Complementary and alternative healthcare use by participants in the PACE trial of treatments for CFS

[Dolphin]

Complementary and alternative healthcare use by participants in the PACE trial of treatments for chronic fatigue syndrome.

Lewith G, Stuart B, Chalder T, McDermott C, White PD.
Journal of Psychosomatic Research 2016;87:37-42.
10 June 2016.
http://www.jpsychores.com/article/S0022-3999(16)30327-0/abstract
http://www.sciencedirect.com/science/article/pii/S0022399916303270

Highlights

• CAM use is very common in patients with chronic fatigue syndrome (CFS) involving approximately two thirds of patients.
• Its use is not associated with any important clinical outcomes.
• The main predictors of CAM use are female sex and local ME group membership.
• These observations are important for clinicians and should be discussed with CFS patients.

Abstract
Background
Chronic Fatigue Syndrome (CFS) is characterised by persistent fatigue, disability and a range of other symptoms. The PACE trial was randomised to compare four non-pharmacological treatments for patients with CFS in secondary care clinics. The aims of this sub study were to describe the use of complementary and alternative medicine (CAM) in the trial sample and to test whether CAM use correlated with an improved outcome.

Method
CAM use was recorded at baseline and 52 weeks. Logistic and multiple regression models explored relationships between CAM use and both patient characteristics and trial outcomes.

Results
At baseline, 450/640 (70%) of participants used any sort of CAM; 199/640 (31%) participants were seeing a CAM practitioner and 410/640 (64%) were taking a CAM medication. At 52 weeks, those using any CAM fell to 379/589 (64%). Independent predictors of CAM use at baseline were female gender, local ME group membership, prior duration of CFS and treatment preference. At 52 weeks, the associated variables were being female, local ME group membership, and not being randomised to the preferred trial arm. There were no significant associations between any CAM use and fatigue at either baseline or 52 weeks. CAM use at baseline was associated with a mean (CI) difference of 4.10 (1.28, 6.91; p = 0.024) increased SF36 physical function score at 52 weeks, which did not reach the threshold for a clinically important difference.

Conclusion
CAM use is common in patients with CFS. It was not associated with any clinically important trial outcomes.

 

[Dolphin]

This is uninteresting if you're interested in learning about treatments as there is not a breakdown by therapy. They have 3 categories:
CAM medication use;
CAM practitioner use;
& Any CAM use.

2.2. Use and definition of complementary and alternative medicine

Participants were asked about their use of CAM at baseline and at 52weeks after randomisation using the Client Services Receipt Inventory (CSRI) [22]. CAM was defined as therapies delivered by CAM practitioners as well as ‘over the counter’ CAM medications or supplements purchased with or without therapist consultation. Regarding CAM, the CSRI only recorded acupuncture, osteopathy, homeopathy and herbal medicine and “other”; this was decided a priori based on the common treatments used in the UK. The “other” category was examined as part of the original trial analysis and any CAM therapies (such as reflexology, chiropractic and shiatsu) were grouped with acupuncture, osteopathy, homeopathy and herbal medicine to create a relevant single variable pertaining to the use of CAM practitioners [23]. We categorised each medication listed in the database as CAM or non-CAM. The CAM medications were multiple and included the general categories of herbal, homeopathic or nutritional medication. We could not analyse the CAM medication data by category because these were frequently used together and it would have been far too complex to extract and analyse variable dose and duration data. If a participant left the CSRI CAM usage sections blank we assumed no CAM use.

The most commonly used CAM practitioners were acupuncturists and homeopaths.

No percentage breakdown is given of the use of different CAM therapies or CAM practitioners.

 

[Dolphin]

They do highlight a big problem with the interpretation of the data.

Individualised practitioner delivering CAM may be associated with considerable non-specific treatment effects so the fact that improvement was not associated with practitioner consultations is both surprising and unexpected. Cho et al.'s systematic review [25] concludes that placebo effects are substantially diminished in patients with CFS. However, many people entering this study were already using CAM so any benefits that may be observed from either CAM medications or the non-specific effects of seeing a CAM practitioner are likely to have already been apparent prior to entry into the trial. The small number of people starting CAM during the study would not be expected to have significantly affected the overall study outcome.

There were a number of limitations in this study, the major one being that this was not a randomised controlled trial of CAM interventions, but was a secondary analysis of the associations of CAM use and CFS outcome. Many people entering this study were already talking CAM and therefore this confounds any definitive conclusions that can be drawn about the effects of CAM in CFS.

 

[Dolphin]

They claimed the following, but I'm not sure what evidence there is for it (maybe someone can point it out to me):

It was notable that there were either significant or nearly significant reductions in CAM use at follow up in those participants who received a supplementary therapy in contrast to those who did not. This suggests that CAM use may be a response to not receiving any alternative help, being an intervention that patients can apply without medical referral or assessment.

At one point they say:

Compared to the SMC group there was no statistically significant difference in CAM use by patients in each of the 3 trial arms at 52 weeks after controlling for baseline CAM use, stratification factors and clustering by centre.

 

[sarah darwins]

It was notable that there were either significant or nearly significant reductions in CAM use ...

How does "nearly significant" work, then?

 

[Valentijn]

How does "nearly significant" work, then?

I'll translate: "It wasn't significant at all, even when using a p-value of 0.05 or 0.1, and doing our best to manipulate the data to produce a significant result."

 

[Woolie]

God, what a zero paper. There's just nothing at all in it. No research question, method, no design, no result. How the hell do they get this guff published?

