This has a been good week for advocacy. It’s showing that smart advocacy works. Twice advocates went straight back into “the system” looking for results and twice – once with the PACE trial and here with AHRQ report in the U.S. – they got them.
Spearheaded by Mary Dimmock who for years has lead the fight to secure a better definition, advocates asked the AHRQ to reanalyze it’s finding regarding CBT and GET.
As part of the Pathways to Prevention (P2P) project, the AHRQ had been tasked determined with determining which treatments were effective for chronic fatigue syndrome (ME/CFS). The group applied tough, tough standards – so tough that it rejected 90% of ME/CFS studies, and only 30 treatment studies made it to the analysis stage. Of those only four studies were deemed “high quality”.
Since most of the trials that made it to the final analysis were small, unreplicated studies, the AHRQ deemed the evidence insufficient to say anything about any treatments other than CBT/GET.
That lead to some strong outcries from some advocates, including some of the same advocates requesting another secondary analysis. The AHRQ was too tough, some said, for such an underfunded field. It made ME/CFS researchers and the chronic fatigue syndrome (ME/CFS) field look bad. That same rigor the AHRQ applied to its first analysis, however, paid dividends this time.
CBT and GET didn’t fare so well in the initial report. The AHRQ stated it had only “moderate” confidence that CBT was able to reduce fatigue or produce “global improvement”, and had low confidence it was able to improve overall functioning, enhance quality of life, increase working hours, or reduce work impairment in ME/CFS.
Mary Dimmock noted the conclusions had a flaw. The P2P report stated that the Oxford definition was so broad – only requiring that unexplained fatigue be present – that any study using it was likely to include people who did not have ME/CFS.
The Oxford definition is an anomalous definition used mostly by a small group of CBT/GET researchers based mostly in the U.K. and the Netherlands. The vast majority of researchers have always used the Fukuda definition for research but in the CBT/GET field the Oxford definition has dominated. The P2P report recommended that the Oxford definition be trashed.
In their CBT and GET analyses, though, the AHRQ treated the Oxford definition like it was valid, lumping studies using it and the Fukuda definition studies together. When that logical inconsistency was pointed out they agreed to reanalyze the CBT and GET studies, comparing studies using the Oxford definition to those using Fukuda definition.
The expectation was that studies using the Oxford definition would fare better than those using the stricter (but still not particularly strict) Fukuda definition, and, indeed, the AHRQ’s conclusions regarding the effectiveness of CBT/GET in ME/CFS changed dramatically once the analysis was done.
For the first time the AHRQ panel asked these questions”
Does CBT Increase Functioning In Fukuda Patients?
In the original report the AHRQ found “low evidence” that CBT does not improve functioning at all. When studies using the Oxford definition were removed the four studies left found that two studies produced benefit and two did not The AHRQ downgraded their evaluation of CBT stating that given the inconsistent results, the imprecise nature of the results and the mixed quality of the studies made it impossible for them to determine if CBT was effective or not in a Fukuda patient.
Does CBT Reduce Fatigue in Fukuda Patients?
When studies using the Oxford case definition were removed, the AHRQ was left with four fair-quality studies, three of which found benefit (n=327), one which found no benefit (n=65), and one poor-quality study finding no benefit (n=58). In general the trend was upwards but the quality of the studies was only fair: the AHRQ determined that this body of work provided a “low strength of evidence that CBT improves fatigue”.
Does CBT Improve the Employment Status of Fukuda Patients?
A starker difference was found in the effects of CBT/GET on employment. Once the AHRQ removed the Oxford definition studies they found that CBT had no impact at all on employment.
Does CBT Improve Quality of Life in Fukuda Patients?
Because all three studies measuring quality of life already used the Fukuda criteria, the AHRQ’s assessment remained the same. Two of the three found no effects on quality of life, leading the AHRQ to conclude that the strength of evidence that CBT/GET did not affect quality of life was low.
Does CBT Result in Global Improvement in Fukuda Patients?
