Open-Label Placebo Treatment for Cancer-Related Fatigue: A Randomized-Controlled Clinical Trial, 2018, Hoenemeyer et al

[Andy]

Abstract

The purpose of this 21-day assessor blinded, randomized-controlled trial was to compare an open-label placebo (OLP) to treatment as usual (TAU) for cancer survivors with fatigue. This was followed by an exploratory 21-day study in which TAU participants received OLPs while OLP participants in the main study were followed after discontinuing placebos. Cancer survivors (N = 74) who completed cancer treatment 6 months to 10 years prior to enrollment reporting at least moderate fatigue (i.e., ≥4 on a 0–10 scale) were randomized to OLP or TAU. Those randomized to OLP took 2 placebo pills twice a day for 21 days.

Compared to those randomized to TAU, OLP participants reported a 29% improvement in fatigue severity (average difference in the mean change scores (MD) 12.47, 95% CI 3.32, 21.61; P = 0.008), medium effect (d = 0.63), and a 39% improvement in fatigue-disrupted quality of life (MD = 11.76, 95% CI 4.65, 18.86; P = 0.002), a large effect (d = 0.76). TAU participants who elected to try OLP for 21-days after the main study reported reductions in fatigue of a similar magnitude for fatigue severity and fatigue-disrupted quality of life (23% and 35%, respectively). OLP may reduce fatigue symptom severity and fatigue-related quality of life disruption in cancer survivors.

Open access at https://www.nature.com/articles/s41598-018-20993-y

 

[Trish]

Fascinating. So even when patients are told they are taking a placebo, they report less fatigue after 21 days, and this is maintained over the next 21 days after stopping the placebo.

It would be interesting to use a more objective measure such as actometer, and to do a long term follow up. I don't think the questionnaires they used would have picked up whether there were able to function better.

So the question in my head is, are they really less fatigued after taking the placebo, or is there some way it is fooling them into thinking they are less fatigued and filling the questionnaires in differently?

It reminds me of the asthma study where patients on placebo thought they were less breathless, but objective spirometer readings showed the asthma was really just as bad.

 

[Andy]

Fascinating. So even when patients are told they are taking a placebo, they report less fatigue after 21 days, and this is maintained over the next 21 days after stopping the placebo.

It would be interesting to use a more objective measure such as actometer, and to do a long term follow up. I don't think the questionnaires they used would have picked up whether there were able to function better.

So the question in my head is, are they really less fatigued after taking the placebo, or is there some way it is fooling them into thinking they are less fatigued and filling the questionnaires in differently?

It reminds me of the asthma study where patients on placebo thought they were less breathless, but objective spirometer readings showed the asthma was really just as bad.

Due to this, doesn't it mean that the PACE results are even more useless than we thought anyway? That even when people are given something that doesn't work, they still report that they are less fatigued? All this shows surely that questionnaires are likely to encourage people to report they are better than they were previously.

 

[Jonathan Edwards]

Due to this, doesn't it mean that the PACE results are even more useless than we thought anyway? That even when people are given something that doesn't work, they still report that they are less fatigued? All this shows surely that questionnaires are likely to encourage people to report they are better than they were previously.

Absolutely bang on right Andy. It shows that people report feeling better simply because that is their role in a charade.

Yet again we see a study conclusion that indicates that the authors do not a have a scrap of common sense.

So the question in my head is, are they really less fatigued after taking the placebo, or is there some way it is fooling them into thinking they are less fatigued and filling the questionnaires in differently?

Neither, it is fooling them into playing a charade.

It is really quite an elegant study in that respect.
Like the SMILE trial it is another sort of positive control for PACE showing that if you tell people that their role is to say they are better they probably will.

 

[JohnM]

You just need to skim the "Study Procedures" section to demonstrate the points made above, and the "PI delivered these discussion points in a way that fostered emotional support, hope and trust." and "we comfortably acknowledge disbelief is understandable and encouraged them to 'see what happens.'"

As to figures of improvement reported in the Abstract, together with a brief read of rest of paper, I'm pretty sure I don't need to be a statistician to see the potential for 'massaging' them for maximum impact .. so I didn't go to the trouble. All the stats 101 lessons I find here on the forum in abundance .. must be rubbing off on me.