 

[MEMum]

Let's just hope there's no skewed press coverage of this zero paper.

 

[Daisymay]

In my innocence I had presumed that anyone taking part in any scientific trial would have to agree to not start any other treatments for the duration of the trial? Obviously wrong.

And why has it taken 5 years to come out with this info? Oh off course, they need to prolong the PACE trial papers for as long as possible as it's one of their excuses for not publishing the raw data.

Surely as a potential confounding factor this should have been addressed at the very beginning of the PACE saga not now?

How can they differentiate as to whether some, just some of the supposed improvement they claim to be due to CBT/GET wasn't down to some patients being helped to say sleep better, be in less pain, shade more eneregy with say acupuncture, osteopathy, magnesium etc so might these statistically insignificant CAM improvements have helped to skew the CBT/GET scores in a positive direction without them knowing? Or the placebo effect of CAM?

 

[Chrisb]

Tell me if you think I am over-interpreting this.

Do they really have any interest in the clinical outcome of CAM use? The message which they might wish to have come across is that the majority of patients are female proponents of CAM, and we all know what that means; nudge, nudge.

The audience to which this message might speak would not understand that there is no comparison with use of CAM in other established conditions. So what is the point of it?

 

[Cheshire]

Do they really have any interest in the clinical outcome of CAM use? The message which they might wish to have come across is that the majority of patients are female proponents of CAM, and we all know what that means; nudge, nudge.

Yes I wouldn't be surprised if it were their aim.

A rapid search brings numbers quite similar for the use of CAM in cancer patients:

According to the 2007 National Health Interview Survey (NHIS), which included a comprehensive survey on the use of complementary health approaches by Americans, 65 percent of respondents who had ever been diagnosed with cancer had used complementary approaches, as compared to 53 percent of other respondents.

https://nccih.nih.gov/health/cancer/camcancer.htm#use

Overall, 66.7% of the respondents reported using CAM, most often in an attempt to boost the immune system.

http://jco.ascopubs.org/content/18/13/2515.abstract

 

[snowathlete]

This is uninteresting if you're interested in learning about treatments as there is not a breakdown by therapy. They have 3 categories:
CAM medication use;
CAM practitioner use;
& Any CAM use.





No percentage breakdown is given of the use of different CAM therapies or CAM practitioners.

Which basically makes the whole thing a completely useless piece of research. You can't group together dozens, or hundreds, maybe even thousands of alternative treatments and measure them together as if they are one thing. Of course you will find what you find. I'm not a big fan of alternative treatments, mostly they are a waste of time, but I long for some decent quality research from the UK, not this kind of junk.

 

[snowathlete]

It was notable that there were either significant or nearly significant reductions in CAM use at follow up in those participants who received a supplementary therapy in contrast to those who did not. This suggests that CAM use may be a response to not receiving any alternative help, being an intervention that patients can apply without medical referral or assessment.

'nearly significant'....I've not seen that in a paper before. Surely, the point is that it is either significant or it is not. Nearly significant is the same as 'not significant' in my book.

I am not at all surprised that people who are in a treatment arm of a trial lower their use of alternative treatments. No doubt they do the same when trying a new alternative treatment, how else can you tell what is working? Common sense, surely.

 

[user9876]

'nearly significant'....I've not seen that in a paper before. Surely, the point is that it is either significant or it is not. Nearly significant is the same as 'not significant' in my book.

I am not at all surprised that people who are in a treatment arm of a trial lower their use of alternative treatments. No doubt they do the same when trying a new alternative treatment, how else can you tell what is working? Common sense, surely.

It will also depend on whether therapists and doctors discouraged their use.

It does make me wonder what counts as CAM for example does taking vitamin supplements (unless suggested by a doctor or dietician)?

 

[Daisymay]

It will also depend on whether therapists and doctors discouraged their use.

It does make me wonder what counts as CAM for example does taking vitamin supplements (unless suggested by a doctor or dietician)?

Yes, could taking one vitamin D capsule each day count as CAM?

 

[CFS_for_19_years]

How does "nearly significant" work, then?

Nearly significant is the same as "nearly insignificant."

 

[sarah darwins]

Nearly significant is the same as "nearly insignificant."

I was going to agree but ... isn't it the same as "insignificant".

 

[Esther12]

CAM was defined as therapies delivered by CAM practitioners as well as ‘over the counter’ CAM medications or supplements purchased with or without therapist consultation.

That's not much of a definition of CAM.

CAM is defined as being delivered by CAM practitioners, who are defined by delivering CAM.

I'd be interested in knowing how they justified splitting CAM from CBT/GET/APT.

 

[Dolphin]

It will also depend on whether therapists and doctors discouraged their use.

It does make me wonder what counts as CAM for example does taking vitamin supplements (unless suggested by a doctor or dietician)?

Not 100% clear. They say:

CAM was defined as therapies delivered by CAM practitioners as well as ‘over the counter’ CAM medications or supplements purchased with or without therapist consultation

and

The CAM medications were multiple and included the general categories of herbal, homeopathic or nutritional medication

I'm sceptical of a lot of alternative medicine. However treatments like l-carnitine are based on plausible theories and have RCT evidence. These look like they would be counted as CAM in the trial; probably vitamins also.

 

[Jonathan Edwards]

God, what a zero paper. There's just nothing at all in it. No research question, method, no design, no result. How the hell do they get this guff published?

Remember it is the Journal of Psychosomatic Research - the authors are convinced that they have undertaken real physical research but this is merely a false belief on their part.