Dodgy science abounds in alternative medicine; this is perhaps particularly true for homeopathy. A brand-new trial seems to confirm this view.
The aim of this study was to test the hypothesis that homeopathy (H) enhances the effects of scaling and root planing (SRP) in patients with chronic periodontitis (CP).
The researchers, dentists from Brazil, randomised 50 patients with CP to one of two treatment groups: SRP (C-G) or SRP + H (H-G). Assessments were made at baseline and after 3 and 12 months of treatment. The local and systemic responses to the treatments were evaluated after one year of follow-up. The results showed that both groups displayed significant improvements, however, the H-G group performed significantly better than C-G group.
The authors concluded that homeopathic medicines, as an adjunctive to SRP, can provide significant local and systemic improvements for CP patients.
Really? I am afraid, I disagree!
Homeopathic medicines might have nothing whatsoever to do with this result. Much more likely is the possibility that the findings are caused by other factors such as:
- patients’ expectations,
- improved compliance with other health-related measures,
- the researchers’ expectations,
- the extra attention given to the patients in the H-G group,
- disappointment of the C-G patients for not receiving the additional care,
- a mixture of all or some of the above.
I should stress that it would not have been difficult to plan the study in such a way that these factors were eliminated as sources of bias or confounding. But this study was conducted according to the A+B versus B design which we have discussed repeatedly on this blog. In such trials, A is the experimental treatment (homeopathy) and B is the standard care (scaling and root planning). Unless A is an overtly harmful therapy, it is simply not conceivable that A+B does not generate better results than B alone. The simplest way to comprehend this argument is to imagine A and B are two different amounts of money: it is impossible that A+B is not more that B!
It is unclear to me what relevant research question such a study design actually does answer (if anyone knows, please tell me). It seems obvious, however, that it cannot test the hypothesis that homeopathy (H) enhances the effects of scaling and root planing (SRP). This does not necessarily mean that the design is necessarily useless. But at the very minimum, one would need an adequate research question (one that matches this design) and adequate conclusions based on the findings.
The fact that the conclusions drawn from a dodgy trial are inadequate and misleading could be seen as merely a mild irritation. The facts that, in homeopathy, such poor science and misleading conclusions emerge all too regularly, and that journals continue to publish such rubbish are not just mildly irritating; they are annoying and worrying – annoying because such pseudo-science constitutes an unethical waste of scarce resources; worrying because it almost inevitably leads to wrong decisions in health care.
“If ever there was a permanent cure for migraine, homeopathic medicines are the only one that can do this miracle. It may sound like an overstatement and quite quackerish, but it’s true. Long term treatment with homeopathy has an excellent cure for migraine headaches.” Statements like this can be found by the thousands on the internet, not just in relation to migraine but also about osteoarthritis. Both migraine and osteoarthritis are important domains for homeopathy, and most homeopaths would not doubt for a second that they can treat these conditions effectively. This is why it is so important to highlight the few sources which are not misleading consumers into making the wrong therapeutic decisions.
‘Healthcare Improvement Scotland’ (HCIS) have just published advice for patients suffering from migraine and osteoarthritis (the full document with all the evidence can be found here). I think it is worth having a close look and I therefore cite it in full:
Homeopathic remedies are prepared by repeated dilution and vigorous shaking of substances in water. Remedies are prepared from substances that in healthy people cause the signs and symptoms of the condition being treated. The more dilute the remedy is the more potent it becomes so that the most potent remedies are unlikely to contain any of the original substance.
People in Scotland have access to homeopathy through some GPs or a referral to homeopaths in the private sector, regional NHS clinics or the Centre for Integrative Care (CIC) (formerly Glasgow Homeopathic Hospital). Not all NHSScotland health boards provide funding for homeopathy; investment varies widely among those that do, and individual boards have begun to review funding for homeopathy services.
- Evidence of clinical effectiveness was reviewed from systematic reviews of four placebo controlled randomised trials of homeopathy for migraine published between 1991 and 1997; and systematic reviews of four active treatment controlled randomised trials of homeopathy for osteoarthritis published between 1983 and 2000. The quality of the evidence was low to moderate.
- Homeopathy for migraine has not been compared with active treatment in randomised controlled trials (RCTs). Of four RCTs comparing homeopathy with placebo, only one found homeopathy to be superior.