Wishing everyone improved health and every happiness. John

 

[Indigophoton]

You just need to skim the "Study Procedures" section to demonstrate the points made above, and the "PI delivered these discussion points in a way that fostered emotional support, hope and trust." and "we comfortably acknowledge disbelief is understandable and encouraged them to 'see what happens'"

I think this was entirely deliberate. The PI is an "oncology health behaviour specialist". So I think they purposefully did everything they could to maximize the conceptual impact on participants to maximize the apparent resulting effects.

One of the authors of this paper was interviewed a couple of years ago:

Dr. Ted J. Kaptchuk, a professor of medicine at Harvard Medical School and director of the Harvard-wide Program in Placebo Studies and the Therapeutic Encounter (PiPS) at Beth Israel Deaconess Medical Center in Boston, has been studying placebos for more than 20 years. ... Kaptchuk says placebos won’t work for every medical situation—for example, they can’t lower cholesterol or cure cancer. But they can work for conditions that are defined by “self-observation” symptoms like pain, nausea, or fatigue.

(My bold)

 

[Esther12]

One of the authors of this paper was interviewed a couple of years ago:
(My bold)

Yeah, Kaptchuk is a big promoter of the 'powerful placebo'.

This placebo study does look interesting. I've not had a chance to look at the paper yet but it sounds like it could be useful.

I thought I'd post some old notes about a biopsychosocial cancer fatigue study here.

This one from Knoop and Bleijenberg:

Examining the role of physical activity in reducing postcancer fatigue

Marieke F. M. Gielissen,

1 Jan F. Wiborg,2 Constans A. H. H. V. M. Verhagen,3 Hans Knoop,2 and Gijs Bleijenberg2
Author information ► Article notes ► Copyright and License information ►
Go to:
Abstract

Purpose

Persistent fatigue is a long-term adverse effect experienced in about a quarter of patients cured of cancer. It was shown that cognitive behavior therapy (CBT) especially designed for postcancer fatigue is highly effective in reducing severe fatigue. However, it is unclear by what mechanism the fatigue reduction is reached. In many fatigue reduction programs, an increase in physical activity is assumed to reduce fatigue. The purpose of the present study is to determine whether the effect of CBT on fatigue is mediated by an increase in physical activity.
Methods

Data of a previous randomized controlled trial on the efficacy of CBT for postcancer fatigue were reanalyzed (CBT, n

=

41; waiting list condition, n

=

42). Actigraphy was used to assess the level of objective physical activity. Cancer survivors were assessed prior as well as after the CBT and waiting list period. The mediation hypothesis was tested according to guidelines of Baron and Kenny. A non-parametric bootstrap approach was used to test statistical significance of the mediation effect.
Results

Although CBT effectively reduced postcancer fatigue, no change in level of objective physical activity was found. The mean mediation effect was 1.4% of the total treatment effect. This effect was not significant.
Conclusion

The effect of CBT for postcancer fatigue is not mediated by a persistent increase in objective physical activity.
Keywords: Fatigue, Postcancer fatigue, Activity, Exercise, Cancer survivors, Actigraphy

Results are the same as for CFS, with CBT leading to improvements in questionnaire scores, but not objective measures of illness. They're more open about this being a problem than they were with their CFS results, but still try very hard to avoid mentioning even the possibility that their questionnaires are not reliable measures of outcome for cognitive or behavioural interventions (and succeed!).

Full paper: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3360858/

I picked some bits out, but it's quite a short paper, so people might prefer to just read it themselves.

Couple of bits making it clear that, if CBT were working as intended, activity levels would increase:

A previous RCT of our research group demonstrated that cognitive behavior therapy (CBT) especially designed for postcancer fatigue is highly effective in reducing fatigue severity in cancer survivors [5, 6]. In clinical practice, physical exercise is seen as a promising strategy for treatment of fatigue during and after cancer treatment. Because an important element in CBT for postcancer fatigue is a physical activity program, the current study aimed to test a treatment model in which the effect of CBT on fatigue is mediated by an increase in physical activity.

