The methodology flaws in Hinman’s acupuncture clinical trial, Part II: Zelen design and effectiveness dilutionsco.


In the October 2014 publication of JAMA, Dr. Hinman and colleagues published the study “Acupuncture for Chronic Knee Pain: A Randomized Clinical Trial,” which concluded that “in patients older than 50 years with moderate or severe chronic knee pain, neither laser nor needle acupuncture conferred benefit over sham for pain or function. Our findings do not support acupuncture for these patients[1].” As pointed out in my former article, Part I[2], there were serious flaws in the trial design and statistics, as well as in the interpretation of the results. This article attempts to address problems in the Zelen design used by Hinman et al[1]. There are some advantages to using a Zelen design for a randomized controlled trial (RCT). First, a Zelen design has a post-randomization consent design, which means that consent is only sought for one treatment each time, without the uncertainty of randomization. Researchers can be more comfortable knowing that they have the participants’ consent each time they undergo a treatment. Patients can also be more comfortable with this design because they know which type of treatment they are receiving; unlike traditional RCTs, patients are not ignorant of whether they are receiving the placebo or experimental treatment. Effects such as resentful demoralization and what is known as the “Hawthorne effect” (altered behavior or performance resulting from awareness of being a part of an experimental study) become less of an issue as patients are not weary of being part of a new alternative group, only the “standard” therapy will applies to them. However, it does have some disadvantages, and therefore can cause biases.

1 High drop-out rate:

The drop-out rates were 2.82% (2/71) in the control group; 22.86% (16/70) in the acupuncture group; 18.31% (13/71) in the laser acupuncture group; and 22.86% (16/70) for the sham laser acupuncture group. According to the acceptable standards for an RCT, dropout rates less than 10% are acceptable, drop-out rates between 10% and 20% mean that the resulting data quality is poor, and drop-out rates of more than 20% mean that the data quality is considered very poor and should not be used in analysis. In this trial analysis, the data quality in the acupuncture and sham laser acupuncture groups are very poor as the drop-out rates are over 20%; the authors should not have directly used them in any statistical analysis, unless they had re-adjusted and re-balanced the sample among the groups during the study. As outlined by the National Institutes of Health, if there is a differential drop-out rate of 15% or higher between study arms, such as between the control group and the treatment group in this clinical trial, then there is a very high potential for bias. This is a flaw that can decrease the quality of the study results.

2 The effectiveness in intervention groups was diluted by various factors

The dilution rates should then be 21.87% in the laser acupuncture group, 13.80% in the sham laser acupuncture group, and 31.27% in the acupuncture group (the dilution rate calculations were shown in Tables 1–3). The dilution rate was very significant in the acupuncture group, which causes the effectiveness to be undervalued in the acupuncture group, by almost 1/3.

The effective significance was masked by limited sample size due to the Zelen design of this study.

3.The sample size calculation in this study is questionable.

4 Conclusion The effectiveness of the acupuncture group was diluted 31.27%, and its drop-out rate was 22.86%, much higher than that of the other groups in Hinman’s clinical trial, which constitutes major flaws in how this study is analyzed and interpreted[8]. Based on the bias of Zelen design used in the study, and incorrect sample size calculation, the conclusions drawn from this study are of poor quality, inaccurate, and invalid.



1 Hinman RS, McCrory P, Pirotta M, Relf I, Forbes A, Crossley KM, Williamson E, Kyriakides M, Novy K, Metcalf BR, Harris A, Reddy P, Conaghan PG, Bennell KL. Acupuncture for chronic knee pain: a randomized clinical trial. JAMA. 2014; 312(13): 1313–1322.

2 Fan AY. The methodology flaws in Hinman’s acupuncture clinical trial, Part I: Design and results interpretation. J Integr Med. 2015; 13(2): 65–68.

« How large are the nonspecific effects of acupuncture? A meta-analysis of randomized controlled trials

via Acupuncture for Chronic Pain is effective -Individual Patient Data Meta-analysis says.


Acupuncture for Chronic Pain Individual Patient Data Meta-analysis

Andrew J. Vickers, DPhil; Angel M. Cronin, MS; Alexandra C. Maschino, BS; George Lewith, MD; Hugh MacPherson, PhD; Nadine E. Foster, DPhil; Karen J. Sherman, PhD; Claudia M. Witt, MD; Klaus Linde, MD ; for the Acupuncture Trialists’ Collaboration
Arch Intern Med. 2012;172(19):1444-1453. doi:10.1001/archinternmed.2012.3654.


Although acupuncture is widely used for chronic pain, there remains considerable controversy as to its value. We aimed to determine the effect size of acupuncture for four chronic pain conditions: back and neck pain, osteoarthritis, chronic headache, and shoulder pain.


We conducted a systematic review to identify randomized trials of acupuncture for chronic pain where allocation concealment was determined unambiguously to be adequate. Individual patient data meta-analyses were conducted using data from 29 of 31 eligible trials, with a total of 17,922 patients analyzed.


