No, JAMA does not now recommend chiropractic care first for low back pain


Of all the alternative medical systems out there, chiropractic is one of the oddest. Unlike many of the others, it has a modicum of plausibility, at least for back problems due to musculoskeletal strains. After all, the science-based specialty of physical therapy uses spinal manipulation to treat back problems. Of course, the big difference between chiropractic and physical therapy is that chiropractic is based on a delusion, namely the concept of subluxations. To science-based specialties, a subluxation a painful partial dislocation. This is different from a chiropractic subluxation, which is claimed to cause disease by interfering with nerve supply to organs. Such subluxations have never been shown to exist, nor has it ever been shown that “innate intelligence,” which to chiropractors is the organizing property of living things and a term originally coined by Daniel David Palmer, the founder of chiropractic. Palmer proclaimed that subluxations interfered with the body’s expression of “innate intelligence,” that controls the healing process. One notes that this vitalistic concept is very much like qi in acupuncture, except that instead of needles redirecting or unblocking the flow of qi in chiropractic adjusting subluxations unblocks the flow or function of the “innate intelligence.” Unfortunately, chiropractic subluxations have never been shown to exist.

Such is the reason why I like to refer to chiropractors as physical therapists with delusions of grandeur. They claim to be able to treat all manner of disease, not just back problems, by adjusting the spine. Even more unfortunately, among the alternative medical specialties, chiropractic is arguably the most entrenched, the most accepted, of them all.

Still, chiropractors have a much-earned inferiority complex. They desperately crave the acceptance that they don’t deserve and the scientific acceptance they don’t have. That’s why, whenever a story like this pops up, it’s spread far and wide as though it’s some sort of validation of chiropractic. In this case, the story is entitled JAMA recommends chiropractic as first means of back pain treatment. Let’s just say that the title of this article is profoundly misleading:

Medical doctors and chiropractors are often at odds with each other. The Journal of the American Medical Association now recommends chiropractic as a first means of treating back pain.

Dr. Alex Vidan wasn’t going to let that pass by without a comment. He stops by FOX2 with his pal Mr. Spine with more information.

As is the video itself:

My first inclination when I saw this article and the accompanying video was to wonder what the source of Vidan’s claims were? What article in JAMA was he referring to? I couldn’t find it, search as I could. The article didn’t seem to exist. Then I finally found it when I came across another chiropractor touting the article, and the reason why there’s much less there than meets Vidan’s eye became apparent. Basically, it’s an article by Denise M. Goodman, MD, MS; Alison E. Burke, MA; Edward H. Livingston, MD entitled, Low Back Pain. It’s a brief article on the causes and treatment of, yes, low back pain. It’s very simple, and it’s very short.

To be fair, the article does mention chiropractic—once. It’s less than a glowing endorsement:

Many treatments are available for low back pain. Often exercises and physical therapy can help. Some people benefit from chiropractic therapy or acupuncture. Sometimes medications are needed, including analgesics (painkillers) or medications that reduce inflammation. Surgery is not usually needed but may be considered if other therapies have failed.

All in all, it’s pretty vague, and saying that “some people” benefit says nothing about evidence or science. It’s actually not a very good article, at least not with respect to discussing treatment. More importantly, just because JAMA published a brief Patient Page on low back pain does not mean that JAMA endorses chiropractic, any more than a medical journal publishing on any topic implies endorsement. Vidan is suffering from self-delusion at best and is being disingenuous at worst. He should know that publishing an article in a journal doesn’t necessarily imply endorsement of the concepts of that article by the journal. If Vidan doesn’t know that, he’s more clueless than the average chiropractor.

But what about this further claim:

The JAMA`s recommendation comes on the heels of a recent study out of the medical journal Spine where sufferers of lower back pain all received standard medical care (SMC) and half of the participants additionally received chiropractic care. The researchers found that in SMC plus chiropractic care patients, 73% reported that their pain was completely gone or much better after treatment compared to just 17% of the SMC group.

