATLAS neck pain study was reported on by Time, NPR, Fox News Health, and Harvard Health, among others.
The current study of subjects with knee osteoarthritis is not a clinical randomized control trial. It is basic research, in a laboratory environment that is more controlled than usually possible in large clinical trials. It also uses more concrete measurement. As such, it can establish robust findings with many fewer subjects. And the purpose of such basic research is not only to establish whether the Alexander Technique might be an effective way to treat knee pain, but why. It’s an exciting window into researchers at work. Let’s unpack the study.
Why do we hurt?
Chronic pain is a surprisingly mysterious thing.
Take knee osteoarthritis. Knee osteoarthritis is a condition where the cartilage that cushions the bones of the knee joint starts the wear down. In this study, the 21 participants all had received an x-ray diagnosis of knee osteoarthritis. An x-ray doesn’t show cartilage: but it will show that the space between bones has narrowed, indicating osteoarthritis. Blood work can further confirm that the condition isn’t systematic (as in rheumatoid arthritis).
It sounds obvious, but if you have knee osteoarthritis, your knees usually hurt. But why? It has long been assumed that the pain that accompanies knee osteoarthritis is the result of the breakdown of cartilage in the joint. Less cushioning equals more pain. But researchers have been unable to show a relationship between the severity of pain and the degree of cartilage loss. Some people with minor cartilage loss in the joint suffer a great deal of pain. Others with major cartilage loss experience little discomfort—they may not even know they have the condition. The same is true in some forms of chronic back pain: there are individuals with significant damage to their vertebrae or intervertebral discs, yet experience little pain, and vice versa.
This new study explores two possible explanations for the pain that accompanies knee osteoarthritis.
1. Is pain a self-fulfilling prophecy?
One possibility is that the patients with knee osteoarthritis become more sensitive to pain. You might call this the “self-fulfilling prophecy” explanation of pain. The idea is that people with chronic pain begin to anticipate that pain, and therefore experience more pain. Previous research has found that Cognitive Behavioral Therapy or mindfulness meditation can reduce a heightened sensitivity to pain in subjects with osteoarthritis or fibromyalgia.
While the subjects in this new study reported a significant reduction in pain after their Alexander Technique lessons, the researchers didn’t find evidence that their Alexander Technique lessons had reduced their sensitivity to pain. The research didn’t disprove the idea that subjects with osteoarthritis might have a heightened sensitivity to pain, just that the benefits of AT didn’t seem to function along those lines. (The authors caution that the study might be too small to know for sure.)
2. Is pain the result of how you move?
The other possible explanation for knee pain that this study explores is that the patients with knee osteoarthritis use excessive co-contraction of their leg muscles during walking and other everyday activities. To understand co-contraction, it helps to know a bit about how muscles work.
Muscles often work in pairs. The muscle that’s doing the work is called the agonist. The muscle that is allowing the work is called the antagonist. Take a simple action like the biceps curl. When your biceps works to close the elbow joint, its antagonist, the triceps, releases. When the triceps works to open the arm at the elbow, the biceps releases. The biceps and triceps are an antagonistic pair.
Your hamstrings and quads muscles in your legs are an antagonistic pair as well. Check out this animation of the muscles that are active during walking. While walking is a much more complicated movement than a biceps curl (this animation also includes the iliopsoas, the glutes, and muscles of the lower leg called the tibialis anterior and calf) you can see the quads and hamstrings taking turns during the gait cycle. If you find the action of the muscles hard to follow, watch the bars of activity on either side of the walking figure.
Co-contraction is when muscles that are usually antagonistic activate simultaneously. In everyday activity, co-contraction functions to brace a joint and isn’t necessarily unhealthy. But previous research has found that individuals with knee osteoarthritis often use excessive co-contraction in their leg muscles during walking and other everyday activities. In this study, the subjects with knee osteoarthritis showed significantly higher levels of co-contraction in their leg muscles than the healthy control group as measured by EMG at the start of the study.
In the study, the 21 participants with knee osteoarthritis had 20 Alexander Technique lessons. The lessons were spaced out over 12 weeks: they had lessons twice a week for 8 weeks and then once a week for the final four weeks.
After their Alexander lessons, the participants showed a dramatic reduction in pain: 56% less pain than at the start of the study. 15 of the study participants regularly took pain killers (analgesia) at the start of the study. 10 stopped taking medication after their Alexander lessons ended. 11 of the participants also reported experiencing less pain in other areas, including neck, shoulder and back.
The subjects also exhibited significantly less co-contraction during walking than at the start of the study. Interestingly enough, the patients did not show an increase in strength over the course of the study—the measurements of leg strength were the same before and after their Alexander lessons.
None of the subjects used any other therapy during their Alexander Technique study. When the researchers followed up 15 months after the start of the study, the subjects had retained the reductions in pain, reporting 51% less pain than before their Alexander lessons.
Evidence & Measurement
We are living in an increasingly “evidence-based” world. Evidence depends on valid and reliable measurement. Alexander teachers have learned that their experience and the experience of their students isn’t considered very robust evidence for scientists studying health and movement. However compelling the anecdote—of pain diminished, increased ease, health restored—saying, “I saw it happen” or “It happened to me,” doesn’t really count.
Most studies of the effectiveness of a particular health intervention use some kind of self-report. For the current study, the subjects filled out the WOMAC questionnaire—a common tool used to evaluate knee and hip pain, stiffness and functioning—before and after their Alexander lessons. It’s on the basis of that self-report that we can say that pain was reduced 56% by Alexander Technique lessons. This kind of self-report is quite a bit more reliable than anecdotal data, but researchers are always looking for other sources to corroborate self-report. Bias too easily creeps in.
The advantage of basic research is the chance to experiment with different types of measurement. With only 21 subjects, the researchers can hook them up to EEG to try to measure the anticipation of pain. They can measure muscle activity with EMG and put the subjects on a force platform to assess joint loading. This kind of laboratory research is much more time intensive and expensive than having someone fill out a questionnaire.
The results of the study suggest that this kind of EMG measurement is worth it in studying the effects of AT on knee pain and osteoarthritis. The researchers found that the self-reported reduction in pain was correlated with a measurable change in the subjects movement coordination—a physiological change in the activity of their musculature. It points in a promising direction, both for scientists who study human motor control and the causes of musculoskeletal pain, as well as doctors who treat knee osteoarthritis. And it might even inspire one of those large randomized control trials to see if the findings hold up in the messy world of clinical medicine.
Many thanks to Tim Cacciatore, one of the authors of the current study, for his feedback on an earlier draft of this post.