- Three RCTs in osteoarthritis comparing homeopathy with medicines for pain relief found either no difference between the interventions, or that analgesic treatment had a better effect than homeopathy. A further RCT comparing intra-articular injection of a homeopathic remedy with hyaluronic acid injections showed similar pain reduction in both groups.
- Published systematic reviews of homeopathy for migraine and osteoarthritis contain insufficient information to inform conclusions about safety.
- No evidence on the cost effectiveness of homeopathy for migraine was identified; and the evidence from a single cost-minimisation analysis of one homeopathic preparation for osteoarthritis is not generalisable to the UK.
- Homeopathy for migraine has not been compared with standard care in RCTs and no evidence of cost effectiveness has been identified..
- There is insufficient evidence to determine whether or not homeopathic treatment for osteoarthritis is clinically effective compared with standard care, and no relevant evidence of cost effectiveness has been identified.
- The evidence does not support treating migraine or osteoarthritis with homeopathy.
Before the fans of homeopathy start shouting “THIS IS ALL RUBBISH AND DISREGARDS IMPORTANT EVIDENCE!!!”, I should mention that the top experts in homeopathy were asked to contribute their evidence and were unable to find any convincing data that would have changed this negative verdict. And it is important to point out that HCIS is a respected, independent organisation that issues statements based on thorough, unbiased reviews of the evidence.
As I reported a while ago, the Australian ‘NATIONAL HEALTH AND MEDICAL RESEARCH COUNCIL’ has assessed the effectiveness of homeopathy. The evaluation looks like the most comprehensive and most independent in the history of homeopathy. Its draft report concluded that “the evidence from research in humans does not show that homeopathy is effective for treating the range of health conditions considered.”
So, the HCIS is in excellent company and I have no doubt whatsoever that this new statement is correct – but I also have little doubt that homeopaths will dispute it.
Blinding patients in clinical trials is a key methodological procedure for minimizing bias and thus making sure that the results are reliable. In alternative medicine, blinding is not always straight forward, and many studies are therefore not patient-blinded. We all know that this can introduce bias into a trial, but how large is its effect on study outcomes?
This was the research question addressed by a recent systematic review of randomized clinical trials with one sub-study (i.e. experimental vs control) involving blinded patients and another, otherwise identical, sub-study involving non-blinded patients. Within each trial, the researchers compared the difference in effect sizes (i.e. standardized mean differences) between the two sub-studies. A difference <0 indicates that non-blinded patients generated a more optimistic effect estimate. The researchers then pooled the differences with random-effects inverse variance meta-analysis, and explored reasons for heterogeneity.
The main analysis included 12 trials with a total of 3869 patients. Ten of these RCTs were studies of acupuncture. The average difference in effect size for patient-reported outcomes was -0.56 (95% confidence interval -0.71 to -0.41), (I(2 )= 60%, P = 0.004), indicating that non-blinded patients exaggerated the effect size by an average of 0.56 standard deviation, but with considerable variation. Two of the 12 trials also used observer-reported outcomes, showing no indication of exaggerated effects due lack of patient blinding.
There was an even larger effect size difference in the 10 acupuncture trials [-0.63 (-0.77 to -0.49)], than in the two non-acupuncture trials [-0.17 (-0.41 to 0.07)]. Lack of patient blinding was also associated with increased attrition rates and the use of co-interventions: ratio of control group attrition risk 1.79 (1.18 to 2.70), and ratio of control group co-intervention risk 1.55 (0.99 to 2.43).
The authors conclude that this study provides empirical evidence of pronounced bias due to lack of patient blinding in complementary/alternative randomized clinical trials with patient-reported outcomes.
This is a timely, rigorous and important analysis. In alternative medicine, we currently see a proliferation of trials that are not patient-blinded. We always suspected that they are at a high risk of generating false-positive results – now we know that this is, in fact, the case.
What should we do with this insight? In my view, the following steps would be wise:
- Take the findings from the existing trials that are devoid of patient-blinding with more than just a pinch of salt.
- Discourage the funding of future studies that fail to include patient-blinding.
- If patient-blinding is truly and demonstrably impossible – which is not often the case – make sure that the trialists at least include blinding of the assessors of the primary outcome measures.
