Just out of interest - how might you blind controls in a trial such as this with a 'therapy' such as this?
I guess you could 'fake it' or get the kids to practice relaxation therapy or something... then the outcome would be compared to 'listening to relaxing music'.
Wonder how LP would stack up against CBT - if that were used as a 'control'.
I don't think it makes sense to talk about blinding participants for 'treatments' like this, and blinded assessment is really hard for CFS too. More objective outcome measures that should be less prone to bias would be things like actometers and time spent at work/education.
'Relaxation' seems like a pretty poor control for bias to me. Something like homeopathy would probably be better so long as included a sense that the patients were responsible for taking medication correctly and in a way that would lead to their recovery (which seems ethically no worse than the LP approach... bleurgh!)