Contribution of individual psychological and psychosocial factors, Dubbo Infection Outcomes Study, 2019, Cvejic, Hickie et al

Discussion in 'Psychosomatic research - ME/CFS and Long Covid' started by Hutan, May 7, 2020.

  1. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,219
    Location:
    Aotearoa New Zealand
    Contribution of individual psychological and psychosocial factors to symptom severity and time-to-recovery after naturally-occurring acute infective illness: The Dubbo Infection Outcomes Study.
    BRAIN BEHAV IMMUN. 2019.
    The role of psychosocial factors in disease severity in post-infective fatigue.

    https://www.ncbi.nlm.nih.gov/pubmed/31376496

    We don't seem to have a thread on this study yet

    Abstract
    BACKGROUND:
    Substantial heterogeneity exists in both the severity of symptoms experienced as part of the sickness response to naturally-occurring infections, and the time taken for individuals to recover from these symptoms. Although contributing immunological and genetic factors have been previously been explored, less is known about the role of individual psychological and psychosocial factors, which may modulate the host immune response, or contribute independently, to symptom severity and duration.

    METHODS:
    Longitudinally-collected data from 484 Caucasian participants (mean age: 33.5 years; 51% women) experiencing a naturally-occurring acute infective illness enrolled in the prospective Dubbo Infection Outcome Study (DIOS) were analysed. At intake and subsequent follow-up assessments, self-report questionnaires were used to ascertain individual psychological and psychosocial characteristics and symptom information. Principal component analysis was applied to symptom data to derive endophenotype severity scores representing discrete symptom domains (fatigue, mood, pain, neurocognitive difficulties) and an overall index of severity.

    The contribution of individual psychological (trait neuroticism, locus of control, and illness behaviours) and psychosocial factors (relative socioeconomic advantage) to endophenotype severity at baseline were examined using multivariable linear regression models; interval-censored flexible parametric proportional hazards survival models were used to explore time to recovery (defined using within-sample negative threshold values).

    RESULTS:
    After controlling for time since symptom onset, greater levels of trait neuroticism consistently predicted greater symptom severity across all symptom domains (all p's < 0.015). Similarly, greater relative socioeconomic disadvantage was significantly associated with greater severity across all endophenotypes (p's < 0.025) except neurocognitive disturbance.

    Locus of control and illness behaviours contributed differentially across endophenotypes. Reduced likelihood of recovery was significantly predicted by greater initial symptom severity for all endophenotypes (all p's < 0.001), as well as higher levels of trait neuroticism.

    CONCLUSIONS:
    Individual psychological and psychosocial factors contribute to the initial severity and to the prolonged course of symptoms after naturally-occurring infective illnesses. These factors may play an independent role, represent a bias in symptom reporting, or reflect increased stress responsivity and a heightened inflammatory response. Objective metrics for severity and recovery are required to further elucidate their roles.
     
    Last edited: May 7, 2020
    Simone, Woolie, Sean and 3 others like this.
  2. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,219
    Location:
    Aotearoa New Zealand
    My recollection of an earlier Dubbo study was that they said that none of the psychological measures predicted time to recovery; only initial symptom severity was related.

    It looks as though they have re-analysed the results, adjusting for time since symptom onset ( :confused: - not sure what they were doing there) in order to find that more neuroticism predicted greater symptom severity.

    :banghead: The abstract provides no evidence for this. At best, (assuming whatever data manipulation done was valid), individual psychological and psychosocial factors (as measured by the surveys) correlate with initial severity and likelihood of recovery.

    I note the results section gives P values for the relationship between initial symptom severity and reduced likelihood of recovery only. It does not give P values for a relationship between neuroticism and likelihood of recovery. It also does not give P values, or even direction of correlation, for hinted-at-relationships between locus of control and 'illness behaviours' (whatever they are) and reduced recovery for subsets of the patient population.

    The rest of the conclusion is a bit more non-committal, but the damage has already been done by then.

    It would be good to see the full paper. I find this all a bit demoralising. Presumably we were all quoting that initial finding too much, the one that said that psychological factors didn't affect recovery, so they had to write something new? Something with all the BPS buzz words and a lot of obfuscation.

