Is reflexology an effective intervention? A systematic review of randomised controlled trials

Reflexology is one of the most popular complementary therapies.1 It has been defined as

Other authors have explained that “‘reflex’ in the context of ‘reflexology’ means the ‘reflection’ of all the organs, systems and structures of the body onto the feet or the hands”.3

In 1997, I (with Kerstin Köder) published a review of the trial data pertaining to reflexology.4 We were able to include five randomised controlled trials (RCTs) and two non-randomised studies. Our conclusion from the totality of this evidence was that “it seems possible, even probable, that its perceived benefit is brought about by non-specific effects”.4 Since then, numerous new RCTs have been published.

The aim of this systematic review was to summarise and critically evaluate the data from RCTs of reflexology as a treatment for any human condition.

Non-randomised trials,5,6 studies of reflexology not delivered by trained reflexologists,7 studies of non-reflexology foot massage,8 and trials with healthy volunteers9-11 were excluded.

The key data (condition studied, study design and controls, primary outcome measures, follow-up, and main results) from all included trials were extracted according to the above criteria. The methodological quality of all included RCTs was assessed by two independent reviewers using the Jadad scale.12 Scores on the scale range from 0 to 5, with 5 indicating the highest quality. Because therapist blinding was not possible, studies were considered “double blind” if the patient and the evaluator of the results were blinded to the group allocation.

A meta-analytic approach had been envisaged; however, it had to be abandoned due to the heterogeneity of the primary data.

Of 217 publications identified, 18 RCTs met the inclusion criteria.13-30 Their key features are summarised in Box 1. Twelve of these studies failed to show convincingly that reflexology is an effective treatment.13-15,18,21-25,27,28 Five RCTs suggested positive effects,16,19,20,26,29 and the direction of the result of one was unclear.17

The methodological quality of the RCTs varied, but, in most cases, it was poor. There were nine RCTs with a Jadad score of 3 or higher.16,18,21-23,26-28,30 Of these higher-quality RCTs, two generated positive16,26 and seven negative results.18,21-23,27,28,30 Many of the RCTs did not adequately control for non-specific effects.13-15,17,20,25,29

Of the nine placebo-controlled studies, three suggested specific effects of reflexology — symptom control in premenstrual syndrome,16 improved quality of life during cancer palliation,19 and symptomatic treatment of multiple sclerosis.26 The other six placebo-controlled RCTs failed to demonstrate specific therapeutic effects of reflexology.18,21-24,28

Most of the included RCTs had extremely low sample sizes. Only five studies18,22,25-27 had more than 50 participants, and seven trials had 30 or fewer participants.13,15,19,20,23,29,30 The two large RCTs18,27 (130 and 243 participants, respectively) both generated negative results.

The range of conditions being treated with reflexology was remarkably wide. For most of them, a single RCT was available. Only for asthma (two studies),15,21 post-operative state (2),13,18 cancer (3),19,20,23 and multiple sclerosis (2)26,30 were independent replications available. For asthma and multiple sclerosis, the results were contradictory (one positive21,26 and one negative15,30 for each condition); for cancer, two studies suggested benefit19,20 and one suggested no benefit;23 and for postoperative state, both RCTs failed to demonstrate effectiveness.13,18

Interest in conducting RCTs and the methodological quality of the published RCTs both seem to have increased during recent years. Twelve RCTs have become available since 2000; seven of these newer studies were placebo controlled,19,21-24,26,28 and two had a Jadad score higher than 3.21,28

This systematic review failed to produce convincing evidence to suggest that reflexology has health benefits beyond a placebo response. Multiple RCTs were available for only four conditions, indicating a general lack of independent replication. The methodological quality of the primary studies was often poor. Most high-quality RCTs did not demonstrate effectiveness. The most promising evidence seems to be in the realm of cancer palliation.19,20

Reflexology has been tested for an impressive range of conditions, implying that reflexologists believe it to be effective in many different situations. The claims made of reflexology by some professional organisations seem to confirm this notion (Box 2). There is little doubt that a foot massage is pleasantly relaxing;31 however, specific medical claims should always be supported by sound evidence. In the case of reflexology, this unfortunately does not appear to be the case.

Reflexologists use “maps” on which one particular area (usually on the sole of the foot) is assumed to represent one particular organ or organ system. The maps are based on the assumption that 10 “energy zones” run longitudinally through the body. Each foot has five of these lines, and all body organs are believed to lie along one or more of these lines.32 By palpating “blockages” on the foot, reflexologists also engage in diagnostic procedures. A certain finding in a particular area and the patient’s reaction to manual pressure are thought to provide clues about a tissue injury or malfunction.33 Previous studies suggest that the diagnostic validity of this method is insufficient.34,35

Reflexologists postulate that malfunctioning of an organ or body system leads to deposits of uric acid or calcium crystalline.3,36 These, in turn, would impinge on the nerve endings on the feet and obstruct lymph flow. Massaging these areas would break down the crystalline deposits so that they can be reabsorbed and eliminated.37 Other hypotheses relate to improvement of blood flow; the theory that reflex points are nerve receptors, the stimulation of which reduces muscular or psychological tension, inducing “deep relaxation”,37 or emits “impulses to all parts of the body, as to Golgi tendon and muscle spindle cells”;32,35 the involvement of the lymphatic system; the notion that waste products, like lactic acid, are removed through reflexology massage; and the enhancement of general homoeostasis.32,38 These theories have, so far, not been submitted to sufficient experimental testing. Future investigations might address these issues, and employ rigorous control for a placebo effect, as incorporated in some of the recent trials.26,28

Most proponents of reflexology would argue that this method is free of risks. However, if used as a diagnostic tool, it will generate false-positive and false-negative diagnoses.34,35 Moreover, if employed as an alternative therapy to treat serious conditions, reflexology can be life-threatening.39 Thus, the notion of an entirely benign intervention does not withstand critical evaluation.

Why, then, do people pay for reflexology? As mentioned before, the treatment can be a pleasant and relaxing experience. There is also evidence to suggest that “a strong therapeutic relationship with providers who listen and provide time and knowledgeable advice”40 might contribute to reflexology’s popularity. If that is true, the current high level of use of “alternative” treatments would be a biting criticism of conventional health care, which often seems to fail patients’ needs in this respect.41 A further reason was recently pointed out by Raymond Tallis:

Three systematic reviews have previously assessed the value of reflexology. Our previous evaluation is now outdated;4 this article is an attempt to update it. Hughes and colleagues recently published a review of massage techniques in paediatric cancer care.43 Even though it included several RCTs of reflexology, its aim was not to summarise the totality of the evidence for or against reflexology. Wang and colleagues recently published a systematic review of the efficacy of reflexology and found that “there is no evidence for any specific effect of reflexology in any conditions”.44 Unfortunately, this systematic review included less than 50% of the available RCT data.

My systematic review has several limitations. Although efforts were made to find all relevant RCTs, I cannot be sure that this aim was achieved. Publication bias might have led to the disappearance of negative studies. In this case, the (already quite disappointing) overall picture generated here might be too positive. Finally, the paucity and the poor quality of the existing studies prevents definitive judgements about the value of reflexology.

In conclusion, this systematic review failed to demonstrate that reflexology is clinically effective for any of the wide range of conditions for which it has been tested.