It seems obvious that the non-specific effect of any intervention is closely related to its credibility and plausibility as a therapeutic intervention i.e. its “face validity”. Now, some of the effects associated with meditation must be non-specific, i.e. comprising a mixture of placebo, therapeutic contact, spontaneous improvement, and so on, whereas some, hopefully, are specific to meditation alone. One might even propose that different meditation techniques have varying proportions of specific and non-specific effects. Within the context of an RCT, the control strategy should ideally:

• elicit all the non-specific effects that meditation might have, but have none of meditation’s specific effects;
• not have any specific effects of its own.

By fulfilling these criteria the control strategy makes the RCT methodology sensitive to any specific effects of meditation that might be detectable.

Dr Ramesh Manocha

none

It may not be practically possible to devise and implement an ideal control method for meditation trials, nevertheless it is important to select a strategy that approximates that ideal. The bare minimum criteria for a control process in meditation research should therefore be:

• First, high face validity as a therapeutic/stress management intervention in its own right. It should actually appear to be a credible meditation technique if that is the expectation of participants.
• Second, a process that involves relaxation and reduction of somatic arousal since this is the nearest conventionally understood phenomenon that meditation resembles and from which it needs to be distinguished.

Given these considerations there are two ideal strategies: sham meditation and the head-to-head comparison.

none

Examining randomised controlled trials exploring meditation in my review, control methods were presumptively categorized according to their face-validity into low, moderate or high face validity categories.

The low face-validity controls used strategies that were:
Passive and unstructured: Participants were involved in minimal or no activity relating to the trial and had no interaction with researchers as a result of being allocated to the control group (e.g. waiting list, no treatment, self-directed reading, or referral to community resources). This kind of comparator controls for minor non-specific effects, such as regression to the mean, the natural history of disease states and environmental factors common to all participants. It does not however, control for any non-specific effects that may be elicited by behaviour therapies.

The moderate face-validity controls use strategies:
Passive and structured. These involved some sort of regular and structured interaction with personnel associated with the experiment (e.g. regular lectures, specific reading, structured educational sessions on unrelated topics, regular blood pressure checks). This controls for the same confounders as Category 1 in addition to the effects of therapeutic contact and sense of active involvement.
• That were active in nature and generated some expectation of benefit but did not have effects or credibility as either a method of relaxation or meditation e.g. support groups, education about health factors measured in the study, or lectures on stress and lifestyle management. This controls for the same as Categories 2 and 3 in addition to the effects of social support, improved lifestyle, etc. Social support has been repeatedly demonstrated to be effective in improving mood and quality of life and reducing the severity of disease symptoms. “Standard treatment” was included in this category.

High face-validity controls use strategies that were:
• That were active in nature but not designed to generate significant expectation of therapeutic benefit (e.g. exercise). This controls for the same as Category 2 in addition to the effects of regular physical activity. Regular physical exercise has been shown to improve mood.
Active in nature, generated some expectation of benefit and elicited the simple physiological effects on rest but did not have specific credibility as a meditative method (e.g. progressive muscle relaxation, other relaxation methods, hypnosis, biofeedback, psychotherapy).
• The same as above but also had credibility as a meditative method (e.g. meditation techniques, strategies designed to convincingly mimic meditation) or constituted a legitimate form of psychotherapy (e.g. desensitisation, cognitive behaviour therapy, counselling).

Dr Ramesh Manocha

none

The design of RCTs for meditation (or any behaviour-based therapy for that matter) involves a number of unique challenges compared with pharmacological trials. While both categories of trial use an inactive placebo, the pharmaceutical trial uses an inert “sugar tablet” which appears similar to the medication being administered. The participant taking the “sugar tablet” is unable to ascertain whether or not they are taking the active or placebo treatment thus allowing the trial to control for confounding factors that may contribute to changes in the participants condition other than that caused by the treatment being studied.

More information about the use of placebos in meditation research can be found at Dr Manocha’s website.

Ramesh Manocha.

none

Sham meditation involves designing control strategies that overtly resemble the intervention, but which do not actually trigger the effects purported to be specifically associated with meditation.

Sham techniques are used in research when the researcher wishes to examine the specific effects a meditation technique may elicit, while controlling for the non-specific effects.

Sham meditation procedures necessarily involve deception of participants and the ethicality of this in clinical trials is open to dispute. Further, this kind of strategy can be logistically challenging and there is always a risk that the deception might be uncovered, thereby immediately invalidating the entire study.

Ramesh Manocha.

none

Dr Ramesh Manocha explains why strong controls are necessary when researching the real-world relevance of meditation in this excerpt from his thesis:

“Some might argue that controlling for non-specific effects is an academic exercise with little real-world relevance. This is supposedly because factors such as the placebo effect, expectancy of relief and demand characteristics are a critical component of most health interventions when administered to patients in the field. Thus trials directed at assessing the ecological (“real-world”) relevance and validity of an intervention have no need to control for factors which will be operating in conjunction with the intervention effect anyway. Authors of such reports add that studies with these kinds of controls reflect real-world scenarios since study participants offered either meditation or the “usual care” (i.e. usually nothing) reflect the reality of their environment. This line of argument assumes that the only value of proper controls is its ability to help answer theoretical questions about meditation, but that such controls fail to allow meditation to demonstrate its “practical relevance”.

“Yet commercial purveyors of meditation frequently claim that their often expensive proprietary techniques are uniquely effective in order to justify expensive fees. Moreover, many meditation techniques can be arduous and culturally challenging. It is important to determine whether it is justifiable to demand these significant costs and efforts or whether the same effects might be elicited by simpler strategies that are similarly rich in non-specific effects but possibly cheaper and easier to implement. By controlling for non-specific effects, it is possible not only to provide important theoretical information about whether meditation has any unique effects, but also whether it is an economically justifiable option, whether it offers any more of an advantage over accepted strategies and whether the claims of meditation enthusiasts have any basis.”

More information about Dr Manocha’s research can be found at his website.

none

archives