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!
The ePatient Revolution
Ryan Prior shares his experience and his thoughts from attending the Stanford Medicine X Conference as he contemplates the rising of the ePatient Revolution ...
Discuss the article on the Forums.

Chalder & Godfry 2009 - CFS in adolescents: Do parental expectations of their child's IQ...

Discussion in 'Latest ME/CFS Research' started by Esther12, Apr 2, 2013.

  1. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    Just a short thing because I keep seeing this paper being cited around the place as evidence that the parents of CFS patients tend to have unusually high and distorted expectations of their children, without the limitations of the study being explained.

    It seems so common in CFS for weak a study to be published, and the paper will include provisos, but then go on to be cited uncritically in other papers and presentations.

    It's not very interesting, but I thought I might as well post my notes up:

    It was not a prospective study, so the testing is only done once people already have CFS.

    As it says: "Chronic fatigue syndrome (CFS) is characterized by prolonged physical and mental fatigue of at least 6 months' duration that severely affects functioning[1]"

    It is not surprising that children who are suffering from a condition characterized by mental fatigue tend to under perform their parent's expectations on IQ tests more than health controls.

    The paper actually says:

    Who knows what other confounding factors there may be - perhaps parents of ill children try to be 'supportive' by speaking more highly of them? Who knows. There's some reason to think that there's a problem with selection bias for patients who end up in tertiary care for CFS, and their results show that there is for for students/parents willing to volunteer for IQ tests at school. (edit: As Snow Leopard mentions beneath: "There is clearly heavy selection bias in the group. 21% of the CFS group and 52% of the control group tested in the top 10% (effectively 90th percentile IQ). You can't draw any conclusions when the groups are so heavily skewed like that.") This study doesn't really let us say much about anything.

    Also:

    Seperate minor point: This study reported higher (although really insignificant 7% vs 11%) levels of family history of fatigue in controls than patients. I've seen others talk about parental fatigue as a psychosocial risk factor (although that always seemed a bit risky considering our poor understanding of any genetic component).

    Here's the Chalder slide which prompted this, but I've seen similar things around the place. (Compare and contrast with what's said in the paper: "However, the discrepancy between expectation and actual IQ did not differ significantly between groups")
    [​IMG]

    https://docs.google.com/viewer?a=v&q=cache:JmsJB03WQOEJ:www.rcpsych.ac.uk/pdf/Chalder%20T.pdf &hl=en&gl=uk&pid=bl&srcid=ADGEESgV7Vm8xtaG6kaKTp08HEMAT8eRJZJI-6iLy8COa97xRQQPEgP4OHUsIm6RpCzMuQSBRbVuXDtNV8yPVygHuVXSHe0Yl7rGiSbjuB6X2g_F1po4ni1LY9vUWlwg2atWknfNOcEk&sig=AHIEtbRJpimAIgsKIwxt22wbZOBniWkJzA


    There's a clearer example of her exaggerating the value of her work here: http://forums.phoenixrising.me/inde...o-educational-intervention-to-aid-reco.13326/
     
    snowathlete, Roy S and Valentijn like this.
  2. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    Not to self: I want to check out this reference they used too. I just looked at the abstract, and I thought it might be less than convincing.

     
  3. Snow Leopard

    Snow Leopard Senior Member

    Messages:
    2,410
    Likes:
    2,054
    Australia
    There is clearly heavy selection bias in the group. 21% of the CFS group and 52% of the control group tested in the top 10% (effectively 90th percentile IQ). You can't draw any conclusions when the groups are so heavily skewed like that.
     
    alex3619, Enid, snowathlete and 2 others like this.
  4. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    Yeah.

    With a lot of these sorts of CFS papers, I just find myself wondering "what question are they trying to answer? How could they think that this would move forward our understanding?" So often their studies' designs do not allow meaningful results to be gathered. Why do they bother? Just to get something published? On the assumption that they can spin shoddy results according to their own interests? Did they not realise these problems would prevent us from usefully interpreting their results?

    A lot of the time they are just churning out papers which are worse than nothing.
     