There must be well over 10 000 clinical trials of acupuncture; Medline lists ~5 000, and many more are hidden in the non-Medline listed literature. That should be good news! Sadly, it isn’t.
It should mean that we now have a pretty good idea for what conditions acupuncture is effective and for which illnesses it does not work. But we don’t! Sceptics say it works for nothing, while acupuncturists claim it is a panacea. The main reason for this continued controversy is that the quality of the vast majority of these 10 000 studies is not just poor, it is lousy.
“Where is the evidence for this outraging statement???” – I hear the acupuncture-enthusiasts shout. Well, how about my own experience as editor-in-chief of FACT? No? Far too anecdotal?
How about looking at Cochrane reviews then; they are considered to be the most independent and reliable evidence in existence? There are many such reviews (most, if not all [co-]authored by acupuncturists) and they all agree that the scientific rigor of the primary studies is fairly awful. Here are the crucial bits of just the last three; feel free to look for more:
Or how about providing an example? Good idea! Here is a new trial which could stand for numerous others:
This study was performed to compare the efficacy of acupuncture versus corticosteroid injection for the treatment of Quervain’s tendosynovitis (no, you do not need to look up what condition this is for understanding this post). Thirty patients were treated in two groups. The acupuncture group received 5 acupuncture sessions of 30 minutes duration. The injection group received one methylprednisolone acetate injection in the first dorsal compartment of the wrist. The degree of disability and pain was evaluated by using the Quick Disabilities of the Arm, Shoulder, and Hand (Q-DASH) scale and the Visual Analogue Scale (VAS) at baseline and at 2 weeks and 6 weeks after the start of treatment. The baseline means of the Q-DASH and the VAS scores were 62.8 and 6.9, respectively. At the last follow-up, the mean Q-DASH scores were 9.8 versus 6.2 in the acupuncture and injection groups, respectively, and the mean VAS scores were 2 versus 1.2. Thus there were short-term improvements of pain and function in both groups.
The authors drew the following conclusions: Although the success rate was somewhat higher with corticosteroid injection, acupuncture can be considered as an alternative option for treatment of De Quervain’s tenosynovitis.
The flaws of this study are exemplary and numerous:
- This should have been a study that compares two treatments – the technical term is ‘equivalence trial – and such studies need to be much larger to produce a meaningful result. Small sample sizes in equivalent trials will always make the two treatments look similarly effective, even if one is a pure placebo.
- There is no gold standard treatment for this condition. This means that a comparative trial makes no sense at all. In such a situation, one ought to conduct a placebo-controlled trial.
- There was no blinding of patients; therefore their expectation might have distorted the results.
- The acupuncture group received more treatments than the injection group; therefore the additional attention might have distorted the findings.
- Even if the results were entirely correct, one cannot conclude from them that acupuncture was effective; the notion that it was similarly ineffective as the injections is just as warranted.
These are just some of the most fatal flaws of this study. The sad thing is that similar criticisms can be made for most of the 10 000 trials of acupuncture. But the point here is not to nit-pick nor to quack-bust. My point is a different and more serious one: fatally flawed research is not just a ‘poor show’, it is unethical because it is a waste of scarce resources and, even more importantly, an abuse of patients for meaningless pseudo-science. All it does is it misleads the public into believing that acupuncture might be good for this or that condition and consequently make wrong therapeutic decisions.
In acupuncture (and indeed in most alternative medicine) research, the problem is so extremely wide-spread that it is high time to do something about it. Journal editors, peer-reviewers, ethics committees, universities, funding agencies and all others concerned with such research have to work together so that such flagrant abuse is stopped once and for all.
We all know, I think, that chronic low back pain (CLBP) is common and causes significant suffering in individuals as well as cost to society. Many treatments are on offer but, as we have seen repeatedly on this blog, not one is convincingly effective and some, like chiropractic, is associated with considerable risks.
Enthusiasts claim that hypnotherapy works well, but too little is known about the minimum dose needed to produce meaningful benefits, the roles of home practice and hypnotizability on outcome, or the maintenance of treatment benefits beyond 3 months. A new trial was aimed at addressing these issues.
