Differential effects with GET in CFS: Systemic review and meta-analysis

[deleder2k]

Differential effects of behavioral interventions with a graded physical activity component in patients suffering from Chronic Fatigue (Syndrome): An updated systematic review and meta-analysis

• M.M. Marquesa, b, c, , ,
• V. De Guchta,
• M.J. Gouveiab,
• I. Lealb,
• S. Maesa

http://www.sciencedirect.com/science/article/pii/S0272735815000896

Highlights


•Interventions including physical activity have beneficial effects on chronic fatigue.

•The number of trials is modest and there is heterogeneity between them.

•Type of setting and provider of treatment moderate fatigue severity effect sizes.

•Minimal direct contact interventions are promising.

Abstract
An updated systematic review and meta-analysis was conducted to (1) evaluate the effects of behavioral and psychological interventions containing a graded physical activity component upon fatigue severity, physical functioning, physical activity and psychological distress, and to (2) examine potential moderator effects of trial characteristics (type of control, setting, provider, length of treatment, psychological component, flexibility in physical activity, and minimal face to face patient-provider contact). Pertinent content of selected studies was extracted and rated on a scale of methodological quality. Sixteen randomized controlled trials (N = 2004) were included in the meta-analyses. Significant small to medium effect sizes (Hedge’s g= 0.25 to g= 0.66) were found for all outcomes at post-treatment (M = 5.2 months) and follow-up (M = 11.7 months), with the exception of physical activity at post-treatment (g= 0.11). The largest effects were found for fatigue severity (g= 0.61 to g= 0.66). Subgroup analyses revealed that minimal contact interventions had additional beneficial effects upon fatigue (g= 0.96) and depression (g= 0.85). Interventions provided by psychologists-psychotherapists and interventions conducted in secondary-tertiary settings also resulted in more beneficial effects on fatigue. We found some indication of publication bias. The small number of studies and variability between them are limitations of this study. Future research should explore additional moderating effects in order to improve the effectiveness of interventions.

And the criterias:

Eligibility criteria

Types of participants

Studies were included if they were conducted in adult patients presenting (Idiopathic) Chronic Fatigue or Chronic Fatigue Syndrome/Myalgic Encephalomyelitis (CFS/ME).

Types of interventions

Studies had to include an arm of a behavioral and/or psychological intervention with a graded physical activity/exercise component, targeting chronic fatigue management.

Types of comparisons

Studies had to include a control condition, consisting of usual care, waiting list control, or another type of intervention (e.g. relaxation).

Types of outcomes

Studies had to present statistical data allowing the calculation of effect sizes (in the published study or provided by the author(s) upon request), on at least one of the following outcomes - fatigue severity, functional impairment/physical functioning, physical activity and/or physical capacity, and psychological distress (depression and/or anxiety), measured at baseline (pre-treatment), at post-treatment and/or at follow-up.

Types of studies

Studies were included if they were randomized controlled trials (RCTs) published in peer review journals in English.

There were no restrictions with respect to the type of diagnostic criteria used, setting, format and source of delivery of the intervention, as well as with respect to the length of the intervention and follow-up measurement point(s).

Conclusion

This meta-analysis of behavioral and psychological interventions targeting graded activity suggests that these interventions have sustained beneficial effects on chronic fatigue management, in particular on fatigue severity reduction for which a medium effect was found. The finding that minimal contact interventions have similar and in some cases higher effects on fatigue severity and depression compared to more intensive interventions is important as these interventions can be more easily implemented in standard health care, can be useful for patients presenting difficulties in regularly attending health care facilities (Burgess et al., 2012), and can be suitable for patients who do not need more intensive forms of treatment (Tummers et al., 2012). All trials included in this meta-analysis had an initial face to face contact with patients, which may have led to increased motivation of patients to engage in a behavior change process (Burgess et al., 2012). Most of these minimal interventions also included patient (self-help) manuals and allowed flexible physical activity/exercise levels that take into consideration the patients’ own resources, which can add to chronic fatigue management. Notwithstanding the beneficial effects of the behavioral and psychological interventions included in this meta-analysis and the valuable indications about targets and format of future interventions, more research is needed to identify optimal features of interventions for chronic fatigue management.