    Last edited: Sep 13, 2014
    Tito, wdb and Valentijn like this.
  5. PhoenixDown

    PhoenixDown Senior Member

    Messages:
    258
    Likes:
    235
    UK
    Something to do with not wanting to turn their backs on ideas that they built their careers on.
     
    ukxmrv, Enid, SilverbladeTE and 2 others like this.
  6. MeSci

    MeSci ME/CFS since 1995; activity level 6

    Messages:
    3,824
    Likes:
    4,586
    Cornwall, UK
    I think you have hit the nail on the head there.
     
  7. perovyscus

    perovyscus

    Messages:
    61
    Likes:
    7
    So strange that I was reading this today and this thread comes up.

    Genders are not matched, ages are not matched and the method of IQ testing shouldn't be used in literature. Furthermore, 52% of control groups tested in the A percentile, so why would the control parents overestimate their children's ability. Which parent was tested?

    The letter method of IQ isn't "extensively tested" it's been replaced by a plethora of other tests, including WAIS, and these people are ostensibly adults anyways.
     
    Snow Leopard likes this.
  8. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    Thank perovyscus. I don't know much about IQ testing, and hadn't thought about how important age is in understanding intelligence for teenagers. I did notice that the testing described seemed to be very brief, but chose not to look in to it. Even before all that, it seems that the results are really uninterpretable!

    If this study does go on being cited and presented as important, it could be worth doing a more thorough examination of it. (Or if this is an area you're familiar with, and you fancy writing something up now, I'd certainly be interested).
     
  9. Tito

    Tito Senior Member

    Messages:
    300
    Likes:
    407
    From what I read, if I understand well, parents are asked to estimate their children's IQ and that number is then compared to the result of a proper IQ test the children take. But how do parents 'guess' their children's IQ. how can they say he/she has 120 and not 110? Or do they say that in comparison to their (perceived) IQ? If they say he/she is "very intelligent", it means 120, 140 or 160?

    Beyond that point, I too find the subject of that research just bizarre... And the treatment will be: "parents, your child is not as clever as you thought. Stop showing interest and he'll get better"?... And then the parents will not seek medical advice any longer and guess what? "That patient is not coming to our clinic any longer, he must be cured"...
     
  10. Tito

    Tito Senior Member

    Messages:
    300
    Likes:
    407
    I found the study :
    http://www.kcl.ac.uk/innovation/gro...ssets/2009/Godfrey2009IQinchildrenwithCFS.pdf

    About the issue of parental assessment:

    "This measure was constructed specifically for this research. Parents' expectations of their child's intelligence were assessed as a percentage using a visual analogue scale. Parents were told that 50% represented a child of average intelligence and that most children fell somewhere in the range of 25% to 75%. Thus parents placed an X at a point they felt was most appropriate on a 10-cm line which was marked with 0%, 50%, and 100% (at 0-, 5-, and 10-cm intervals). This was then scored by measuring in millimeters from 0 to the location of the X giving a score out of 100 equal to the number of millimeters. Adolescents then assessed their own intelligence using the same method as described above".

    How bizarre that is... I had hoped that at least they would have asked for each IQ test question whether their child could have answered it or not, instead of rating their child against other children they may have heard about (ie. hypothetical children).
     
    Valentijn likes this.
  11. Valentijn

    Valentijn Activity Level: 3

    Messages:
    6,686
    Likes:
    10,098
    Amersfoort, Netherlands
    Yeah, I think it's about as useful as the result that something like 90% of therapists rate themselves "above average" :p
     
    biophile and ukxmrv like this.
  12. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    Cross posting here, but I just started another thread on a paper which had a section that reminded moe of this thread: http://forums.phoenixrising.me/inde...n-be-a-nuisance-replication-is-science.22690/


     
  13. maryb

    maryb iherb code TAK122

    Messages:
    2,920
    Likes:
    2,054
    UK

    We'd like the parents to make an appointment though.................
     
  14. perovyscus

    perovyscus

    Messages:
    61
    Likes:
    7
    In case you're still interested, I've attached the study "Cognitive idiosyncrasies among children with the
    chronic fatigue syndrome: anomalies in self-reported activity levels".



     

    Attached Files:

    Esther12 likes this.
  15. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    Thanks for that. I've had this window open on my PC since you posted, and still have not got arround to reading it, but wanted to say 'ta' anyway.
     
  16. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    I'm not sure if I looked at this before. Stumbled upon it now, and posting some notes. Thanks again.