Again, this is something that irritates the hell out of me: News articles that mention a study but don’t give me any easy way of identifying it, no link, no author, no title, no anything. As far as I can tell, this article from November appears to be the right article. it’s from the Palmer Center for Chiropractic Research, Davenport, IA; Physical Medicine and Integrative Care Services, Fort Bliss, TX; Samueli Institute, Alexandria, VA; Palmer College of Chiropractic, Davenport, IA; TriMax Direct, Saint Paul, MN; and Palmer College of Chiropractic, San Jose, CA, and entitled Adding chiropractic manipulative therapy to standard medical care for patients with acute low back pain: results of a pragmatic randomized comparative effectiveness study. It’s a randomized trial of standard medical care (SMC) versus standard medical care plus chiropractic manipulative therapy (SMC+CMT).

Pragmatic trial. It had to be a pragmatic trial.

I’ve discussed pragmatic trials before. Alternative medicine advocates love them because they don’t have rigorous controls. “pragmatic trials” are basically an attempt to determine whether treatments validated in properly designed randomized clinical trials (RCTs) work under “real world” conditions. RCTs are intentionally designed to make the population studied as homogeneous as possible, both to minimize differences between the control group and the experimental groups and to decrease variability within groups, the better to isolate the signal from the difference between treatment and control. However, once a treatment gets out into the community, it becomes more widely used, and the rigid inclusion and exclusion criteria used to select subjects for clinical trials fly out the window. The patients upon whom the treatment is used become much less homogeneous, and differences between academic medical centers and the community can change how the treatment is delivered. So “pragmatic” trials seek to determine effectiveness in the real world, which is a different thing than the efficacy determined in the rarified, tightly controlled world of RCTs. Here’s the problem. Pragmatic trials in CAM are putting the cart before the horse. You need to demonstrate efficacy in RCTs before it’s appropriate to consider doing pragmatic trials to determine real world effectiveness. There’s a reason why the National Center for Complementary and Alternative Medicine (NCCAM) loves pragmatic trials.

One way to look at pragmatic trials is that they often don’t have a placebo control. That’s one of their big problem in many pragmatic trials, and it’s definitely a problem in this trial, which studies two groups, SMC (N=46) and SMC + chiropractic (N=45). SMC consisted of this:

The study did not restrict access to SMC or prescribe a SMC delivery protocol. Thus, both groups had normal access to the SMC typically provided to patients with LBP at WBAMC. Standard care included any or all of the following: a focused history and physical examination, diagnostic imaging as indicated, education about self-management including maintaining activity levels as tolerated, pharmacological management with the use of analgesics and anti-inflammatory agents, and physical therapy and modalities such as heat/ice and referral to a pain clinic.


The number of visits in the SMC group was in the range of 0 to 8, with a mean of 1.4 visits. The majority of participants (n = 24) in this group had only 1 visit. Medications were prescribed for 37% of the participants and included nonsteroidal anti-inflammatory drugs, muscle relaxants, benzodiazepines, analgesic creams, and narcotics. Thirty-three percent were placed on a treatment plan (exercise program, range of motion, stretching and modalities including heat and electrical stimulation) delivered primarily by a physical therapist. Fifty percent were given referrals, with a majority for physical therapy (38%) followed by radiographical evaluation (31%). The SMC group providers were physician assistants (28%), family practice physicians (18%), physical therapists (16%) or aides (12%), nurse practitioners (9%), or specialty providers (physical medicine [3%], athletic trainer [3%], and chiropractor [3%]).