One hundred veterans with CLBP participated in a randomized, four parallel group study. The groups were (1) an eight-session self-hypnosis training intervention without audio recordings for home practice; (2) an eight-session self-hypnosis training intervention with recordings; (3) a two-session self-hypnosis training intervention with recordings and brief weekly reminder telephone calls; and (4) an eight-session active (biofeedback) control intervention.
Participants in all four groups reported significant pre- to post-treatment improvements in pain intensity, pain interference and sleep quality. The three hypnotherapy groups combined reported significantly more pain intensity reduction than the control group. There was no significant difference among the three hypnotherapy groups. Over half of the participants who received hypnotherapy reported clinically meaningful (≥30%) reductions in pain intensity, and they maintained these benefits for at least 6 months after treatment. Neither hypnotizability nor amount of home practice was associated significantly with treatment outcome.
The authors conclude that two sessions of self-hypnosis training with audio recordings for home practice may be as effective as eight sessions of hypnosis treatment. If replicated in other patient samples, the findings have important implications for the application of hypnosis treatment for chronic pain management.
Even though this trial has several important limitations, I do agree with the authors: these results would be worth an independent replication – not least because self-hypnosis is cheap and does not carry great risks. What would be interesting, in my view, are studies that compare several alternative LBP therapies (e.g. chiropractic, osteopathy, acupuncture, massage, various form of exercise and hypnotherapy) in terms of cost, risks, long-term effectiveness and patients’ preference. I somehow feel that the results of such comparative trials might overturn the often issued recommendations for spinal manipulation, i.e. chiropractic or osteopathy.
Niacin – also known as vitamin B3 or nicotinic acid - is a natural compound (formula C
2 ) and an essential nutrient for humans. It is water-soluble, which means it is not stored in the body. Excess amounts of the vitamin leave the body through the urine. That means we need a continuous supply of niacin in your diet.
Niacin is found in variety of foods, including liver, chicken, beef, fish, cereal, peanuts and legumes. It can also be synthesized from tryptophan, an essential amino acid found in protein. Niacin has long been an accepted treatment for high cholesterol. It is well-documented to increase the levels of high-density cholesterol (HDL or “good cholesterol”) and to decrease the levels of low-density cholesterol (LDL or “bad cholesterol”).
But what do these effects really mean? Do they translate into true health benefits? A brand-new study casts doubt on the value of niacin therapy:
After a pre-randomization run-in phase to standardize the background statin-based LDL cholesterol-lowering therapy and to establish participants’ ability to take extended-release niacin without clinically significant adverse effects, the researchers randomly assigned 25,673 adults with vascular disease to receive 2 g of extended-release niacin and 40 mg of laropiprant or a matching placebo daily. The primary outcome was the first major vascular event (non-fatal myocardial infarction, death from coronary causes, stroke, or arterial revascularization).
During a median follow-up period of 3.9 years, participants who were assigned to extended-release niacin-laropiprant had an LDL cholesterol level that was 10 mg per deciliter (0.25 mmol/l) lower and an HDL cholesterol level that was 6 mg per deciliter (0.16 mmol/l) higher than the levels in those assigned to placebo. Thus the lipid-effects of previous studies were confirmed.
However, assignment to niacin-laropiprant, as compared with assignment to placebo, had no significant effect on the incidence of major vascular events Niacin-laropiprant was associated with an increased incidence of disturbances in diabetes control that were considered to be serious and with an increased incidence of diabetes diagnoses as well as increases in serious adverse events associated with the gastrointestinal system, the musculoskeletal system, the skin and infections and bleeding.
Based of these data, the authors arrived at the following conclusion: among participants with atherosclerotic vascular disease, the addition of extended-release niacin-laropiprant to statin-based LDL cholesterol-lowering therapy did not significantly reduce the risk of major vascular events but did increase the risk of serious adverse events.
This extremely well-done trial is a poignant reminder of the fact that, in health care, we must never take our assumptions for granted. Here the underlying assumption was that the Niacin-induced lipid changes lead to a reduction of cardiovascular risks. Not only it proved to be erroneous but, through serious adverse effects, Niacin actually decreased patients’ health status.
The lessons from all this are straight forward, I think:
- ‘Natural’ does not necessarily mean safe.
- Long-established does not necessarily mean efficacious.