In CBT for postcancer fatigue, an increase in the level of physical activity is stimulated. The fatigued cancer survivor selects a simple physical activity that can be performed every day (e.g., walking or cycling) and increases gradually and systematically the duration of this particular activity. The directive for a patient is to perform an activity twice a day, starting at a level of which is certain that symptoms do not aggravate. Mostly this is 5 to 10 min. The activity level is increased with 1 min a day for each time the activity is performed and will end at a maximum of 60 min twice a day (Table 5). Illness-related cognitions are also targeted during the physical activity program. In particular, patients' preoccupation with their condition and their belief that they have little control over their symptoms are challenged. By the time patients are active 2 h a day, most patients will have become aware of the fact that they are capable of doing more without experiencing extreme fatigue and that they recover faster than before. Gradually, walking or cycling is now being replaced by other activities. One might choose to systematically expand mental and social activities, preparing the patient for a return to work or other personal targets.

Nice they can be certain of this:

If a physician was certain that the fatigue had no somatic cause, the patient was invited to participate.

Looks like the control was waiting list only, so lots of room for response bias.

Again they decide to assume that perpetuating factors are entirely reversible and cognitive/behavioural, based on.... wish thinking? (pragmatism):

In understanding postcancer fatigue, it is important to make a distinction between precipitating factors and perpetuating factors. The assumption is that cancer itself and/or cancer treatment may have triggered fatigue (precipitating factors), but that other factors are responsible for persistence of fatigue complaints (perpetuating factors). CBT is focused on six perpetuating factors of postcancer fatigue. They involve insufficient coping with the experience of cancer, a heightened fear of disease recurrence, dysfunctional fatigue-related cognitions, dysregulatory sleep–wake cycles, dysregulatory activity patterns, and insufficient social support and interactions [13].

Worth noting that 'top-up' sessions were available in the second six months:

A mean of 12.5 sessions were given (SD 4.7) with a duration of 1 h during a 6-month period. Patients were offered a maximum of two sessions during a 6-month follow-up period.

For those of you really interested in mediation analyses, there's a fair bit of it, and I'm too tired to dig into that right now (someone explained bootstrapping to me... I understood it for a bit!).

No significant relationship between changes in physical function, and changes in fatigue questionnaire scores:

Regression analyses testing the hypothesized mediating effect are seen in Table 3. Fatigue was reduced significantly more in the CBT condition than in the control group (path c). However, CBT did not produce a significant change in physical activity (path a). There was also no significant relationship between changes in physical activity and changes in fatigue (path b). The effect of treatment remained significant when it was controlled for changes in physical activity (path c′).

I'm going to post all of their discussion, because it's a bit funny to me that they explore so many possibilities, without ever considering whether their highly effective cognitive-behavioural treatment might just be inducing response bias when compared to a waiting list control. We have no evidence that fatigue questionnaires are reliable ways of assessing cognitive and behavioural interventions.

Also, they keep trying to come up with ways in which patients could have increased their activity... but that just didn't show up.

Discussion

The results indicated that the positive effect of CBT on postcancer fatigue was not mediated by an increase in objective physical activity measured with the actometer. CBT did not cause an increase in physical activity at the end of treatment (path a) nor was an increase in physical activity associated with a reduction in fatigue (path b). The finding that there was no effect of the interventions on physical activity already showed that mediation was absent, but as mediation analyses require a large power, a bootstrap analysis was performed. The formal test of the mediation effect confirmed that CBT yielded its effect independent of a change in physical activity. These results are in line with Wiborg et al. [24], who demonstrated that the effect of CBT on fatigue in patients with chronic fatigue syndrome was also not mediated by an increase in physical activity following treatment.

There are several potential alternative explanations for the fact that we did not find support for our mediation hypothesis. One reason could be that the actometer was not obtained from all participants, and it is possible that we introduced a bias through exclusion. Also, by excluding patients, the sample size became smaller. However, no difference on baseline characteristics was found between completers and non-completers, and the effect of CBT was still significant in the smaller group. Additionally, the bootstrap approach is sensitive to detect differences in small samples.

Gradual increase of physical activity is an important part of the treatment protocol; however, no increase was found assessed with actigraphy at the end of treatment. Our study is limited by not documenting the level of physical activity during therapy. Therefore, we do not know whether the patients became temporarily physically more active during CBT. One might assume that this is probable, as they discussed repeatedly their gradual increase in activity with the therapist as demonstrated in Table 5. Anyway, a possible temporary increase was no longer present when the second assessment took place [13].[Patients did also have access to two additional sessions in the second six months, so it seems more likely that patients are just easily encouraged to tell therapists what they want to hear, or else substitute behaviour.]

