So this is a week into a Mindfulness-Based Stress Reduction (MBSR) course.  I've practised the "Body Scan" meditation eight times in the last eight days.  Initially there was the practice session that we did at the first MBSR class that I talked a bit about in last week's blog post "MBSR: first evening of the course".  The next day I used the 35 minute practice provided on the CD's our trainer has given us.  The day after this I was involved in teaching a day on mindfulness with an experienced colleague & mindfulness trainer and we went through the body scan seated over 30 to 40 minutes (and then to a partially silent/meditative meal).  Since then I've twice listened to the approximately 30 minute body scan (both times lying down) taught by Jon Kabat-Zinn on the CD included in the fine multi-authored book "The mindful way through depression", twice listened to the 15 minute practice (once seated & once lying) taught by Mark Williams in his recent publication "Mindfulness: a practical guide to finding peace in a frantic world", and once taken myself through a 30 minute lying down practice.

How has it been?  As someone who can get a bit drowsy lying down for thirty minutes, I found the seated practices somewhat better for me than the lying down ones (even though the instructions "on the tin" are to do this meditation lying rather than sitting).  I'm unconvinced by the need for the practice to be 30 minutes long.  I suspect it's quite often a fairly daunting challenge for people and may both result in less regular practice, poorer quality attentional focus, and increased dropout rates.  I think I'd probably prefer two 15 minute practices with more latitude about whether to practise them seated or lying.  These are difficult questions though.  There is a very good argument that goes something like "We know that the current way of teaching MBSR/MBCT courses is associated with encouraging outcomes.  Unless you're involved in a research study examining alternative delivery formats, or circumstances force you to change the format, it is much more sensible to stay with the approach that has actually been shown to work."  Good point.  And there are a cluster of research studies that throw interesting side-lights on all this. 

For example, in the recent paper "Do mindfulness meditation participants do their homework? And does it make a difference? A review of the empirical evidence", the authors report "Regular between-session practice of mindfulness meditation is among the key factors proposed to produce the therapeutic benefits of mindfulness-based programs. This article reviews the mindfulness intervention literature with a focus on the status of home practice research and the relationship of practice to mindfulness program outcomes. Of 98 studies reviewed, nearly one-quarter (N = 24) evaluated the associations between home practice and measures of clinical functioning, with just over half (N = 13) demonstrating at least partial support for the benefits of practice. These findings indicate a substantial disparity between what is espoused clinically and what is known empirically about the benefits of mindfulness practice."  So not a crashing endorsement here of a key importance for home practice.  To add further complexity, studies using tape/CD players that keep a surreptitous record of time playing, show that participants' self-estimates of how long they spend listening to relaxation/meditation recordings as homework only correlate poorly with the time actually noted on the players being used.  How about the overall length of the training course itself?  Another paper "How long does a mindfulness-based stress reduction program need to be? A review of class contact hours and effect sizes for psychological distress" commented "The mindfulness-based stress reduction (MBSR) program was designed to be long enough for participants to grasp the principles of self-regulation through mindfulness and develop skill and autonomy in mindfulness practice. It traditionally consists of 26 hours of session time including eight classes of 2-1/2 hours and an all-day class. The circumstances of some groups exclude them from participating in this standard form and a number of trials have evaluated programs with abbreviated class time. If lower program time demands can lead to similar outcomes in psychological functioning, it would support their utility in these settings and might lead to greater participation. However, the effect of variation in class hours on outcomes has not been systematically studied. To obtain preliminary information related to this question we examined effect sizes for psychological outcome variables in published studies of MBSR, some of which had adapted the standard number of class hours. The correlation between mean effect size and number of in-class hours was nonsignificant for both clinical and nonclinical samples and suggests that adaptations that include less class time may be worthwhile for populations for whom reduction of psychological distress is an important goal and for whom longer time commitment may be a barrier to their ability or willingness to participate. However, the standard MBSR format has accrued the most empirical support for its efficacy and session time may be important to the development of other kinds of program outcomes. The result points to the importance of empirical studies systematically examining this question."

Yup, it's the ubiquitous "Further research studies are needed" comment!  Actually I think it's very likely that regular home practice adds considerably to the benefits participants can gain from attending a MBSR/MBCT course, it's just that what we know is tentative and approximate.  The message on frequency and duration of practice is probably, all other things being equal, it makes sense to stick to the standard course recommendations for now. 

What about how the "Body scan" is approached?  How is it best to do the exercise itself?  How much is it helpful to encourage relaxation, even subtlely?  How much is it helpful to encourage effort?  Is it useful to underline the benefits that can be achieved?  Are we teaching a skill where one can hope to get better at it?  I've listened to four experts teaching the body scan in the last eight days.  They teach it differently.  Not a big surprise.  The basic structure is the same, then it's like listening to different jazz musicians interpreting a piece in their varied ways.  Are all equally valid, or are some interpretations "better" than others?  The crunch is ... what is likely to be most helpful?  And this sits on the bigger question, what are the mechanisms that make a MBSR/MBCT course useful?  What precisely is it most important for participants to learn? 

For more on this, see the post "MBSR: what's likely to be the most effective way of teaching the bodyscan?"