1. Patients launch $1.27 million crowdfunding campaign for ME/CFS gut microbiome study.
    Check out the website, Facebook and Twitter. Join in donate and spread the word!
ME/CFS and Beating the Clock
For Jody Smith, the ticking of a clock was enough at one time to chase her back to her bed. But with the passage of time, she has been able to reclaim her living room ...
Discuss the article on the Forums.

Wessely studies from the 90s that keep cropping up

Discussion in 'Latest ME/CFS Research' started by oceanblue, May 15, 2011.

  1. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Normally we comment on recent research but there were a series of papers from Simon Wessely and colleagues that are still cited and have become quite influential, so I thought they should have their own thread. Some have already been discussed in part on other threads. The papers seem to come out of a single large study done in the early 1990s:

    Population based study of fatigue and psychological distress, 1994
    T. Pawlikowska, T. Chalder, S. R. Hirsch, P. Wallace, D. J. Wright, and S. C. Wessely
    The ultimate source of the normative fatigue data used in the PACE trial, and source of the graph linking psychological morbidity to fatigue

    Postinfectious fatigue: prospective cohort study in primary care, 1995 (pdf), 1995, Wessely S, Chalder T, Hirsch S, Pawlikowska T, Wallace P, Wright DJ.
    Basis for claim there is no link between common viral infectons and CFS


    Psychological symptoms, somatic symptoms, and psychiatric disorder in chronic fatigue and chronic fatigue syndrome: a prospective study in the primary care setting, 1996 (pdf).
    S Wessely, T Chalder, S Hirsch, P Wallace and D Wright
    Source of a graph linking number of CDC symptoms to psychological morbidity.

    The prevalence and morbidity of chronic fatigue and chronic fatigue syndrome: a prospective primary care study. 1997. S Wessely, T Chalder, S Hirsch, P Wallace and D Wright
    Claims a prevalence of 2.6% for CFS (CDC 94 criteria) in the UK.
  2. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Fatigue vs Psychological morbidity

    This relates to the Pawlikowska paper and had been discussed on the PACE thread after featuring in a recent talk by Simon Wessely.

    Fig 3 from this paper shows Chalder Fatigue score vs mean score for the General Health Questionnaire (GHQ-12), with a very strong correlation between the two. For convenience, I've reproduced it here as it's freely available in the open access document above; if anyone thinks this is a breach of copyright please let me know.
    fatigueghq.jpg

    I think that graph was CDC symptoms (inc fatigue) vs psychological morbidity (using the CIS-R) from a sister paper, but it's a similar idea.

    I'd agree that GHQ-12 points such as not 'been able to concentrate on whatever you are doing' or a feeling that you can't play a useful part in things, can be a direct result of ME rather than any useful measure of psychological health. Also, as the study isn't prospective it doesn't say anything about the direction of the relationship. Fatigue and psychological health may both be linked by a strong correlation to physical health.

    However, there is another, simpler explanation for the findings (and I like simple).

    All particpants were asked what cause they attributed to their fatigue:
    - 40% cited psychosocial causes eg work stress
    - 17% cited psychological causes
    - 15% cited physical causes
    - 20% didn't answer this question, 7% cited other causes and 2% cited pregnancy.
    Only 0.25% cited CFS.

    So what we are looking at here is a graph of explained fatigue and, unsurprisingly, the most common cause of fatigue is psychological/psychosocial, but physical illness also causes fatigue. So what? CFS is about unexplained fatigue, and the graph tells us nothing about that.

    Nice graph, strong correlation but it tells us nothing about whether or not psychological factors account for unexplained fatigue. So it's "disappointing" that Simon Wessely used the slide to imply a causal relationship between psycholgical problems and unexplained fatigue.
  3. Enid

    Enid Senior Member

    Messages:
    3,309
    Likes:
    840
    UK
    I am glad you are able to analyse oceanblue (am sure many others can too) but when these names come up I tend to get a serious health warning. Still being cited and influential is a worry, so the "research" should be exposed for what it is.
  4. Dolphin

    Dolphin Senior Member

    Messages:
    6,525
    Likes:
    4,929
    Yes, good points. And good you corrected my error.