Not sure whether this belongs in the ME research forum or in the other forum (with respect to the non existent criteria used)

 

[Esther12]

Thanks.

I've not seen the full paper, but had just posted about this in a thread on the 4-steps RCT, which is related and part of the same PhD project: http://forums.phoenixrising.me/inde...4-steps-for-unexplained-cf.32694/#post-604628

 

[Bob]

Significant small to medium effect sizes (Hedge’sg= 0.25 to g= 0.66) were found for all outcomes at post-treatment (M = 5.2 months) and follow-up (M = 11.7 months), with the exception of physical activity at post-treatment (g= 0.11). The largest effects were found for fatigue severity (g= 0.61 to g= 0.66).

So an intervention that is intended to increase physical activity and reduce disability, in patients with idiopathic chronic fatigue and chronic fatigue syndrome, failed to significantly increase activity post-treatment. I think that says it all really. Instead, the modest post-treatment changes are seen only in subjective outcomes in open label (I assume) trials. It's no surprise to us, because the lack of significant improvement in physical activity is a repeating pattern for GET in CFS trials and clinical practice. (Note that there may have been a significant effect for physical activity at follow-up but, from the abstract, we don't know if that is a small or medium effect.)

 

[deleder2k]

So an intervention that is intended to increase activity and reduce disability, in patients with idiopathic chronic fatigue and chronic fatigue syndrome, failed to significantly increase activity post treatment. I think that says it all really. Instead the modest changes are seen in subjective outcomes in open label (I assume) trials. No surprise to us, because the lack of increase in physical activity is a rotating pattern for GET in CFS. (Note that there may have been a significant effect for physical activity at follow up but, from the abstract, we don't know if that is a small or medium effect.)

Hear hear

 

[Sean]

So an intervention that is intended to increase physical activity and reduce disability, in patients with idiopathic chronic fatigue and chronic fatigue syndrome, failed to significantly increase activity post-treatment.

Zombie science at its finest.

 

[Snow Leopard]

With respect to the effect sizes, such effects only make sense with respect to the underlying measurement itself (eg change in questions answered on the questionnaire, or physical activity).
So such effect sizes are comparable to other interventions using the same underlying measurement, but are not really comparable to other underlying measurements. Eg. this does not meant that an effect of 0.6 or whatever actually means there is a moderate improvement. The effect size for physical activity of 0.11 and on fatigue questionnaires of 0.6 could in fact be the same effect, simply that the questionnaire was more responsive to small changes.

edit - the methodology is weird, they've included CBT studies as "behavioral interventions with a graded physical activity component" aka GET.

Secondly, basically no discussion of harms/adverse effects.

 

[Dolphin]

This study was excluded:

Lloyd, A. et al. (1993). Immunological and psychological therapy for patients with chronic fatigue syndrome: a double-blind, placebocontrolled trial. American Journal of Medicine, 94,97. Statistical data unsuitable

This was also excluded from the Malouff et al. review and possibly others because Andrew Lloyd (the corresponding author) didn't give reviewers extra data they needed e.g. standard deviations.

As this study found no benefit for CBT, excluding it increases average improvement for CBT.

Here's the abstract:

Am J Med. 1993 Feb;94(2):197-203.
Immunologic and psychologic therapy for patients with chronic fatigue syndrome: a double-blind, placebo-controlled trial.
Lloyd AR1, Hickie I, Brockman A, Hickie C, Wilson A, Dwyer J, Wakefield D.
Author information

• 1Department of Immunology, Prince Henry Hospital, Sydney, Australia.

Abstract
PURPOSE:
To evaluate the potential benefit of immunologic therapy with dialyzable leukocyte extract and psychologic treatment in the form of cognitive-behavioral therapy (CBT) in patients with chronic fatigue syndrome (CFS).

PATIENTS AND METHODS:
Immunologic and psychologic treatments were administered to 90 adult patients who fulfilled diagnostic criteria for CFS in a double-blind, randomized, and placebo-controlled study. A four-cell trial design allowed the assessment of benefit from immunologic and psychologic treatment individually or in combination. Outcome was evaluated by measurement of global well-being (visual analogue scales), physical capacity (standardized diaries of daily activities), functional status (Karnofsky performance scale), and psychologic morbidity (Profile of Mood States questionnaire), and cell-mediated immunity was evaluated by peripheral blood T-cell subset analysis and delayed-type hypersensitivity skin testing.

