The
Journal of Psychosomatic Research seems to be suffering from some sort of identity crisis. Earlier this year, the editor and his two immediate predecessors published
an admirable editorial in which they noted the serious risk of bias in subjective outcomes in studies that are not rigorously blinded. Yet the journal’s editorial advisory board is studded with investigators who specialize in unblinded studies that rely solely on subjective outcomes—in particular, Professors Peter White and Michael Sharpe, two of the lead PACE investigators, and Professor Per Fink, their Danish colleague.
And the journal keeps publishing studies based on such flawed research, in the process undermining its own credibility and raising questions about whether its editorial team is actually in charge of editorial policy. How else to explain
the recent paper from Professor White and colleagues of one-year follow-up results from GETSET, a study of self-guided graded exercise therapy (GET) for the illness they call chronic fatigue syndrome (CFS)? This follow-up managed to present null results at one year as a success by engaging in blatant outcome-switching.
How exactly did the authors accomplish this feat? They prioritized a “within-group” comparison for the GETSET arm as the most important result rather than the “between-group” comparison that is the outcome of interest in any clinical trial. In other words, they highlighted first and foremost the fact that the purported early benefits from GETSET were maintained at follow-up rather than the most salient detail–that in the end there were no demonstrable differences between the intervention and comparison arms..........