    I have found it frustrating on occasion coming across very flawed old papers, which nobody else replied to, but not having the chance to write in as the time had passed. It's one of the things that motivates me to prioritise letter writing.
  5. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Pawlikowska population fatigue study vs Norwegian fatigue study

    The Pawlikowska study was the ultimate source for the 'normative' fatigue levels used by PACE trial (see below for full post on this, taken from the main PACE thread). However, a Norwegian study that also used the Chalder Fatigue Questionnaire found lower levels of fatigue than Pawlikowska, even though it used an older population (age 19-90).

    Crucially, the Norwegian study was based on a representative sample of the population, taken from Norway's National Population Register. Both studies used the same definition of fatigue 'caseness' as the PACE protocol, a bimodal score of 4 or more. Here are the results compared:
    So, using the same scale but a representative sample the Norwegian study found much lower levels of fatigue than Pawlikowska, a study that did not use a representative sample. Nb a separate study on the SF36 scale did not indicate substantial differences in fatigue between the UK, US and Sweden, indicating that the Norwegian data is likely to be representative of the UK too.


    that post from the PACE thread:
    Esther12 likes this.
  6. Enid

    Enid Senior Member

    Messages:
    3,309
    Likes:
    840
    UK
    Thanks for your analysis oceanblue - looks like the psychs are trying to psychologise us all in one way or another - "gods of the gaps" as another member described them. (Chalder's training - psychiatric nursing).
  7. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Summary of the Wessely Primary Care study

    It would probably help to explain the basic structure of this study: the overall design is pretty good, perhaps one reason why the papers that emerged from the study became quite influential. The study has 4 stages:

    Stage I: The Prequel
    Everyone registered with 6 primary care practices in the south of England and aged 18-45 was mailed the Chalder Fatgiue Questionnaire (CFQ) and the General Health Questionnaire (GHQ, used as a measure of psychological health). The data was used to measure 'pre-visit' health.

    Stage II: The Visit
    2 cohorts, each of around 1,000 patients:
    • Viral cohort: patients visiting the practice with suspected viral infections, age 18-45
    • Non-viral cohort: next patient to visit with non-viral complaint, age 18-45
    Again, the GHQ-12 and CFQ questionnaires were used.

    Stage III: 6 month follow-up
    Both cohorts were contacted 6 months later, again using the GHQ-12 and CFQ questionnaires.
    Responders (Viral & non-viral cohorts) were sub-divided into Chronically Fatigued (fatigued for 6 months ie at Stage II & III) and non-fatigued patients

    Stage IV: Detailed work-up of Stage III (case control study)
    All Chronically Fatigued subjects were asked to return for further assessment, along with matched non-fatigued controls (ie for every fatigue case, one non-fatigued patient of the same age and sex was selected) for both viral and non-viral cohorts, giving:
    • Viral cohort: fatigued
    • Viral chohort: matched non-fatigued controls
    • Non-Viral cohort: fatigued
    • Non-Viral chohort: matched non-fatigued controls
    This was the point at which full-blown CFS was assessed for the first time, along with psychological morbidity via the CIS-R interview and symptoms via the Somatic Symptom Checklist (including the CDC-94 symptoms).

    So far, so good...
  8. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Serious flaw in study showing 'common infections don't lead to CFS'

    Postinfectious fatigue: prospective cohort study in primary care, 1995 (pdf), Wessely S, Chalder T, Hirsch S, Pawlikowska T, Wallace P, Wright DJ.

    This study compared a cohort of around 1,000 patients who went to see their doctor with a viral illness with a matched cohort of patients seeing their doctor for other reasons - and found no difference in their rate of CFS 6 months later:
    Unfortunately, it appears that the study failed to exclude any existing cases of CFS in both cohorts - it wasn't looking at NEW cases after infection, it was looking at total cases (of which new cases would only be a small proportion). So it's unsurprising there was little difference between the 2 cohorts.

    See the above post (#7) for an outline of the methodology.

    The main results for CFS 6 months after visiting the surgery show similar rates for the viral and non-viral cohort

    Chronic Fatigue: 9.9% (Viral), 11.7% (non-viral)
    Oxford CFS: 1.3%, 1.9%
    CDC '94 CFS: 1.0%, 2.0%

    But wait, 2% of patients have CFS 6 months after visiting their doctor with a non-viral complaint?! Something odd is going on here, and that odd something is that the paper does not state at any point that existing CFS cases are excluded. Oh. In fact, it does note that:
    Ah. These results look rather different:

    Chronic Fatigue: 4.6% (Viral), 4.2% (non-viral)
    Oxford CFS: 0.8%, 0%
    CDC '94 CFS: 0.9%, 0.4%

    As the number of cases are so small the differences between viral and non-viral is not significant but the picture looks very different.