RESULTS:
Neither dialyzable leukocyte extract nor CBT (alone or in combination) provided greater benefit than the nonspecific treatment regimens.

CONCLUSIONS:
In this study, patients with CFS did not demonstrate a specific response to immunologic and/or psychologic therapy. The improvement recorded in the group as a whole may reflect both nonspecific treatment effects and a propensity to remission in the natural history of this disorder.

 

[Dolphin]

Another fail for CBT that wasn't included because the author(s) wouldn't pass on the necessary statistical information:

Whitehead L. & Champion, P. (2002). Can general practitioners manage chronic fatigue syndrome? A controlled trial. Journal of Chronic Fatigue Syndrome, 10,55-64.
Statistical data unsuitable

Original Article

Can General Practitioners Manage Chronic Fatigue Syndrome?
A Controlled Trial


2002, Vol. 10, No. 1 , Pages 55-64
Lisa Whitehead, and Peter Campion
The Health and Community Care Research Unit, University of Liverpool,
The Department of Primary Care Medicine, University of Hull,
†Correspondence: Ms.Lisa Whitehead, The Health and Community Care Research Unit, Thompson Yates Building, Quadrangle, Brownlow Hill, Liverpool, L69 3GB



Abstract
Background: Chronic Fatigue Syndrome/Myalgic En-cephalomyeltis (CFS/ME) is now recognised as a condition that results in substantial disability with a prevalence of around 0.6%.

Aim: The study aimed to test the hypotheses that general practitioners could (a) diagnose and (b) treat patients with the Chronic Fatigue Syndrome (CFS).

Method: All practices in two health authorities were contacted with a 35% uptake. Fifty percent of practices then entered a patient into the study. Practices were randomised to either intervention or control groups, and were encouraged to recruit patients. It was intended that the intervention practices would introduce a form of brief cognitive behavioural therapy. Control practices were invited to manage their patients as usual, which often included referral to secondary care.

Results: The study suffered from both poor recruitment and high drop out. However, we were able to show that this intervention had no effect on the illness of the patients enrolled, and that patients with CFS remained highly disabled over the 12 month study period, whatever their treatment.

Conclusion: The study suggests that general practitioners in this study were unable to effectively treat the condition. This accords with the Royal Colleges' report (1996), that the only evidence for effective treatment thus far has come from specialist units. The study suggests that general practitioners are unable to provide a management programme of this nature, and possibly effective treatment programmes for CFS in primary care.




Read More: http://informahealthcare.com/doi/abs/10.1300/J092v10n01_05?journalCode=wcfs

 

[Dolphin]

The authors found some evidence that there might have been reporting bias i.e. unsuccessful trials or results that have not been published:

Meta-regressions with respect to methodological quality did not reveal any significant slope for any of the outcomes assessed (Table 4). Visual inspection of the funnel plots revealed asymmetry for fatigue severity. Egger's test showed a significant asymmetric funnel plot at post-treatment (intercept: 2.36, 95% CI 0.29–4.42,p= 0.03). After adjustment with the trim-and-fill procedure the magnitude of the effect sizes decreased from g = 0.61 to g = 0.40 at post-treatment (95% CI 0.21–0.59; number of trimmed studies = 5). At follow-up, inspection of the funnel plot for fatigue severity showed the presence of asymmetry, although Egger's test did not indicate a significant asymmetric plot (intercept: 2.37, 95% CI−1.75–6.49,p= 0.22). After adjustment the point estimate dropped from g = 0.66 to g = 0.46 (95% CI 0.14–0.77; number of trimmed studies = 2). Furthermore, inspection of the funnel plot of the effects for physical impairment/functioning at posttreatment revealed the presence of asymmetry. Egger's test did not show a significant asymmetric funnel plot (intercept: 0.25, 95% CI −2.47–2.97,p= 0.84). After adjustment with the trim-and-fill procedures, the magnitude of the effect dropped from g = 0.29 to g = 0.23 (95% CI 0.07–0.39, number of trimmed studies = 2). We also found presence of asymmetry for physical capacity at post-treatment. Egger's test indicated an asymmetric funnel plot (intercept: −8.79, 95% CI −17.31–0.28,p= 0.05). After adjustment the point estimate dropped from g = 0.27 to g = 0.18 (95% CI−0.06–0.42, number of trimmed studies = 1). There was no indication of asymmetry for other outcomes.