    The main problem with this study is that there's no ill health control group. Parents speaking positively about their children's activity level before they got ill, or individuals doing the same about themselves, could be a common theme, or could be a reaction to the psychosocial approaches to CFS, and reflect a desire to show that they do want to be active rather than adopt the sickness role. The same potential biases could effect claims about desired activity levels.

    Without trying to take account of these problems, it's difficult to see how any sort of conclusions can be drawn from this sort of study.

    These sorts of things, and the prejudices of staff, can effect clinical observations:

    2 months of fatigue in teenagers is not that long, is it?

    I see so much potential for bias, and the parents of those with CFS to attempt to display 'positivity' in the following questionnaire:

    Figure 1 on page 6 makes it look as if the measures of activity counted what the children considered to be rest/inactivity as activity. Looking at those results, it seems very difficult to draw any conclusion.

    Some more results:
    (Pardon the funny formatting)

    One other thing: Why don't people desire to e super-dooper active? What's bad about that? If we're talking about magic desires, then I would want pure pleasurable activity. If we're talking about desires tempered by reality, then I would expect CFS patients to want to be less active. It's not really clear to me from the way questions asked, exactly what was meant by them.

    They've got no ill-health control. So do not know (or have not reported) what sort of readings one should expect from those whose health does seriously interfere with normal activities.

    Also, they did select the patients themselves, as teenagers who had reported two months of fatigue. There probably was considerable variability in the health of this group.

    They do mention this on the final page, after a load of speculation about cognitive perpetuating factors:

    At this point? Maybe in the intro would have been better!

    Oh, and they mention another point I'd made:

    So after a load of meandering speculation, they basically admit that really their study is of little value.

    Why is this still being cited?
    Reliable and objective measures of activity levels would be a preferable way of measuring outcomes.
    Wrote this while half-asleep, so sorry if a bit confused!​
     
  17. alex3619

    alex3619 Senior Member

    Messages:
    7,668
    Likes:
    12,428
    Logan, Queensland, Australia
    I haven't read the entire thread yet, but I thought I would raise a point. IQ testing in CFS patients, even children? I have a university medal. I was a PhD candidate. Yet there are days when I probably function in the bottom five percent of the population ... no math skills, pathetic visual skills, almost no verbal skills. If parents rate their children's IQ on how they used to be, or were perceived to be, then they will always rate them higher than their current effective IQ. Its called brain fog for a reason.

    I really wish they would do some work testing IQs on patients to see how much they drop with ME or CFS, and how much they drop during a crash, before they jump to unjustified conclusions.

    PS Of course like all their studies, they seem to be really studying CF not ME.
     
    Valentijn and MeSci like this.
  18. Snow Leopard

    Snow Leopard Senior Member

    Messages:
    2,410
    Likes:
    2,054
    Australia
    None of these studies used unbiased or carefully controlled selection of participants and none have been replicated, so it is hard to draw any conclusions.
     
    Valentijn likes this.
  19. Esther12

    Esther12 Senior Member

    Messages:
    5,375
    Likes:
    5,864
    This is what they Chalder and Godfrey 2009 paper cited the older paper for:

    Even if we assume that the results from the study were accurate and meaningful, they did not show that.

    Both CFS and healthy children rated themselves as being 'active' for less time than the activity monitor, and this could just indicate that the 'activity monitor' has a different view as to what being 'active' means than most children. Equally, the study did not show that the children and their parents had higher expectations for their post-CFS activity levels than was realistic for children of that age. Fry and Martin did not show what would be a realistic expectation or desires for children's post-CFS activity levels.

    Actually, the results for 'expected' activity were:

    CFS child: 71.7 (16.9)
    CFS parent: 61.7 (16.5)
    Healthy child: 71.2 (19.6)
    Health parent: 72.4 (16.7)

    So the parents had lower expectations, and the children had basically the same expectation.

    They're just building 'evidence based' theories out of nothing? I was really knacked when I was reading this paper yesterday, so I'm a bit worried I went zzzzz for an important part of the paper, and missed the bit which showed it was of real value, but it looks like this is just another example of Chalder totally misrepresenting work in order to exaggerate her own expertise.

    It reminds me of that Ioannidis paper I was reading yesterday:

    http://forums.phoenixrising.me/inde...e-false-john-p-a-ioannidis.25381/#post-388736
     

See more popular forum discussions.

Share This Page