    In short, there's nothing you can conclude from this study about the link between common acute infections and subsequent develoopment of CFS

    Also, the 'viral' cohorts in not quite what you might expect: 36% were diagnosed as Flu (19%), Cold (16%) and Glandular Fever (1%) but the most common diagnosis was 'sore throat' (33%) and the rest were other ailments.
  9. Dolphin

    Dolphin Senior Member

    Messages:
    6,525
    Likes:
    4,929
    Thanks oceanblue. Those incidence figures are a lot lower than say the Dubbo study found for a few confirmed specific infections; generally the incidence is around maybe 10% although given that a lot of these seem to get better pretty quickly, I can be slightly reluctant to say some are "proper" cases.

    A general problem I have with a lot of these studies that look at pre-disposing factors e.g. before a viral infection (I think some of these studies looked at that) is that many patients many not be in a "pre" state at all but instead have undiagnosed ME/CFS; some may have symptoms while others may not have that many symptoms but this is because they have adjusted their life in some way e.g. for the years I was going around undiagnosed, I generally had to give up any competitive sports and ration my social activities which also made me feel a bit unhappy and frustrated - not depressed as I could laugh easy but sometimes unhappy that I couldn't do everything I wanted, etc. It could hit other people worse e.g. people around them could be getting annoyed with them for not "pulling their weight" which could cause extra strife, etc. So I'll feel more confident with studies looking at pre-disposing factors when they are better at testing for it (and also when studies actually employ such tests i.e. it's no good having good tests if "cheap" paper research doesn't actually use them).
  10. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Study claiming 2.5% prevalence rate for CFS didn't diagnose properly

    The prevalence and morbidity of chronic fatigue and chronic fatigue syndrome: a prospective primary care study. 1997. S Wessely, T Chalder, S Hirsch, P Wallace and D Wright
    Full text pdf


    This study found a CFS prevalence rate of 2.6% for CDC '94 (Fukuda) criteria but failed to evaluate patients according to the published critieria. It excluded only a suspiciously-small proportion of them due 'explained fatigue' and so the stated rate is almost certainly a huge overestimate.

    Reality check
    These prevalence figures fail a basic reality check: 1 in 40 people having CFS/ME is not credible. In the UK, primary care doctors have around 1,500 registered patients each (almost the entire population is registered) which would give them nearly 40 CFS patients each. At the time the study came out, many doctors pointed out they have only a handful or less of CFS patients each.

    2 population studies found much lower rates for CFS: Jason 1999 found 0.42% and Reyes 2003 found 0.24%. The Reeves 2007 study, which used the wacko Empirical Criteria that allows fatigue purely due to depression, found a similar rate of 2.5% - which again makes the Wessely figures look suspiciouly high.

    Study flaws
    See the above post (#8) for an outline of the methodology.
    The key problem seems to be that the study didn't properly clinically evaluate patients as required by the CDC/Fukuda definition (pdf)
    Fukuda requirements:
    1. Thorough history of medical and psychological circumstances at the onset of fatigue - as well as many other factors
    2. A mental status examination
    3. A thorough physical examination
    4. A minimum battery of lab screening tests
    By contrast, there the study used no physical examination; only some of the lab tests; no thorough medical history (they just used existing patient notes); and the CIS-R questionnaire that only looks at overall psychological health but doesn't diagnose any specific disorders which is necessary to apply exclusions.

    Tiny level of exlusions
    Only 3 patients (about 7%) of patients were excluded from CFS due to medical exclusions. Another 3 were excluded from the Chronic Fatigue group due to lab test results: these 3 may or may not have met the critieria for full CFS.

    This level of exclusion is minimal compared with the Jason and Reyes studies, both of which used a detailed medical history, physical examination, proper psychiatric evaluation and battery of lab tests as required by Fukuda. Their clinical evlauations excluded 73% (Jason) and 77% (Reyes) of their 'CFS-like' patients for physical or psychiatric reasons. However, a substantial number of patients had been excluded by the questionnaires even before the clinical evaluation, so the total exclusion rates were even higher.

    The Wessely study suggested that one reason for the low level of exclusions is they only looked at patients aged 18-45, but 67% of the Jason population and 61% of the Reyes population were in the same age group so this is highly unlikely to explain the massive differences (especially as psychological disorders are common in younger people).

