Title, Date of Publication & Journal

Title: ‘Randomised clinical trial of five ear acupuncture points for the treatment of overweight people’

Date of publication: Monday 16 December 2013.

Journal: Acupuncture in Medicine.

Claim supported by evidence?

The paper provides some evidence that ear acupuncture may produce small improvements in weight loss in overweight/obese subjects, compared to a control group, although the results are not conclusive. Larger improvements were seen in five-point ear acupuncture compared to single-point ear acupuncture. Although these effects were small, they were seen consistently across six different measures of obesity. However there were some major flaws in the study (see below) and more research is needed before acupuncture can be recommended as an effective treatment for obesity.

Summary

• Acupuncture group showed a small improvement in measures of weight loss compared to control
• Subjects were randomised, which increases confidence in results
• However there were major flaws in the study design: not well blinded and lack of control of dietary intake

Study Conclusions

For each of the six measures of obesity, the group receiving five-point acupuncture showed the largest improvement, followed by single-point acupuncture, with the control group showing the smallest improvement. The consistency of this pattern across all of the obesity measures is impressive, and supports the claim that five-point acupuncture “may be better at reducing abdominal fat … than single point stimulation”. However the overall magnitude of the improvements was relatively small (e.g. the largest average weekly improvement in BMI was less than 1%), so it is arguable whether the improvements observed are clinically significant. Also, the differences between the groups (particularly between the two different types of acupuncture) was modest and in several cases not statistically significant. Care should be taken to express the results in absolute terms (e.g. 6.1% improvement in BMI in the 5-point acupuncture group after 8 weeks, versus 3.1% improvement in the control group) rather than relative terms (e.g. doubling of weight loss compared to control).

The lack of statistical significance between the two types of acupuncture may reflect the relatively small number of subjects in the study (the press release admits that the study was small), and a larger study is needed to provide more robust conclusions. Because there were flaws in the study design, e.g. subjects were likely able to determine which treatment they had received, there may be alternative explanations for the observed results, and further research is needed before ear acupuncture can be recommended as an effective treatment for obesity.

Strengths/Limitations

There were some positive elements to the study design, i.e. this was a prospective, randomised study which included a control group. However, there were also some major flaws with the study design. Although the study was ostensibly ‘single blinded’, curious subjects could determine their treatment group by examining the adhesive dressing (as admitted by the authors). This may have led to differences in behaviour (e.g. diet and exercise) between the groups.

The results for this study are highly dependent on subjects’ compliance with the specified diet and exercise regime. However the authors admitted that compliance with this regime was poor, and this may have serious confounded the study results. An alternative explanation for the observed results might be that patients receiving acupuncture received a lower calorific intake and/or took more exercise compared to the control group, rather than the observed weight loss being a direct result of the five-point or single-point acupuncture interventions. A summary of compliance in each of the treatment groups was not provided in the main results, so this possibility cannot be ruled out. Also, compliance was self-reported by subjects, and may not be reliable.

The statistical analysis was flawed, because only including subjects who completed the study can produce false conclusions. This is a problem when a large proportion of the subjects withdraw from the study, especially where this occurs at different rates between the treatment groups, as was observed in this study. There are modern statistical methods available to address this problem, but these were not applied in this study.