Only two studies (but containing 540 people) assessed global improvement. Both the Oxford and CDC definition studies found improvement in global improvement but the authors deemed the evidence insufficient to make a recommendation.
Does Graded Exercise Therapy Improve Outcomes in Fukuda Patients?
Four trials, three using the Oxford Definition and one using the CDC definition were included in the original analysis. The results from all the trials were consistent; there was moderate strength of evidence that GET improved functioning (4 trials, n=607) and global improvement (3 trials, n=539), low strength of evidence that it reduced fatigue (4 trials, n=607) or decreased work impairment (1 trial, n=480).
When they removed the Oxford trials from the analysis, the one study left showed improvement, but it alone provided insufficient evidence that GET was successful in Fukuda patients.
The AHRQ re-analysis of the CBT and GET trials resulted in it downgrading its recommendations significantly. In two cases (CBT’s effects on functioning and employment) the reanalysis suggested that CBT was less effective in Fukuda patients than in Oxford criteria patients.
In other cases (CBT – Global Improvement; GET – functioning, global improvement, fatigue) the Fukuda studies suggested that some improvements did occur but too few studies were done for the AHRQ to make a recommendation.
In the end the only mildly positive statement the AHRQ could make was that there was low strength of evidence that CBT improved fatigue in ME/CFS. Otherwise it found no evidentiary basis for the use of CBT/GET in ME/CFS. It didn’t state that these techniques are helpful or not helpful; it simply stated that they are unproven.
Meanwhile we should note that the Fukuda definition, which has been the gold standard definition in ME/CFS research for over 20 years is nothing to shout about either. It’s more than past time for a new, statistically based research definition for ME/CFS. What better way to start the NIH’s new era of ME/CFS research than with a definition that works?
The AHRQ has spoken. Now that they have basically demolished the idea that anything positive can be said with certainty regarding the effectiveness of CBT/GET in ME/CFS where do we go from here?
The AHRQ, we should note, applies the most rigorous standards of any review body.
One might assume that the more rigorous reports might be authoritative in the field. The AHRQ is highly respected but it’s competing with the Cochrane Reports which are less rigorous and probably more used.
The Cochrane Reports probably include many studies that AHRQ rejected in their reports and are cited frequently.
A look at major medical websites indicates that the idea that CBT/GET is not just an effective treatment but is the only effective treatment for ME/CFS is deeply embedded in some.
These websites tend to provide rudimentary, imprecise analyses of treatments. They don’t get into the niceties regarding different definitions. Time will tell, but it may be difficult to get them to change their ways. Let’s look at what four prominent medical websites say about CBT/GET and ME/CFS
Up to Date may be the most critical website since it’s commonly used by physicians seeking to get up to date information on treatments. A pay to play website, UpToDate probably provides the most comprehensive overviews of treatments on the web. It’s CBT/GET section starts off with some proviso’s:
“a systematic review of 35 randomized trials evaluating therapies for SEID/CFS concluded that counseling therapies and graded exercise therapy may have benefits for some patients with SEID/CFS . However, neither of these modalities is curative.”
But then goes on to state that
“Cognitive behavioral therapy (CBT) has been effective in patients with SEID/CFS”.
The first CBT and GET trial UpToDate cites is the PACE trial. It notes that the trial involved 600 plus patients – thus providing automatic validation for many practitioners – but provides no hint that any issues exist with this trial. It then provides the kind of imprecise summary so common in medical websites. UpToDate states that
“CBT in combination with specialist medical care (in the PACE trial) was associated with less fatigue and better physical function compared with specialist medical care alone”
without noting that the effects weren’t considered clinically significant. UpToDate does the same thing with the GET arm of the PACE trial stating that
“GET in combination with specialist medical care was associated with less fatigue and better physical function compared with specialist medical care alone”
without noting that the step test results indicated that the participants were still severely disabled.
In another study UpToDate stated that CBT in combination with managed care was more effective than managed care without saying how more effective it was
Each of the vague statements UpToDate makes could be and probably are often interpreted to suggest that CBT/GET is really helpful in ME/CFS.
