Treatment outcome in adults with CFS: a prospectiv​e study in England based on the CFS/ME NOD

[Dolphin]

Free full text:
http://qjmed.oxfordjournals.org/content/early/2013/03/28/qjmed.hct061.full
pdf- http://qjmed.oxfordjournals.org/content/early/2013/03/28/qjmed.hct061.full.pdf html

(The comparison trial is the PACE Trial)

Treatment outcome in adults with chronic fatigue syndrome: a prospective study in England based on the CFS/ME National Outcomes Database.

QJM. 2013 Mar 28. [Epub ahead of print]

Crawley E, Collin SM, White PD, Rimes K, Sterne JA, May MT; CFS/ME National Outcomes Database.

From the Centre for Child & Adolescent Health, School of Social & Community Medicine, University of Bristol, Oakfield House, Oakfield Grove, Bristol BS8 2BN, Wolfson Institute of Preventive Medicine, Barts and The London School of Medicine and Dentistry, Charterhouse Square, London EC1M 6BQ, Department of Psychology, University of Bath, Claverton Down, Bath BA2 7AY and School of Social and Community Medicine, University of Bristol, Canynge Hall, 39 Whatley Road, Bristol BS8 2PS, UK.

Abstract*

Background:

Chronic fatigue syndrome (CFS) is relatively common and disabling.

Over 8000 patients attend adult services each year, yet little is known about the outcome of patients attending NHS services.

AIM:

Investigate the outcome of patients with CFS and what factors predict outcome.

DESIGN:

Longitudinal patient cohort.

METHODS:

We used data from six CFS/ME (myalgic encephalomyelitis) specialist services to measure changes in fatigue (Chalder Fatigue Scale), physical function (SF-36), anxiety and depression (Hospital Anxiety and Depression Scale) and pain (visual analogue pain rating scale) between clinical assessment and 8-20 months of follow-up.

We used multivariable linear regression to investigate baseline factors associated with outcomes at follow-up.

Results:

Baseline data obtained at clinical assessment were available for 1643 patients, of whom 834 (51%) had complete follow-up data.

There were improvements in fatigue [mean difference from assessment to outcome: -6.8; 95% confidence interval (CI) -7.4 to -6.2; P < 0.001]; physical function (4.4; 95% CI 3.0-5.8; P < 0.001), anxiety (-0.6; 95% CI -0.9 to -0.3; P < 0.001), depression (-1.6; 95% CI -1.9 to -1.4; P < 0.001) and pain (-5.3; 95% CI -7.0 to -3.6; P < 0.001).

Worse fatigue, physical function and pain at clinical assessment predicted a worse outcome for fatigue at follow-up.

Older age, increased pain and physical function at assessment were associated with poorer physical function at follow-up.

CONCLUSION:

Patients who attend NHS specialist CFS/ME services can expect similar improvements in fatigue, anxiety and depression to participants receiving cognitive behavioural therapy and graded exercise therapy in a recent trial, but are likely to experience less improvement in physical function.

Outcomes were predicted by fatigue, disability and pain at assessment.

PMID:23538643[PubMed - as supplied by publisher]

*I gave each sentence its own paragraph

 

[Dolphin]

This journal takes letters if anyone wants to reply:

Letters to the Editor should be focused and succinct with fewer than 250 words.

 

[Simon]

Comparison is vs baseline (no control group), on a par with PACE Trial SMC group
Note that these figures are a before/after comparison, with no control group to allow for any natural improvement or placebo effect. This study uses the same Fatigue and Physical Function measures as the PACE trial, and the patients in this study did about as well as the SMC group ("specialised" medical care, no CBT or GET), and less well than the PACE CBT/GET patients who had the same treatments.

This study vs PACE SMC, baseline to 12 months
Fatigue: -6.8 vs -4.5 (ie improved more* than PACE SMC)
Physical Function: +4.4 vs +11.6 (ie improved less* than PACE SMC)

*Physical Function difference between this study and PACE is almost certainly statistically significant due to the very large sample sizes, though the difference in fatigue might not be.

This suggests that 'out in the wild', CBT and GET (and any othr treatment offered) are, overall, no better than SMC.

 

[maryb]

Thanks to those with a scientific bent for explaining things for us 'foggies'

 

[Dolphin]

Minor point:

Worse fatigue, physical function and pain at clinical assessment predicted a worse outcome for fatigue at follow-up.

Older age, increased pain and physical function at assessment were associated with poorer physical function at follow-up.

