The PACE trial principal investigators answer questions about graded exercise therapy and the trial
The PACE trial was the largest randomised controlled trial of non-drug treatments for chronic fatigue syndrome (CFS) ever undertaken. Detailed descriptions of the trial and its published papers and treatment manuals are available free to download at www.pacetrial.org. In essence, 641 patients with CFS were randomly allocated to one of four treatments: all received specialist medical care (SMC), provided by a doctor. One group received that alone, and the other three groups received SMC complemented by up to 15 sessions of a therapy: adaptive pacing therapy (APT) (based on the principle of living within the limits of the illness by balancing activity and rest), cognitive behaviour therapy (CBT) or graded exercise therapy (GET). Full descriptions of the therapies are available from the trial website. Patients were followed up for a year.
We found that both CBT and GET were more effective than both SMC alone and APT in improving fatigue and physical functioning, both reported by the patients themselves. About six out of ten patients made a clinically useful improvement in both fatigue and functioning after CBT or GET. CBT and GET were also more effective in improving: general ability to do things, global improvement in health, sleeping, post-exertional malaise. The latter finding is particularly reassuring for those who are concerned that GET might make this core symptom worse; it didn’t - it improved it. We disagree with those people who are concerned that these improvements are somehow less valid since they relied on the subjective reports of patients rather than objective measures; we believe that the patient is the best judge of how unwell they are. Of the three therapies, patients were most confident of being helped by APT and least confident of CBT before they started their treatment; yet APT was ineffective while CBT was effective. We also examined whether patients who met other criteria for the illness, including having myalgic encephalomyelitis (ME), had a different pattern of responses to these treatments, and found that they didn’t; those who also met the criteria for ME were as likely to respond.
We went on to use a combination of self-reported measures (being within the population normal range for both fatigue and physical functioning, plus no longer meeting trial entry criteria, plus a self-rating of feeling “much” or “very much” better) to measure the numbers of patients who had recovered from their current ill-health, and found that both CBT and GET were about three times more likely to lead to recovery than the other two treatments.
In order to carefully examine whether any of these treatments made patients worse, we measured six safety outcomes. Serious adverse events, serious adverse reactions to treatment, and non-serious adverse events were judged independently by experienced CFS doctors. We also noted both the numbers of patients who withdrew from treatment because of feeling worse, those dissatisfied with their treatment, as well as those who reported feeling “much” or “very much worse”. There were no significant differences in any of these safety measures across the four treatment arms, with the one exception that more patients who had received specialist medical care alone, with no added therapy, were more dissatisfied with their treatment. We think this all means that these treatments are as safe as each other. Because the proportions thought to have had a serious adverse reaction to a treatment were 2% or less in all treatment groups, we think this means that all these treatments, including GET, are safe. Improvements in post-exertional malaise, after GET in particular, add to this reassurance.
The one important caveat is that these treatments are safe so long as they are delivered by the appropriately qualified, trained, and supervised therapists, as described on the trial website and in published papers. We suspect that some patients who have deteriorated after GET, for example, have done so because their treatment was either not graded or not therapist supported. We would stress the importance of therapy being mutually agreed between therapist and patient, with no coercion involved, and it being reviewed if not helping.
What is graded exercise therapy?
In a nutshell, the therapist (usually a physiotherapist) plans a programme of graded exercise therapy (GET) with the patient, taking into account daily living tasks and pattern of sleep and activities. The first period of therapy is spent finding a baseline of physical activities that can be done even on a bad day. Once this is achieved, the patient and therapist carefully plans increased periods of (preferably enjoyable) physical activity, such as walking, at a pace that is no different from their current pace (i.e. an increase in time being physically active, with no increase in intensity). This is individually planned according to the patient’s current abilities and increases occur every week or two, depending on symptoms. If a patient develops increased symptoms, with no other cause found (such as having a cold), then activity is held at the same level, with no increase until symptoms abate, to allow adaptation to occur. Once a patient is doing about 30 minutes physical activity a day (possibly divided up into intervals, if easier to achieve), then the patient and therapist plan graded increases in the intensity of exercise. No incremental increase of either time or intensity should be more that a 20% increase from the current level (i.e. 1-2 minutes extra walking if already doing 10 minutes a day). More detailed treatment manuals are available at www.pacetrial.org.
Changes to the trial protocol
Some people have criticised the trial for changing some things after we started the trial, such as how we analysed the results. This happens not infrequently in trials, and we stated at the very beginning that we would write a detailed statistical analysis plan as we went on. This was done in order to improve the analysis and interpretation of the data, and was completed before any outcome data were examined and approved independently by the two trial oversight committees. This link gives more explanations for those changes that were made http://www.pacetrial.org/faq/
We suggest that this trial shows that both cognitive behaviour therapy and graded exercise therapy are moderately effective and safe treatments, and can be recommended to those patients able to attend clinics. Yes, we do need more research to find more effective treatments for this chronic disabling illness, but we now know that there are two currently available treatments that may help and are safe to try.
Professor PD White,
Professor M Sharpe,
Professor T Chadler