CBT might have resulted in a temporary increase in physical activity. This temporary increase in physical activity during therapy might have been sufficient to facilitate a persistent change in illness-related cognitions and reduce fatigue. The gradual increase of physical activity can help the patient to feel (more) confident that he or she is actually capable of achieving these activities despite his or her symptoms, thereby enhancing his or her sense of control and bringing about a positive self-efficacy. This would be an interesting hypothesis to explore in future research.

In addition, CBT for postcancer fatigue is also focused on five other perpetuating factors of severe fatigue. It seems logical to assume that the positive effect of CBT on postcancer fatigue was mediated by change in the other perpetuating factors. Future research should therefore be focused on examining the exact mechanisms of change in CBT by monitoring illness-related cognitions and physical activity repeatedly during therapy and on determining the exact role of each perpetuating factor.

The association between fatigue and depression is complex. Fatigue can occur as a symptom of depression or may precipitate feelings of depression because of its interference with mood, work, and leisure activities. However, it is important to realize that in most patients, postcancer fatigue occurs independently from depression [25–27].

The current study sample does not reflect the incidence and types of cancer in the Dutch population, and the sample is quite young compared to the general oncological population. Therefore, a positive effect of physical activity on fatigue cannot be ruled out, and replication is necessary in survivors matching the general oncological population.

In the literature, there is a lack of consistent evidence whether a physical activity program reduces postcancer fatigue [7–12]. Because in the current study we did not find an increase in physical activity, it remains unclear if there is a positive effect of an increase in physical activity for reducing postcancer fatigue.

However, even without an increase in physical activity, CBT was highly effective in reducing postcancer fatigue. Therefore, a persistent increase in physical activity seems not to be a requirement in reducing postcancer fatigue, and a change in illness-related cognitions is likely to play a more crucial role in CBT for postcancer fatigue.

There's also this 2012 presentation from Chalder and Kathering Rimes that claimed psychological factors were more important than illness or treatment related factors for cancer related fatigue. That claim seems a bit premature to me.

http://www.babcpconference.com/archive/leeds2012/programme/abstract book v2.pdf

A Multivariate Study to explore the involvement of Cognitive and Behavioural factors in Cancer-related Fatigue
Sahil Suleman, Guy's & St Thomas' NHS Foundation Trust; Trudie Chalder, Institute of Psychiatry, King's College London; Kate Rimes, Department of Psychology, University of Bath

Background & Aims: Fatigue is a highly prevalent and debilitating problem in women with breast cancer undergoing chemotherapy. It is becoming evident that psychological factors are more important than illness-related and treatment-related factors when seeking to understand the development and maintenance of cancer-related fatigue. This study investigated the cognitive behavioural, social and affective responses associated with cancer-related fatigue and related functional impairment. Method: 100 women diagnosed with breast cancer completed a range of measures. Correlational and multiple regression analyses were used to explore associations between fatigue severity, social adjustment, physical functioning and a range of psychological, social demographic and clinical variables. 33 patients completed measures prior to chemotherapy and were followed up prospectively to examine the relationship between psychosocial variables and fatigue and functional impairment as measured after three cycles of chemotherapy.

Results: A range of cognitive, behavioural and affective variables were associated with increased fatigue severity, poorer social adjustment and increased physical impairment. Key cognitive behavioural correlates included increased embarrassment avoidance (cancer-related) cognitions, all-or-nothing behaviour, avoidance behaviour, health anxiety, unhelpful beliefs about negative emotions and perceptions of punishing responses from their significant other. Furthermore, exploratory analyses suggested that increased presence of maladaptive behaviours and unhelpful cognitions prior to chemotherapy predicted the presence of fatigue and functional impairment after three cycles of chemotherapy.

Conclusions: In line with our hypotheses, endorsement of the majority of cognitive, behavioural and affective variables was associated with increased fatigue and functional impairment. Some of these variables also continued to predict fatigue and functional impairment over time. These findings make a significant contribution to a cognitive behavioural understanding of cancer-related fatigue and also point to particular factors that are amenable to change within the context of cognitive behavioural therapy approaches.

 

[Mithriel]

For a while, lots of my friends with MS were taking part in clinical trials. As they discussed them, it became clear that "a little bit better" is where the placebo effect occurs. None of them wanted to dismiss a possible treatment out of hand so if it did not actually make them throw up or fall over or something they answered this way.