A possible corollary of this is that services that use the criterion that fatigue must be the most significant symptom, like Barts, and indeed trials like the PACE Trial, might (at least marginally) help get better results for themselves as it would have the effect of the pain levels of their patients being a bit lower on average.

 

[Dolphin]

This presumably helps isolate the services involved:
(my bolding)

Acknowledgements

We thank the clinical leads and team members of all services participating in the National Outcomes Database and, in particular, the clinical leads of the services that contributed data to this study: Dr Hazel O’Dowd (North Bristol NHS Trust), Karen Butt and Dawn Dunn (Somerset Partnership NHS Foundation Trust), Sue Pemberton (Leeds and York Partnership NHS Foundation Trust), Dr Maurice Murphy (East London Foundation NHS Trust and Barts and the London NHS Foundation Trust) and Dr Yasmin Mullick and Dr Amolak Bansal (Epsom and St Helier University Hospitals NHS Trust). E.C. had the original idea for the study; S.C. performed all data management and statistical analyses; M.M. and J.S. provided statistical and methodological input; S.C. and E.C. wrote the first draft and incorporated comments from all authors; all authors contributed to the interpretation and writing of the paper and read and approved the final manuscript.

Does this mean Dr. Murphy would cover two services?

3 services used CBT/GET and three activity management.

Can we work out which ones are which?

Barts would be CBT/GET.
I think Dr Hazel O’Dowd (North Bristol NHS Trust) would be also.

I think Sue Pemberton (Leeds and York Partnership NHS Foundation Trust) uses activity management.
I imagine Dr Yasmin Mullick and Dr Amolak Bansal (Epsom and St Helier University Hospitals NHS Trust) do too.

Barts is a large service. The paper says something about one large service offering CBT/GET and one offering activity management. I imagine the CBT/GET one is thus Barts? And the large activity management one would be which one?

 

[Dolphin]

Figure 1.
is

Predicted values of fatigue against time from generalized estimating equation regression model incorporating a
fractional term for time (time_) (N= 882). Vertical lines indicate follow-up interval adopted for this study (8–20 months).

Here is some info on this:

Follow-up interval Each team sent out follow-up questionnaires at 12 months. Variation in when the questionnaires were sent and delays in return of 12-month follow-up questionnaires led to variation in the exact time of follow-up. Also, some teams obtained data at additional follow-up points (e.g. 6 and 24 months). To maximize the availability of follow-up data for our analysis, we determined a margin of follow-up either side of 12 months. We did this by fitting fractional polynomial generalized estimating equation (GEE) regression models of fatigue against time.21,22 <more technical details in the paper>

Afraid I don't really understand polynomial generalized estimating equation (GEE) regression models of fatigue but believe I have an idea what they did.

Judging by the shapes of the graphs, the lowest point (best value) for fatigue was at around six months and scores then actually gradually got worse (graph is cut off at 600 days). I wonder does this mean that rather than patients improving with time, patients actually were regressing a little on average?
(This isn't discussed explicitly in the paper, only that points between 8 months and 20 months were considered comparable to scores at 12 months).

 

[Dolphin]

In Table 3:

For SF-36, coefficient <1 indicates that higher levels of disability are positively associated with higher levels of fatigue at 8–20 months.

and
Table 5:

For SF-36, coefficient <1 indicates that higher levels of disability are positively associated with higher levels of anxiety, depression or pain at 8–20 months.

I think, although am not 100% sure, it should be <0.

 

[Esther12]

I just stumbled over this old thread on outcome measures from these clinics, and thought I would post:
Report from CCRNC for APPG inquiry into patient services (2009)


http://forums.phoenixrising.me/inde...-into-patient-services-2009.8177/#post-157348

Dolphin picked this out as the most interesting and complete data at the time:

Service 16: 12 Month Outcome Data

Initial Follow up
Average hours of employment per week: 5.5 8.9
Average SF36 Function Scale scores (Range 100 - 0): 39.8 44.6
Average Pain VAS Scores: 38.2 34.0
Mean Chalder Physical Fatigue Scale score (Range 0 - 7): 6.4 3.4
Mean Chalder Mental Fatigue Scale score (Range 0 - 4): 3.5 2.1
Mean Chalder Fatigue combined score (Range 0 - 11): 6.9 5.5
Average Hospital Anxiety Scale score: 10.4 8.9
Average Hospital Depression Scale score: 10.1 7.8

There's also plenty there to indicate that 'activity management' can be quite different from the CBT/GET approach.