 

[Dolphin]

The following is included in Appendix C

2.4. Quality and risk of bias assessment The methodological quality of the included trials was assessed using a 29-item modified version of the Cochrane Collaboration Depression, Anxiety and Neurosis Review Group (CCDAN) quality rating scale (Lackner, Mesmer, Morley, Dowzer, & Hamilton, 2004; Moncrieff, Churchill, Drummond, & McGuire, 2001), validated byLackner et al. (2004). The scale assesses characteristics of both internal and external validity of trials. Each item is scored 0 (not done and/or not reported), 1 (done and/or reported to some extent) or 2 (adequately done and/ or adequately reported), with the exception of items 23 (Interests declared) and 29 (Consecutive subjects), which are scored 0 or 2. Total scores range from 0 to 58; higher scores indicate higher methodological quality. Risk of bias (high/low/uncertain) was classified based on the following items from this scale: Selection bias—concealment of allocation (item 6); detection bias—blinding of assessors (item 13); attrition bias (incomplete outcome data)—information on attrition (item 16) and inclusion of drop-outs in analyses (item 19), following the guidelines contained in the Cochrane collaboration's tool for assessing risk of bias (Higgins et al., 2011).

Discrepancies in quality rating were resolved by consensus between the two coders (MM and MJG). Overall, inter-rater agreement on the 29 items of the methodological quality scale was satisfactory (Cohen's kappa = 0.68).

The worst item by far is:
24: Allegiance to therapy:
All but one score 0 (that includes the PACE Trial). The odd one out is Prins et al. (2001) that scores 1 (out of 2).

Anyone know what this is? I'm guessing it's whether the paper reports any allegiances of the investigators to the therapies.

 

[Dolphin]

Another category where trials did very badly was:
13-Assessor blinded.
Score of 0: 13
Score of 1: 2 (Wearden et al., (2010); Deale et al. (1997))
Score of 2: 1 (O'Dowd et al. (2006))

 

[Dolphin]

Another category where trials did very badly was:
15-Treatment side effects

This presumably refers to the reporting of treatment side effects/harms.
Score of 0: 11
Score of 1: 4 (Fulcher & White (1997); Wearden et al. (2010), Sharpe et al. (1996), O'Dowd et al. (2006))
Score of 2: 1 (White et al. (2011))

These results are in line with:

Reporting of Harms Associated with Graded Exercise Therapy and Cognitive Behavioural Therapy in Myalgic Encephalomyelitis/Chronic Fatigue Syndrome

Tom Kindlon

ABSTRACT

Across different medical fields, authors have placed a greater emphasis on the reporting of efficacy measures than harms in randomised controlled trials (RCTs), particularly of nonpharmacologic interventions. To rectify this situation, the Consolidated Standards of Reporting Trials (CONSORT) group and other researchers have issued guidance to improve the reporting of harms. Graded Exercise Therapy (GET) and Cognitive Behavioural Therapy (CBT) based on increasing activity levels are often recommended for Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS). However, exercise-related physiological abnormalities have been documented in recent studies and high rates of adverse reactions to exercise have been recorded in a number of patient surveys. Fifty-one percent of survey respondents (range 28-82%, n=4338, 8 surveys) reported that GET worsened their health while 20% of respondents (range 7-38%, n=1808, 5 surveys) reported similar results for CBT.

Using the CONSORT guidelines as a starting point, this paper identifies problems with the reporting of harms in previous RCTs and suggests potential strategies for improvement in the future. Issues involving the heterogeneity of subjects and interventions, tracking of adverse events, trial participants’ compliance to therapies, and measurement of harms using patient-oriented and objective outcome measures are discussed. The recently published PACE (Pacing, graded activity, and cognitive behaviour therapy: a randomised evaluation) trial which explicitly aimed to assess “safety”, as well as effectiveness, is also analysed in detail. Healthcare professionals, researchers and patients need high quality data on harms to appropriately assess the risks versus benefits of CBT and GET.