- Assumptions are merely assumptions, nothing more; if we want to make sure that they hold, we need to test them.
- When we finally do test assumptions, we better do it rigorously.
Some chiropractors claim that their main intervention, spinal manipulation, works for nonspecific neck pain by improving inter-vertebral range of motion (IV-RoM). But IV-RoM is difficult to measure, and whether it is related to clinical outcomes seems uncertain. Researchers from the Institute of Musculoskeletal Research & Clinical Implementation and the Anglo-European College of Chiropractic have just published a study that might throw some light on this issue. According to its authors, it was aimed at answering the following research questions:
- Does cervical spine flexion and extension IV-RoM increase after a course of spinal manipulation?
- Is there a relationships between any IV-RoM increases and clinical outcomes?
- How does palpation compare with objective measurement in the detection of hypo-mobile segments?
Thirty patients with nonspecific neck pain and 30 healthy controls matched for age and gender received quantitative fluoroscopy (QF) screenings to measure flexion and extension IV-RoM (C1-C6) at baseline and 4-week follow-up. Patients received up to 12 neck manipulations and completed NRS, NDI and Euroqol 5D-5L at baseline, plus PGIC and satisfaction questionnaires at follow-up. IV-RoM accuracy, repeatability and hypo-mobility cut-offs were determined. Minimal detectable changes (MDC) over 4 weeks were calculated from controls. Patients and control IV-RoMs were compared at baseline as well as changes in patients over 4 weeks. Correlations between outcomes and the number of manipulations received and the agreement (Kappa) between palpated and QF-detected of hypo-mobile segments were calculated.
QF had high accuracy (worst RMS error 0.5σ) and repeatability (highest SEM 1.1σ, lowest ICC 0.9σ) for IV-RoM measurement. Hypo-mobility cut offs ranged from 0.8σ to 3.5σ. No outcome was significantly correlated with increased IV-RoM above MDC and there was no significant difference between the number of hypo-mobile segments in patients and controls at baseline or significant increases in IV-RoMs in patients. However, there was a modest and significant correlation between the number of manipulations received and the number of levels and directions whose IV-RoM increased beyond MDC (Rho=0.39, p=0.043). There was also no agreement between palpation and QF in identifying hypo-mobile segments (Kappa 0.04-0.06).
The authors concluded that this study found no differences in cervical sagittal IV-RoM between patients with non-specific neck pain and matched controls. There was a modest dose-response relationship between the number of manipulations given and number of levels increasing IV-RoM – providing evidence that neck manipulation has a mechanical effect at segmental levels. However, patient-reported outcomes were not related to this.
This conclusion seems a little odd to me. In my view the study suggests a clearly negative answer to all the three research questions formulated above. An interesting paragraph from the authors’ discussion section provides further insight: The lack of a relationship between symptomatic improvement and increased IV-RoM is also of interest. Clearly other mechanisms that improved the comfort and functional capacity of the patients in this study were in play, including spontaneous recovery. Other important biological factors may have included chemical factors in joint and muscle and activation patterns in the latter. However, this study seemed to rule out central pain hypersensitivity as a factor, as this was not detected at baseline in any of the patients. Psychological and social factors and their influence on functional behavior may also have had a role and may have been influenced by the interventions received.
So, spinal manipulation does not seem to work by improving IV-RoM. Could this be because spinal manipulation does not work at all?
‘Red ginseng’ is an herbal medicine prepared by steaming raw ginseng. This process is believed to increase its pharmacological activity. Further conversion through fermentation is thought to increase its intestinal absorption and bioactivity to diminish its toxicity.
Red ginseng (RG) is traditionally used for diabetes. Our own systematic review of 4 RCTs concluded that the evidence for the effectiveness of RG in controlling glucose in type 2 diabetes is not convincing. Few included studies with various treatment regimens prohibit definitive conclusions. More rigorous studies are needed to clarify the effects of RG on this condition.
Now a new RCT has become available. This study was conducted to investigate the effects of daily supplementation with fermented red ginseng (FRG) on blood sugar levels in subjects with impaired fasting glucose or type 2 diabetes. It was a four-week long, randomized, double-blind, placebo-controlled trial. Forty-two subjects with impaired fasting glucose or type 2 diabetes were randomly allocated to two groups assigned to consume either placebo or FRG three times per day for 4 weeks. Fasting and postprandial glucose profiles during meal tolerance tests were assessed before and after the intervention.