In the primary analysis including all eligible trials, acupuncture was superior to both sham and no acupuncture control for each pain condition (all p<0.001). After exclusion of an outlying set of trials that strongly favored acupuncture, the effect sizes were similar across pain conditions. Patients receiving acupuncture had less pain, with scores 0.23 (95% C.I. 0.13, 0.33), 0.16 (95% C.I. 0.07, 0.25) and 0.15 (95% C.I. 0.07, 0.24) standard deviations lower than sham controls for back and neck pain, osteoarthritis, and chronic headache respectively; the effect sizes in comparison to no acupuncture controls were 0.55 (95% C.I. 0.51, 0.58), 0.57 (95% C.I. 0.50, 0.64) and 0.42 (95% C.I. 0.37, 0.46). These results were robust to a variety of sensitivity analyses, including those related to publication bias.


Acupuncture is effective for the treatment of chronic pain and is therefore a reasonable referral option. Significant differences between true and sham acupuncture indicate that acupuncture is more than a placebo. However, these differences are relatively modest, suggesting that factors in addition to the specific effects of needling are important contributors to the therapeutic effects of acupuncture.

How large are the nonspecific effects of acupuncture? A meta-analysis of randomized controlled trials.


How large are the nonspecific effects of acupuncture? A meta-analysis of randomized controlled trials

Klaus Linde1*, Karin Niemann1, Antonius Schneider1 and Karin Meissner12

Author Affiliations

1Institute of General Practice, Technische Universität München, Orleansstrasse 47, D-81667 Munich, Germany

2Institute of Medical Psychology, Ludwig-Maximilians-University, Goethestrasse 31, D-80336 Munich, Germany

For all author emails, please log on.

BMC Medicine 2010, 8:75  doi:10.1186/1741-7015-8-75

The electronic version of this article is the complete one and can be found online at:http://www.biomedcentral.com/1741-7015/8/75



While several recent large randomized trials found clinically relevant effects of acupuncture over no treatment or routine care, blinded trials comparing acupuncture to sham interventions often reported only minor or no differences. This raises the question whether (sham) acupuncture is associated with particularly potent nonspecific effects. We aimed to investigate the size of nonspecific effects associated with acupuncture interventions.


MEDLINE, Embase, Cochrane Central Register of Controlled Clinical Trials and reference lists were searched up to April 2010 to identify randomized trials of acupuncture for any condition, including both sham and no acupuncture control groups. Data were extracted by one reviewer and verified by a second. Pooled standardized mean differences were calculated using a random effects model with the inverse variance method.


Thirty-seven trials with a total of 5754 patients met the inclusion criteria. The included studies varied strongly regarding patients, interventions, outcome measures, methodological quality and effect sizes reported. Among the 32 trials reporting a continuous outcome measure, the random effects standardized mean difference between sham acupuncture and no acupuncture groups was -0.45 (95% confidence interval, -0.57, -0.34; I2 = 54%; Egger’s test for funnel plot asymmetry, P= 0.25). Trials with larger effects of sham over no acupuncture reported smaller effects of acupuncture over sham intervention than trials with smaller nonspecific effects (β = -0.39, P = 0.029).


Sham acupuncture interventions are often associated with moderately large nonspecific effects which could make it difficult to detect small additional specific effects. Compared to inert placebo interventions, effects associated with sham acupuncture might be larger, which would have considerable implications for the design and interpretation of clinical trials.

BMJ Editors Hit–and Hit Hard–Over Careless Interpretation of Acupuncture for Chronic Knee Pain Study (JAMA 10/1/2014)

Let’s start with the conclusion, lest I lose you between here and the end.

Acupuncture is more likely to provide relief for chronic knee pain due to osteoarthritisthan any other modality. Pooled studies (a meta analysis) that compared physical interventions for chronic knee pain showed the following “effect sizes:”

(Electro-) Acupuncture: ES of 0.89
Warm baths: ES 0.65
Exercise: ES 0.55

To put this in perspective, the “Minimum Clinically Important Difference” or minimally significant Effect Size, according to patients suffering from chronic knee pain, is 0.39. And the Effect Size threshold the UK’s National Institute for Health and Care Excellence (NICE) uses to determine reimbursement is 0.50.

Call them acupuncture activists. And really brainy ones.

When an Australian group published (in October 1st JAMA) an impressively large (N=282, but there were 4 groups) clinical trial of 12 weeks of acupuncture for chronic (moderate to severe) knee pain in persons 50 years and older, they concluded no difference between the sham control and actual acupuncture. Editors of the premier acupuncture journal in the world (the BMJ group’s Acupuncture In Medicine), however, cried foul. And lamented the lost opportunity to help millions of chronic knee pain suffers. There were determined to set the record straight.

Their protest and clarification letter to JAMA was summarily rejected.