Inclusion criteria included male and female US active-duty military personel between 18 and 35 years of age with acute low back pain (LBP). Acute LBP is defined as LBP of less than four weeks of duration. Exclusion criteria were pretty minor, including not being at the post for at least six weeks, LBP for more than four weeks, pregnancy, or other conditions for which CMT is contraindicated. The specific type of chiropractic manipulation was specified in the study criteria. Outcomes studied included pain measured using the numerical rating scale and physical functioning measured using the Roland-Morris Disability questionnaire (RMQ) and the Back Pain Functional Scale (BPFS). The NRS asks participants to rate their level of pain during the past 24 hours on an ordinal 11-point scale (0 = no LBP; 10 = worst possible pain). The minimal clinically important difference is a change of 2.5 points. 20 The modified RMQ assesses LBP-related disability and the minimal clinically important difference is estimated at 2 to 3.5 points. Finally, secondary outcomes included satisfaction, measured on the same 0-10 point scale as the pain scale to respond to the question, “How satisfied were you with the overall results of your care?” Patients were also ased to rate their improvement on a seven-point Likert scale answering the question, “Compared to your first visit, your back pain is…”

No attempt was made to blind treating personnel or research subjects, although the principal investigator and data analysts were blinded to subject group.

So what were the results? They were pretty much what one would expect for a trial like this. Both groups improved. (After all, most acute back pain resolves on its own.) However, the SMC+CMT group improved a bit more. For instance, the pain scores for the SMC group was 6.1 and 5.2 at two and four weeks, respectively; for the SMC+CMT group, the scores were 3.9 and 3.9 at two and four weeks respectively, for a difference of 2.2 and 1.2. Remember, the minimally clinically important difference in pain scores is considered to be 2.5 points, which tells us that these differences were short-lived and probably not clinically significant. Differences were somewhat higher for the RMQ and BPFS.

Let me note one thing that stood out. Both treatment groups got very few visits to the conventional medical care practitioners. For instance, the number of visits in the SMC group ranged from 0 to 8, with a mean of 1.4 visits, but the majority of participants had only one visit. The same was true of the SMC+CMT group, whose visit number ranged from 0 to 4 for conventional therapy, with a mean of 1. In contrast, in this group, this group had a median of 7 visits for CMT, with a range from 2-8, which tells me that a lot of them got 7 or 8 visits in four weeks. This is a huge difference in contact with conventional therapists. Thus, even though 50% of subjects in SMC got referrals to physical therapy, their visit numbers tell me that few went, or at least went very often. Another big problem with this study is the disproportionate loss to followup. The loss to followup was only 15% in the SMC+CMT group and 35% in the SMC group. Although an analysis was done to try to see if this affected the results, bias due to this disproportionate dropout can’t be ruled out. Also, the investigators didn’t actually track medication use between the two groups.

Now here’s the result that was cherry-picked in the interview with Vidan:

Seventy-three percent of participants in the SMC plus CMT group rated their global improvement as pain completely gone, much better, or moderately better, compared with 17% in the SMC group (Figure 2). The mean satisfaction with care score on a 0 to 10 scale for the SMC plus CMT group was 8.9 at both weeks 2 and 4; the mean for the SMC group was 4.5 at week 2 and 5.4 at week 4 (Table 4).

Of course they did better. They got manipulation similar to physical therapy seven times during the eight weeks, while most of the control population didn’t. And of course they were more satisfied. They saw a practitioner more often and got an active intervention. Obviously, chiropractic is not physical therapy, but in this study clearly everyone in one group got some form of spinal manipulation on a regular basis (roughly twice a week), while most people in the control group did not get anywhere near that level of intervention. The real “pragmatic” trial would have been take physical therapy (PT) out of SMC for purposes of this study and study it as a separate variable, as in SMC, SMC+PT, and SMC+CMT. (Remember, I’m proposing what I’d consider to be a more rigorous pragmatic trial.)

I also note that this is one trial, and that its results conflict with another similar trial done published in 2009 that found that pain reductions and analgesic consumption were similar in all groups and concluded that spinal manipulative therapy “is unlikely to result in relevant early pain reduction in patients with acute low back pain.” Funny how Vidan never mentioned this.

And, of course, a rather weak article in JAMA that mentions chiropractic as a treatment for low back pain does not mean that JAMA recommends chiropractic first.