 

[Dolphin]

Thanks Esther12.

Mean Chalder Fatigue combined score (Range 0 - 11): 6.9 5.5

I just checked and this error is in the original. First number should be 9.9.

 

[Dolphin]

I just stumbled over this old thread on outcome measures from these clinics, and thought I would post:
Report from CCRNC for APPG inquiry into patient services (2009)


http://forums.phoenixrising.me/inde...-into-patient-services-2009.8177/#post-157348
<snip>

There's also plenty there to indicate that 'activity management' can be quite different from the CBT/GET approach.

Good point. This paper makes out it is based on CBT/GET, because "activity management" in the NICE guidelines are. However that's not necessarily what services are using if they're not using CBT/GET.

 

[seanpaul]

We did not find that depression, anxiety or duration of illness at assessment predicted outcome. Clinicians providing assessments should not assume that co-morbid mood disorders or length of illness are predictors of outcome.

White puts his name to this paper - no doubt just to get his name on another paper, with this being Crawley's work *which goes to show the absolute lack of independence in UK ME/CFS research, whereby they state there is no connection between depression and anxiety and outcomes, yet White has pushed many papers saying the absolute opposite, the ME is a psychiatric illness highly correlated and predicted by depression and anxiety.

Inconsistency again and again - using the same old psychiatric derived survey instruments

 

[Dolphin]

We did not find that depression, anxiety or duration of illness at assessment predicted outcome. Clinicians providing assessments should not assume that co-morbid mood disorders or length of illness are predictors of outcome.

White puts his name to this paper - no doubt just to get his name on another paper, with this being Crawley's work *which goes to show the absolute lack of independence in UK ME/CFS research, whereby they state there is no connection between depression and anxiety and outcomes, yet White has pushed many papers saying the absolute opposite, the ME is a psychiatric illness highly correlated and predicted by depression and anxiety.

Inconsistency again and again - using the same old psychiatric derived survey instruments

Good points.

I think a reason Peter White's name may be on it is he/Barts were part of those given research money by Action for ME to assess the services (it was a joint application to AfME headed by Esther Crawley but also mentioned Barts - I didn't think it was the safest or best use of such research funds).

Here is one fairly recent paper where they make an issue out of depression and anxiety: http://forums.phoenixrising.me/index.php?threads/cognitions-behaviours-and-co-morbid-psychiatric-diagnoses-in-patients-with-chronic-fatigue-syndrome.22402/

 

[Simon]

This presumably helps isolate the services involved:
(my bolding)

Does this mean Dr. Murphy would cover two services?
3 services used CBT/GET and three activity management.
Can we work out which ones are which?

Barts would be CBT/GET.
I think Dr Hazel O’Dowd (North Bristol NHS Trust) would be also.

I think Sue Pemberton (Leeds and York Partnership NHS Foundation Trust) uses activity management.
I imagine Dr Yasmin Mullick and Dr Amolak Bansal (Epsom and St Helier University Hospitals NHS Trust) do too.

Barts is a large service. The paper says something about one large service offering CBT/GET and one offering activity management. I imagine the CBT/GET one is thus Barts? And the large activity management one would be which one?

based on my googling, plus your comments:

CBT/GET
- Barts
- East London
- Bristol

Would have thought that Bart's was the biggest, but an AfME interview with Hazel O'Dowd(not sure of date of it) says now seeing 400 pa: the three CBT/GET services in this study saw 372, 160 & 157 patients in 2010

"More funding would make a huge difference. Since the service was established in 2003, our referral rate has grown every year and now we see about 400 patients annually."

"Activity Management"
- Leeds/York
Sue Pemberton, consultant OT, was replaced last year and has set up a private service - the Leeds/York service now run by consultant liaison psychiatrist Dr Hiroko Agaki so think it now does CBT/GET. From her statement it seems likely that Leeds was the big Activity Management centre since it's long established and has played a lead role:

I was asked to design the original therapy programme for the Leeds & West Yorkshire CFS/ME Service in 1990. This programme was then adopted by many other CFS/ME Services across the UK. In 2003 I was selected as the Regional Clinical Champion for the condition for the North, East and West Yorkshire area.

That would leave:
- Somerset
- Epsom & St Hellier

However, I'm not sure how clear-cut the distinction between CBT/GET and Activity Management really is in practice: CBT uses activity management and all the 'activity managment' centres use psychologists in the mix, though I don't know if Leeds did during the time of this study (2005-9).