Compared to the placebo, FRG supplementation led to a significant reduction in postprandial glucose levels and to an increase in postprandial insulin levels. There also was a significant improvement in the area under the curve (AUC) in the FRG group. However, fasting glucose, insulin, and lipid profiles did not differ from the placebo group.
The authors of this trial concluded that daily supplementation with FRG lowered postprandial glucose levels in subjects with impaired fasting glucose or type 2 diabetes.
What should we make of these findings? Do they indicate that FRG might be an alternative to conventional anti-diabetic drugs? I would caution that we have tons of data for the latter, while we know far too little about FRG to recommend it for routine use.
On the contrary, the findings could suggest that diabetic patients who are well-controlled with diet or anti-diabetic medication should be avoiding ginseng products. If they actually work, they might significantly interfere with their metabolic control which, in turn, could even endanger their lives.
What is the best treatment for the millions of people who suffer from chronic low back pain (CLBP)? If we are honest, no therapy has yet been proven to be overwhelmingly effective. Whenever something like that happens in medicine, we have a proliferation of interventions which all are promoted as effective but which, in fact, work just marginally. And sure enough, in the case of CLBP, we have a constantly growing list of treatments none of which is really convincing.
One of the latest additions to this list is PILATES.
Pilates? What is this ? One practitioner describes it as follows: In Pilates, we pay a lot of attention to how our body parts are lined up in relation to each other, which is our alignment. We usually think of our alignment as our posture, but good posture is a dynamic process, dependent on the body’s ability to align its parts to respond to varying demands effectively. When alignment is off, uneven stresses on the skeleton, especially the spine, are the result. Pilates exercises, done with attention to alignment, create uniform muscle use and development, allowing movement to flow through the body in a natural way.
For example, one of the most common postural imbalances that people have is the tendency to either tuck or tilt the pelvis. Both positions create weaknesses on one side of the body and overly tight areas on the other. They deny the spine the support of its natural curves and create a domino effect of aches and pains all the way up the spine and into the neck. Doing Pilates increases the awareness of the proper placement of the spine and pelvis, and creates the inner strength to support the natural curves of the spine. This is called having a neutral spine and it has been the key to better backs for many people.
Mumbo-jumbo? Perhaps; in any case, we need evidence! Is there any at all? Surprisingly, the answer is yes. Recently, someone even published a proper systematic review.
This systematic review was aimed at evaluating the effectiveness of Pilates exercise in people with chronic low back pain (CLBP).
A search for RCTs was undertaken in 10 electronic. Two independent reviewers did the selection of evidence and evaluated the quality of the primary studies. To be included, relevant RCTs needed to be published in the English language. From 152 studies, 14 RCTs could be included.
The methodological quality of RCTs ranged from “poor” to “excellent”. A meta-analysis of RCTs was not undertaken due to the heterogeneity of RCTs. Pilates exercise provided statistically significant improvements in pain and functional ability compared to usual care and physical activity between 4 and 15 weeks, but not at 24 weeks. There were no consistent statistically significant differences in improvements in pain and functional ability with Pilates exercise, massage therapy, or other forms of exercise at any time period.
The authors drew the following conclusions: Pilates exercise offers greater improvements in pain and functional ability compared to usual care and physical activity in the short term. Pilates exercise offers equivalent improvements to massage therapy and other forms of exercise. Future research should explore optimal Pilates exercise designs, and whether some people with CLBP may benefit from Pilates exercise more than others.
So, Pilates can be added to the long list of treatments that work for CLBP, albeit not convincingly better than most other therapies on offer. Does that mean these options are all as good or as bad as the next? I don’t think so.
Let’s assume chiropractic/osteopathic manipulations, massage and various forms of exercise are all equally effective. How do we decide which is more commendable than the next? We clearly need to take other important factors into account:
- acceptability for patients
If we use these criteria, it becomes instantly clear that chiropractic and osteopathy are not favourites in this race for the most commendable CLBP-treatment. They are neither cheap nor free of risks. Massage is virtually risk-free but not cheap. This leaves us with various forms of exercise, including Pilates. But which exercise is better than the next? At present, we do not know, and therefore the last two factors are crucial: if people love doing Pilates and if they easily stick with it, then Pilates is fine.
