The UK’s £5 million, 640-patient PACE trial was a non-blind study of cognitive behavioural therapy (CBT) and graded exercise therapy (GET) for chronic fatigue syndrome that used subjective self-reports of fatigue and physical function as its main outcome measures.
After the trial was over, the study authors abandoned all of the pre-planned main outcome and recovery analyses and calculated what they called “a normal range” for fatigue and physical function.
Bizarrely, patients could score worse at the end of the trial than they had at trial entry and be within this “normal range”. Patients with Class II congestive heart failure have an average physical function score close to the PACE trial’s “normal range” threshold.
The “normal range” clearly didn’t measure recovery but it has been the basis for claims that a substantial proportion of patients recovered following these therapies, made in two leading scientific journals: The Lancet and Psychological Medicine.
The claims were widely reported in the British media and never retracted by the journals or the study authors, even though critics — many of them patients — wrote in to point out the errors.
Investigative journalist and public health scientist David Tuller of the University of California, Berkeley recently confirmed the problems in a damning critique of the trial. He interviewed Dr Bruce Levin, a leading biostatistician at Columbia University and expert in trial design, who said:
“I have never seen a trial design where eligibility requirements for a disease alone would qualify some patients for having had a successful treatment… I find it nearly inconceivable that a trial’s data monitoring committee would have approved such a protocol problem if they were aware of it.”
Clearly, claims of recovery based on these bizarre standards cannot be allowed to stand.