Free at: http://iacfsme.org/ME-CFS-Primer-Ed...of-Harms-Associated-with-Graded-Exercise.aspx

 

[Dolphin]

It would be interesting to see whether Cochrane reviews have reported similar deficiencies in the CBT and GET literature.

 

[Dolphin]

The authors don't give the impression they think this is a biological condition that needs more than just management techniques to recover.

 

[Dolphin]

1.1. Physical activity and chronic fatigue Several studies emphasize the fact that lack of physical activity and prolonged physical inactivity (rest) can result in physical deconditioning as well as in other physiological and psychosocial consequences that may perpetuate fatigue and physical disability (Clark, Clark, & White, 2005; Fulcher & White, 2000; Nijs, Wallman, & Paul, 2011b). It has therefore been recommended that CFS/ME patients engage in physical activity/ exercise instead of refraining from it (National Institute for Health and Clinical Excellence, 2007). Physical activity thatis too vigorous can however perpetuate fatigue symptoms (Nijs, Paul, & Wallman, 2008; Nijs, Wallman, & Paul, 2011b). Patients' perceptions and expectations with respect to symptom exacerbation as a consequence of physical exertion can lead to fear of physical activity (Clark et al., 2005; Nijs et al., 2008; Prins, Van der Meer, & Bleijenberg, 2006) and explain the reduced levels of physical activity found in patients with chronic fatigue (Nijs et al., 2011a). Not surprisingly, it is common to find a “boom-and-bust pattern” (or all-or-nothing behavior) in these patients, which is the systematic alternation between periods of over-activity (when feeling good) and, as a consequence of that, feeling extremely fatigued and having to rest for longer periods of time. For these reasons, physical activity should be balanced, gradually introduced and offered with caution to CFS/ME patients (Clark et al., 2005; National Institute for Health and Clinical Excellence, 2007; Nijs, Wallman, & Paul, 2011b).

If patients fear activity and this causes reduced levels of physical activity, why would you necessarily find a "boom-and-bust pattern"? If people fear things, they tend to avoid them them not do lots of the activity sometimes.

 

[Dolphin]

It is good that the authors recognise there are different types of exercise program:

Graded Exercise Therapy (GET) has been recommended as a treatment for CFS/ME patients (National Institute for Health and Clinical Excellence, 2007). GET is based on the assumption that aerobic exercise (e.g. brisk walking) or physical activity (e.g. housework, gardening) must be initiated at a level (intensity and frequency) that doesn't exacerbate symptoms and must be gradually increased until patients reach an optimal level of activity. GET follows the exercise prescription guidelines of the American College of Sports Medicine (2013), tailored to each patient's initial level of physical capacity. Most GET interventions follow a similar protocol (Fulcher & White, 1998). GET is usually delivered by an exercise physiologist or physical therapist, and consists of supervised aerobic exercise sessions and/or home-based aerobic exercise prescription (e.g. walking). GET focuses on avoiding overexertion by advising patients not to exceed the recommended levels of physical activity/exercise. At the same time, in most GET interventions patients are encouraged not to reduce or stop doing physical activity/exercise when symptoms get worse. Recent approaches to GET have advocated that flexibility in graded exercise programs according to individual tolerance levels can be beneficial for CFS/ME patients. This implies that exercise can be reduced or even stopped when symptoms get worse (Nijs et al., 2008; Wallman, Morton, Goodman, Grove, & Guilfoyle, 2004).

Patients were recommended not to exceed the levels of exercise agreed upon beforehand by the therapist and patient in order to avoid overexertion, and to maintain these levels if symptoms got worse. The exception was the graded exercise program conducted by Wallman et al. (2004) in which patients were advised to reduce their activity levels if symptoms got worse (i.e. flexibility in physical activity levels-pacing).

 

[Snow Leopard]

If patients fear activity and this causes reduced levels of physical activity, why would you necessarily find a "boom-and-bust pattern"? If people fear things, they tend to avoid them them not do lots of the activity sometimes.

Notice how they didn't actually provide a citation showing evidence of a boom-and-bust pattern? Because this isn't an evidence-based statement (it isn't a thing for most patients - there isn't a boom, just a bust after trying normal/modest activities).