I like this new definition but I'd love to see some data to back it up.
At the heart of all this is whether ME is a distinct clincial entity (perhaps entities) as I presume everyone here believes - or, as people like Simon Wessely argue, just one end of the spectrum of Chronic Fatigue. In fact Simon Wessely and Peter White have argued that adding the requirement for more symptoms (as this case definition does) make it more likely that it will include those with a primarily psychological illness.
There may be a way to help resolve this argument that could help prove this definition is better than other definitions and so should be widely adopted.
Basically, it involves
looking to see if there are 'natural boundaries' separating ME as defined here and other illnesses such as "CFS" defined more widely (eg by the Empirical Criteria, Oxford Criteria or even Fukuda), MDD or simple Chronic Fatigue. This is based on the work of a guy called
Kendell who has looked at the issue in general of how you define a disease (interestingly, he doesn't think psychologists are very good at it, even though he is one himself).
What you would do is take a large group of patients with ME defined according to these criteria, as well as patients with the other illnesses mentioned above plus some healthy controls. Then you'd collect a whole load of data on them: symptoms, standard blood tests, tests for possible biomarkers like Natural Killer cells and gene expression; pyschological profiles; fatigue and function scores; maybe even some exercise tests too.
Then you'd analyse the data and see if the ME patients clustered away from the CFS patients, Chronic Fatigue patients, MDD and of course the healthy controls. If they do cluster into a separate group, then you have good evidence that these new Criteria are indeed selecting a more meaningful group of patients.
The CDC tried something a bit like this a few years ago but it was way too small and relied on the Empirical criteria for selection, so it probably only had maybe 25 CFS patients (Fukuda definition) and probably even fewer still Canadian criteria patients. It's arguable they looked at the wrong biomarkers too.