I am sure chiropractors will (yet again) disagree with me but, to me, this logic could hardly be more straight forward.
There is some (albeit not compelling) evidence to suggest that chiropractic spinal manipulation might be effective for treating non-specific back pain. But what about specific back pain, such as the one caused by a herniated disc? Some experts believe that, in patients suffering from such a condition, manipulations are contra-indicated (because the latter can cause the former), while others think that manipulation might be an effective treatment option (although the evidence is far from compelling). Who is correct? The issue can only be resolved with data from well-designed clinical investigations. A new trial might therefore enlighten us.
The stated purposes of this study were:
- to evaluate patients with low-back pain (LBP) and leg pain due to magnetic resonance imaging-confirmed disc herniation treated with high-velocity, low-amplitude spinal manipulation in terms of their short-, medium-, and long-term outcomes of self-reported global impression of change and pain levels
- to determine if outcomes differ between acute and chronic patients using.
The researchers conducted a ‘prospective cohort outcomes study‘ with 148 patients with LBP, leg pain, and physical examination abnormalities with concordant lumbar disc herniations. Baseline numerical rating scale (NRS) data for LBP, leg pain, and the Oswestry questionnaire were obtained. The specific lumbar spinal manipulation was dependent upon whether the disc herniation was intraforaminal or paramedian as seen on the magnetic resonance images and was performed by a chiropractor. Outcomes included the patient’s global impression of change scale for overall improvement, the NRS for LBP, leg pain, and the Oswestry questionnaire at 2 weeks, 1, 3, and 6 months, and 1 year. The proportion of patients reporting “improvement” on the patient’s global impression of change scale was calculated for all patients and for acute vs chronic patients. Pre-treatment and post-treatment NRS scores were compared using the paired t test. Baseline and follow-up Oswestry scores were compared using the Wilcoxon test. Numerical rating scale and Oswestry scores for acute vs chronic patients were compared using the unpaired t test for NRS scores and the Mann-Whitney U test for Oswestry scores.
Significant improvements for all outcomes at all time points were reported. At 3 months, 91% of patients were “improved”, and 88% were “improved” after 1 year. Acute patients improved faster by 3 months than did chronic patients. 81.8% of chronic patients 89.2% felt “improved” at 1 year. No adverse events were reported.
The researchers concluded that a large percentage of acute and importantly chronic lumbar disc herniation patients treated with chiropractic spinal manipulation reported clinically relevant improvement.
Does this new study meaningfully contribute to our knowledge about the effectiveness of chiropractic manipulation for back pain caused by herniated discs? The short answer to this question is NO.
A longer answer might be that the report does tell us something relevant about the quality of this research project. We know from countless studies that ~50% of patients experience adverse effects after spinal manipulations by a chiropractor. This means that any report claiming that NO ADVERSE EFFECTS WERE REPORTED is puzzling to a degree that we have to seriously question its quality or even honesty. In this context, it is relevant to mention that a recent review of the evidence concluded that a cause-effect relationship exists between the manipulative treatment and the development of disc herniation.
The positive outcomes reported in this new study could, of course, be due to a range of factors which are unrelated to the manipulations administered by the chiropractors:
- natural history of disc herniation
- regression towards the mean
- other treatments employed by the patients
- social desirability
To be able to say with any degree of certainty that the manipulations had anything to do with the observed positive outcomes would require an entirely different study-design. Should we assume that this is not known in the world of chiropractic? Or should we consider that chiropractors shy away from rigorous research because they fear its results?
The term prospective cohort outcomes study, seems to be a chiropractic invention (cohort studies are by definition prospective, and observational studies are usually prospective). It seems that, behind this long and impressive word, one can easily hide the fact that this study design fails to make the slightest attempt of controlling for non-specific effects; the term sounds scientific – but when we analyse what it means, we discover that this methodology is little more than a self-serving consumer survey. Most scientists would call such an investigation quite simply an OBSERVATIONAL STUDY.
I think it is time that chiropractors start doing proper research which actually does answer some of the many open questions regarding spinal manipulation.