    The 2.6% prevalence rates lacks credibility.
  11. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Agreed. Having a diagnostic test would make life a lot easier.
  12. Dolphin

    Dolphin Senior Member

    Messages:
    6,525
    Likes:
    4,929
  13. Dolphin

    Dolphin Senior Member

    Messages:
    6,525
    Likes:
    4,929
    Thanks for the analysis of the paper, oceanblue - well done.

    In some studies, if there is not an assessment of cases individually by a clinician, ordering whatever tests they might think is necessary, people are classed as "CFS-like" rather than CFS cases.

    Another thing I remember finding odd about that paper was that the prevalence for the Oxford criteria (without comorbid psychological disorder) was actually lower than the Fukuda prevalence (2.2% vs 2.6%).
  14. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    Very odd, especially as after excluding those with 'psychological morbidity' the Oxford criteria rate was higher than CDC (0.7% vs 0.5%), wheras given the nature of the Oxford Criteria you would expect it to be lower. Further evidence, I think, that the figures are unreliable.
  15. Snow Leopard

    Snow Leopard Senior Member

    Messages:
    2,231
    Likes:
    1,594
    Australia
    The thing about these 2.2-2.6% incidences, is that when you compare them to the associated 'recovery' rates, (a) either those recovery rates are misreported (b) Most of the group suffers from a relapse-remitting disease, resulting in the recovery rates being flawed due to lack of time series studies. (the CDC suggested this in one of their papers - inconsistent results from one period to the next) (c) Or around 4-5% of the population has suffered from 'chronic fatigue syndrome' some time in the past.
    What I'd really like to see is a population study measuring the numbers of those who have greater severity - unable to work for over 3 years for example - then we'd probably find our more familiar 0.2-0.5%.
  16. SilverbladeTE

    SilverbladeTE Senior Member

    Messages:
    2,076
    Likes:
    1,578
    Somewhere near Glasgow, Scotland
    I just see this garbage as good material that one day will be part of the prosecution evidence of a "Crimes Against Humanity" trial for these ass clowns :p
  17. Nielk

    Nielk

    Messages:
    5,106
    Likes:
    4,685
    Queens, NY
    I'm glad to see you woke up:thumbsup:
  18. SilverbladeTE

    SilverbladeTE Senior Member

    Messages:
    2,076
    Likes:
    1,578
    Somewhere near Glasgow, Scotland
    Nielk,
    hehe! ;)

    Oh did you see a report said that troops serving in Afghanistan are suffering higher rates of mental illness?
    yet Weasely had concluded there was no unusual rates of PTSD, from there...
    Ah, the wheels of justice grind oh so slow, but so exceedingly fine! :p

    Silverblade's Thought For Today:
    The impact of the Weasels on the plight of ME Sufferers, is analogous to the impact of diahorea on your underwear! :D
  19. Enid

    Enid Senior Member

    Messages:
    3,309
    Likes:
    840
    UK
    Trying to sweep all into their "spiders" web degree by creeping degree. Just wondering if I dare do the washing up without worrying about my "attitude" to it.
  20. oceanblue

    oceanblue Senior Member

    Messages:
    1,174
    Likes:
    342
    UK
    More concerns about the accuracy of their CFS 'CDC-94' diagnosis

    I've already commented on the implausibly high prevalence rate of 2.5% (1 in 40) for CFS found by the study - probably due to a failure to exclude patients with explained fatigue - but I've noticed a few more factors that might be relevant.

    1. CFS was primarily diagnosed via the 'Chronic Fatigue Questionnaire'; this unpublished, unvalidated questionnaire came from Trudie Chalder's 1990 MSc Thesis. The study fieldwork for all these papers was carried out in 92/93 - yet the Fukuda/CDC-94 criteria the study uses weren't published until 1994 (obviously). Did they use time travel to make sure they had the right criteria for their fieldwork? I emailed Trudie Chalder to request a copy of her questionnaire but as yet have received no reply.

    2. They refer to 9 (or more, Wessely 1996) CDC symptoms when there are only 8. Table 5 of this paper list 9 symptoms first that appear to be their CDC ones yet includes 'muscle weakness' and 'fever/chills' that are NOT CDC symptoms, but misses out 'sore throats' that is a CDC symptom. They may have diagosed some case of CFS on the basis of the wrong symptoms.

    3. The authors say they only did some of the mandatory CDC-94 lab tests 'for reasons of cost'. What kind of an excuse is that?

See more popular forum discussions.

Share This Page