 

[Dolphin]

Thanks.
I've remembered now that Barts supplied two centres for the PACE Trial (seems slightly odd to me that it was called two centres, but maybe the sample sizes of all the other centres were all fairly similar), which suggests it's big centre.

Anyway, it looks like one out of Barts and Bristol had quite poor results. It would be interesting to be more sure. "Barts/PDW" strut around, in a way, as if their service is great.

 

[Simon]

Some thoughts on the paper:

Read the small print first - it matters

• This isn't a clinical trial, instead it uses data from everyday clinical practice, which is a whole lot more messy and harder to control
• Only just over half of all patients completed outcome questionnaires, eg fatigue scores. Although those that completed questionnaires were demographically very similar to those that didn't, as the authors say "we don't know whether patients lost to follow-up had better, the same or worse outcomes at 8-20 months, or how any such differential losses might affect results"
• All comparisons are between outcomes and baseline, not versus a control 'untreated' group. Some of the improvement is likely to be due to natural recovery, or placebo effect, or self-report bias.
• Nonetheless, this much better data than most NHS clincial services collect, and I think, still useful - though findings are not hard and fast. They also cover over 800 patients.

Overall results: "mixed"

• Both fatigue and function improved in patients from baseline to outcome (which varied from 8-20 months). The improvement in fatigue was a drop in the Chalder Fatigue Scale score of 6.8 points, similar to that seen in the PACE trial for CBT/GET (compared with baseline). This is on a 0-33 scale, but 11 is 'normal' and 33 is worst, so it's effectively a 22 point scale, and a 6.8 point improvement is substantial.
• However, Physical Function only improved by 4.4 points (0-100 scale), much less that the +11.6 improvement from baseline of the PACE SMC 'control' group.
• Overall, then, there was a substantial improvement in fatigue and nothing much to report in function - the resutls for function were far worse than reported in the PACE trial (+19 points gain for CBT/GET)
• Results in clincial practice almost always fall short of trials, and not just in CFS. In trials, patients and therapists are likely to be more motivated, there are more resources, standards of therapy are monitored, compliance rates are higher, trials usually have more sessons etc. 'Non-ideal' patients, eg those with additional illnesses are usually excluded from trials. That said, what matters to patients is the effectiveness of the treatment they are likely to receive themselves, and this study is likely to give a better indication than a clinical trial.

cont next post

 

[Simon]

Activity Management vs CBT/GET

In some ways, this is a bombshell:

Comparing the three services that said they offered CBT/GET with the three services that only offered activity management, suggested that patients attending services offering CBT and GET had less im-provement in fatigue at 12 months (fully adjusted coefficient 0.57; 95% CI 0.24 to 0.90; P = 0.001, corresponding to a difference of ~2 points on the Chalder Fatigue Scale).

2 points is huge, given that much of the improvement in baseline is probably self-report/placebo/natural recovery. CBT/GET in PACE only achieved a gain of 3.5 vs SMC. And particularly given that in PACE APT (pacing) came out worse than CBT/GET.

On the other hand, this is comparing 3 services with 3 other services, and there could be many, many confounding factors linked to the particular services inc type of patients (though many baseline factors were adjusted for) and number of sessions, plus group vs individual treatment.

On top of this, looking at clinic websites there seems to be a lot of overlap between CBT/GET and activity management with OTs, physios, psychologists involved in both types of services, and pacing common to all, at least to some extent. Looks to me more like a spectrum of treatments rather than a clear CBT/GET vs Activity Management divide.

The authors argue all this, and I think they have a point. Nonetheless, these results don't support the view that CBT/GET is superior to activity management.

Also, note this:

The lack of patient-level treatment data and a non-intervention group does not allow us to compare treatments or to determine whether observed improvements were due to the interventions themselves

Anxiety, Depression, Illness Duration etc: nothing much predicts response to treatment

• Anxiety and depression at baseline were NOT predictors of outcome so if you had anxiety or depression you would respond the same to treatment as those that did not.
• Illness duration at baseline was NOT a predictor of outcome, which is slightly surprisingly: it was generally thought that chances of recovery are much better in the first years of the illnes, though this is the biggest data set I've seen looking at outcomes so the findings may well be right.
• Worse physical outcome at baseline was associated with a slightly less improvement in fatigue. Other effects, eg worse pain at baseline predicitng slighly less improvement in Depressioin were very small, and probably only statistically significant because of the huge sample size.