But guess what? They just happen to have their own journal. (Take that, AMA.) And not one of the enumerable, embarrassing American titles. As noted above, we’re talking The British Medical Journal (Even if the quality of the studies AIM publishes sometimes causes a cringe here and there, the White and his staff are doing the best with what they have to work with.) And so AIM editor and British Medical Acupuncture Society chief Mike Cummings proceeded to publish their quibbles today for inclusion, I imagine, in the January 2015 print edition, and set about to educate folks about how to consider clinically relevant results.

For expediency’s sake, I will extract the key arguments from their brilliantly prepared letter today. Then over the next week or so I will work on paraphrasing and whittling it down.

1. BACKGROUND OF THE CLINICAL PROBLEM. Patients with OA knee pain are suffering the commonest cause of pain and disability in older people. More than half have inadequate pain relief.2 They face a choice between ineffective paracetamol, non-steroidal drugs that can harm the heart, (kidneys) and gastrointestinal tract, gels that scarcely work, physiotherapy, opioids that cause dependency and lose effectiveness, arthroscopic washouts that do nothing or surgery.3 They deserve a fuller, more considered answer to their question: “Is it worth trying acupuncture?”

2. NIFTY DESIGN OF TRIAL. The neat part of the Zelen design that Hinman et al used was that the control group, who were not given acupuncture, were not even aware that their pain scores were used in a trial of acupuncture so disappointment could not influence their scores, as was claimed for other studies. This ‘no acupuncture’ group was compared with acupuncture (manual) and with sham laser (and with real laser, which is not considered here, to keep things simple).

3. WHERE THE ANALYSIS BEGINS TO STUMBLE. The problems started with the trialists’ choice of the threshold minimum clinically important difference (MCID) to estimate sample size. They chose a value based on one chosen by six self-styled ‘expert’ physicians,4 namely a 35% fall in baseline pain score (1.8/sample mean baseline 5.1). This is equivalent to an effect size (ES) of 0.6, calculated using their assumed baseline SD of 30 (the actual SD was 21, giving a higher threshold ES of 0.86). A different figure for MCID was generated by 192 patients with OA, who registered improvement scores as well as changes in pain.5 This showed a more modest MCID, equivalent to an ES of 0.39 (shown in figure 1). The National Institute for Health and Care Excellence (NICE) did not regard any value for MCID as valid6 and chose a generic value of 0.5 (see figure 1). Hinman et al chose a high threshold and also failed to discuss the effect that alternative threshold MCID values would have on the interpretation of their findings. We also note that the MCID for any treatment should be chosen to take account of acceptability, safety and cost-effectiveness,7 which would argue for a lower threshold for acupuncture for knee pain.

4. THE OLD “BETA ERROR” BUGABOO: SHORT ON STATISTICAL POWER. Hinman et alapplied this ‘clinically important’ difference to a ‘clinically irrelevant’ comparison–acupuncture versus sham laser. Sham laser is not an available therapy. The only reason for comparing acupuncture with sham would be to estimate the effects of the needles themselves, but this is already well known from the Cochrane review8 and an individual patient data meta-analysis (figure 1).9 It is known that the effect of needles alone is small, and so is unlikely to be identifiable reliably with sample sizes of less than about 800.10 The sample size in the study by Hinman et al (n=70) clearly appears to be inadequate for the question, according to the existing evidence, and not best use of resources. The resulting ES of acupuncture against sham that was actually found by Hinman is similar to that shown by the best evidence8 (see figure 1), although the wide CI means the data can only be of any importance when they are included in a meta-analysis in the future.

5. WHO DECIDES WHAT TREATMENT EFFECT IS MEANINGFUL? Hinman et al found that, after 12 weeks, knee pain was significantly reduced by acupuncture compared with no acupuncture control, with an ES of 0.6 (data from their table 2; see figure 1). The difference did not quite meet the MCID they had postulated–although the estimated ES is the same size as the MCID–but it more than meets the MCID chosen by patients themselves (ES 0.39) and that selected by NICE (ES 0.5). In interpreting this result, the secondary outcomes should also have been brought into thoughtful consideration: there were significant differences in favour of acupuncture for six out of eight secondary outcomes (see eTable 5 in their paper) and the response rate, which is the most patient-orientated measure of success,7 was 76% in the acupuncture group compared with 32% in the no acupuncture control group.

6. TO MAKE MATTERS WORSE, THE AUSTRALIANS STUDIED THE LEAST EFFECTIVE ACUPUNCTURE TECHNIQUE. Hinman et al did not apply optimal acupuncture. Use of electroacupuncture has been shown superior to manual stimulation for knee pain in 2010.8

A couple of weeks ago, I also came across a study of “needle-less” acupuncture, also for chronic knee pain of the OA variety. Basically it involved warming the knee with these stick on cones of burning mugwort–sort of like (very carefully) burning incense around your knee cap. That too showed clinical effectiveness, although now I am prompted to dig up the original study and see if we can fit that “effect size” into our acupuncture, warm baths, exercise line-up above. Stay tuned.


Get every new post delivered to your Inbox.

Join 473 other followers