UpToDate then cherry-picked studies to cite one indicating no cardiopulmonary problems exist in ME/CFS and that exercise does not increase symptoms. It could have easily cited studies showing the opposite but chose to cite studies that reflected its conclusion – that exercise therapy is effective in ME/CFS.
The American Family Physician states
“There is substantial evidence for two treatments for CFS: cognitive behavior therapy (CBT) and graded exercise therapy.”
Unless you’re doing the kind of analysis that the AHRQ does, that is probably a true statement. It goes on to say that:
“Persons diagnosed with chronic fatigue syndrome should be treated with cognitive behavior therapy, graded exercise therapy, or both. Cognitive behavior therapy and graded exercise therapy have been shown to improve fatigue, work and social adjustment, anxiety, and postexertional malaise.”
AFP obviously goes far, far beyond the AHRQ’s conclusions, and even goes far beyond the Cochrane Report’s conclusion. It hardly attempts to produce a balanced report of CBT or GET’s limitations or their very moderate effects.
WebMD is a bit more enlightened. CBT is not presented as a cure but as a coping mechanism that can help some people.
“(CBT) is a type of counseling that has been shown to help some people who have CFS feel less tired” that “teaches you how to change the way you think and do things. These changes can help you better cope with fatigue and other symptoms.”
With regard to GET, WebMD states that it may help some people but suggests that those who are severely ill stay away.
“Unless your CFS is severe, try a graded exercise program, starting out with gentle and easy movement. Start with as little as 1 minute, and slowly add more over days or weeks. Studies have shown that a carefully planned graded exercise program can help some people with CFS regain their strength and energy and feel better. “
The Centers For Disease Control
The CDC has muted its recommendations for CBT and GET considerably in recent years. The CBT section is now posted in the Improving Health and Quality of Life section (not the Treatment section). No mention of a cure is made. The CDC contextualizes CBT as a way to effectively pace and reduce stress.
For CFS patients, CBT can be useful by helping them pace themselves and avoid the push-crash cycle in which a person does too much, crashes, rests, starts to feel a little better, and then does too much once again. Often, CBT is prescribed along with other therapies to help CFS patients manage activity levels, stress, and symptoms. CBT can help CFS patients better adapt to the impact of CFS and improve their ability to function and their quality of life.
The GET section is now posted in the management (not the treatment) section. The stated goal is not to cure ME/CFS but to improve fitness and contains many qualifiers: GET can help “some” patients “manage” their illness, and exercise should stop before patients get tired and ME/CFS is reactivated. It has a separate section for the severely ill.
Gradual, guided physical activity can help some CFS patients manage the illness. Appropriate rest is an important element of GET, and patients should learn to stop activity before illness and fatigue are worsened.
The end point of each GET session….. should be reached before the patient becomes tired….Appropriate goals are to prevent tiredness, to avoid activating the syndrome, and to increase overall fitness.
The CDC has been thought of as a leader that other websites will follow but this short analysis suggests this is not so. The CDC has the mildest recommendations regarding CBT and GET but its recommendations should indicate there is now almost no evidentiary base to conclude that CBT/GET are effective in ME/CFS.
Hopefully the AHRQ’s reanalysis will help blunt the emphasis given to CBT/GET in this field. What this field really, really, really needs, though, is to end the dominance of European government funded behavior based trials. No two entities should be able, simply by pouring money into a field, to dictate its treatment options, but that’s essentially what’s happened in ME/CFS.
The major medical websites will shift their emphasis when good treatment trials provide evidence that other treatments are effective in ME/CFS. The NIH can begin this process by funding Ampligen and Rituximab trials.
One successful Ampligen trial – which could be set up quickly and easily – would be enough to get FDA approval. Rituxmab would take longer but the NIH has a great deal of experience running Rituximab trials.
With the AHRQ re-analysis in hand, approvals for Ampligen and Rituximab, if they should occur, could go a long way to blunting the effects that years of over emphasis on CBT/GET have had. The NIH has said it’s serious about ME/CFS; funding these two trials would indicate that it is. It’s time for the U.S. to step up.