Other points

• Progress seems to flatline after 6 months or so, similar to PACE. Looks like there might even be a slight decline but prob not statistically signficant (and note there seem to have been quite a lot fewer sessions per patient than for PACE). Nonetheless, if CBT/GET theories are right, it should be 'onwards and upwards', not flatlining well short of recovery. On the other hand, this is the kind of pattern you might expect from a modestly helpful illness management programme.
• All of the outcome measures in the study were self-reports and as such may be prone to some self-report bias. Ideally objective measures are needed too eg actometers (to measure activity levels), employment levels, benefit levels etc. Where these have been measured in other trials and studies (eg PACE, Wiborg studies, Belgian Audit of CFS services) objective gains have not materialised. However, this ongoing study is now collecting data on school attendance and impact of ill-health on work, so hopefully this will be addressed in future reports.

 

[Esther12]

Thanks a lot for that Simon.

re school attendance being measured in the future: I suspect that this is an outcome measure which could be dramatically improved without one needing to improve patient's health. Generally though, the 'CFS in kids' stuff I've seen seems even less good than the adult stuff, and it's quite possible that a lot of kids given this diagnosis have quite different problems than those found in adults. And things like viral triggers for CFS would be totally different, due to differences in kid's immune responses (no glandular fever!).

re the number of patients looked at: do you think that these represent all of the patients going through these clinics? I had assumed that clinics would get though more patients than this, but I don't really know what it's sensible to assume.

It looks like this BACME report covers the same period (up to 2009). It seems to show pretty variable results from different services, and looked at data more than six services. I wonder how they choose which ones were put in to this new report? I'd love to get my hands on all the raw data, and information about which centre was which.

http://www.bacme.info/document_uploads/BACME_docs/ReportAPPGFINAL.pdf

From page 16, patient reports seem to indicate that there is a significant role for 'pacing':

“Meeting people with CFS. Learning about pacing. Learning to live with the CFS.”

This isn't 'recovery' style CBT:

“Drive to do the shopping (long-term goal).”

I'd like to know how they selected patient comments.

The report also says:

The data represented here has been systematically collected annually across many services and represents a high response rate.

However this is less likely because of the high percentage of patients that responded, the anonymised collection of data, and the consistent outcome in data collection across many teams.

Is it really a high response rate? I'd assumed that many more patients per year were being 'treated' than this. I don't think that they ever said what the response rate was either.

 

[Simon]

Thanks a lot for that Simon.
...
re the number of patients looked at: do you think that these represent all of the patients going through these clinics? I had assumed that clinics would get though more patients than this, but I don't really know what it's sensible to assume.

...This isn't 'recovery' style CBT:


Is it really a high response rate? I'd assumed that many more patients per year were being 'treated' than this. I don't think that they ever said what the response rate was either.

Thanks.

I think generally in clinical practice, and probably in the PACE trial too, CBT isn't 'recovery' style - I suspect that many therapists are pragmatic and recognise the limitations faced by patients; certainly PACE results indicated that if they used 'recovery style' CBT it didn't work as such.

Response rate was 53% for fatigue outcomes, 50% with all outcomes. That's lousy by trial standards, but probably pretty good for clinical practice. The paper says they only included clinics with >40% response rates so excluded clinics were all less than this.

 

[Esther12]

re CBT types: But then they'll talk as if the different CBT trials are all testing the same thing! Treating all CBT approaches to CFS as if they're the same is like assuming all drugs are the same: "Drugs are an evidence based treatment shown to improve symptoms." A failure to explicitly distinguish between them also makes it difficult for patients to give informed consent.

re PACE CBT: There is something from the/a PACE trial PI lumping their CBT with 'recovery style' CBT, and away from Friedberg 'learn to live with' CBT. I can't remember the quote or where it was from though. I could probably dig it up if you wanted.

re Response rates: I may not have been clear, but everything I was quoting from from the older report where they did not provide the response rate.

re the response rate for this paper being over 40%: Do you think that is of all patients seen, or of all patients asked to fill in questionnaires?

We included patients who had attended one of six NHS specialist CFS/ME services during the period 1 January 2005 to 31 December 2009. The six services were chosen because they had been collecting 12-month outcome data for at least 1 year during the study period and had achieved >40% 12-month follow-up. Patients were included if they were age 18 years or older and were diagnosed with CFS according to Centers for Disease Control criteria.1,2

Patient-level data
In each of the six services, patients routinely completed the following inventories prior to their initial clinical assessment (baseline)

I'd guess that the response rate was for all patients, but that different services started using the questionnaires at different times.