White et al. (Reference White, Goldsmith, Johnson, Chalder and Sharpe2013) report various recovery rates from chronic fatigue syndrome (CFS) following the PACE Trial. However, additional information would have been useful.

White et al. use a selection of broad criteria to define recovery, none of which allow one to be confident recovery has been achieved. Firstly, Chalder Fatigue Questionnaire (CFQ) and SF-36 Physical functioning (PF) scores within the normal range are in fact possible at baseline. This means it is possible to have fatigue that is classed as ‘severe, disabling and affected physical and mental function’ and yet satisfy this particular recovery criterion.

Secondly, not satisfying the Oxford criteria only requires a change on just one measure, and the change may be minimal, across a threshold, e.g. going from an SF-36 PF score of 65 to 70 or a CFQ (bimodal) score of 6 to 5. A sign that this criterion is not that stringent can be seen with the fact that 41% of the specialist medical care (SMC) group, which received no active treatment, no longer met the Oxford criteria at 12 months, much higher than recovery rates seen in previous studies (Cairns & Hotopf, Reference Cairns and Hotopf2005).

Finally, a CGI score of 2, which means a participant rated as ‘much better’ but not ‘very much better’ also gives no assurance that somebody had recovered. It seems quite possible that many with CGI scores of 2 have simply improved but not recovered.