    It's probably a paper worth picking apart a bit.
     
    bobbler, RedFox, Simone and 20 others like this.
  3. Invisible Woman

    Invisible Woman Senior Member (Voting Rights)

    Messages:
    10,280
    Based on Hutan's initial post -

    It seems to me that simply assessing someone's mental state when they are ill may alter the response simply by making someone think more their symptoms and the inconvenience and possible worry caused.

    Another issue is that we don't have any data from these patients prior to onset of illness. So there is no comparison pre and post onset. So we do not know that psychological changes haven't been caused by becoming ill, rather than affecting illness severity.

    Isn't it possible that someone might seem more neurotic when faced with a sudden and unexpected, albeit temporary, loss of control in areas that are very important to them? Especially if you are a lone parent or the only wage earner , for example, and have no firm idea how long temporary might last.
     
    bobbler, Simone, sea and 15 others like this.
  4. Hoopoe

    Hoopoe Senior Member (Voting Rights)

    Messages:
    5,417
    Why would you have to control for time since symptom onset in a longitudinal study? I don't understand what was done here and why.
     
    Simone, janice, Forbin and 5 others like this.
  5. Snow Leopard

    Snow Leopard Senior Member (Voting Rights)

    Messages:
    3,860
    Location:
    Australia
    Psychosocial factors certainly affect the reporting of symptoms on symptom questionnaires as stated in the conclusion, so without objective signs of symptoms or their impact (actigraphy etc), the results are inconclusive.

    As @Hutan suggests, this is a post-hoc analysis and therefore is "suggestive" quality at most (the lowest quality evidence).
     
    bobbler, Simone, sea and 8 others like this.
  6. Andy

    Andy Committee Member

    Messages:
    22,989
    Location:
    Hampshire, UK
  7. Hoopoe

    Hoopoe Senior Member (Voting Rights)

    Messages:
    5,417
    The responses to this article could reinforce the belief of the investigators that patients are neurotic. Maybe someday they'll consider how it feels for highly stigmatized and neglected patients to have a group of researchers regularly searching for flaws in patients thinking and behaviour, with often questionable methods and publishing the results in various media. Stigmatization of ME/CFS is no joke. It kills and destroys lives. It is serious threat to the well being of patients.

    I question that I had while thinking about this study is whether this kind of personality research has ever produced anything useful for patients with any illness, or if it's just a way for researchers to keep themselves busy and publish something?
     
    Last edited: May 8, 2020
    bobbler, Sid, Simone and 24 others like this.
  8. Invisible Woman

    Invisible Woman Senior Member (Voting Rights)

    Messages:
    10,280
    So the eminent Sir Simon is saying that not only is it a possibility for at least some viruses to enter the brain, this may pose increase risks of ongoing ill health.

    It seems reasonable to assume depending on the areas of the brain affected and the specific virus involved, different ongoing health problems might ensue.

    This suggests to me that this type of study is pretty much useless unless and until there is more knowledge about which viruses can affect the brain, what areas of the brain are affected and what symptoms and health problems are caused.
     
    Simone, Mithriel, pteropus and 3 others like this.
  9. rvallee

    rvallee Senior Member (Voting Rights)

    Messages:
    13,605
    Location:
    Canada
    The end. Poor questionnaires that ask leading questions make for bad research.

    The questionnaires are vague, non-specific and their very choice represent incorrigible bias, amplified by outcome-seeking analysis. Enough of this weak pseudoscience.
     
    bobbler, sea, JemPD and 6 others like this.
  10. spinoza577

    spinoza577 Senior Member (Voting Rights)

    Messages:
    455
    The abstract doesn´t describe the relationship between the infection and the disease. The reader already needs "to know" it.
    Logically speaking, a contribution is not only not necessary, but even unlikely, b/c if it would be a real contribution it would have caused the disease before the infection.

    Therefore also the psychosocial factors are unlikely to substantially contribute.

    They might be a predisposition though, but this is a completely different thing. E.g. you will not cure an existing cancer if you destroy any genetic predisposition. When the cancer has occured you need to cure the cancer itself.
    This is an important possibility, whatever it can mean in all concrete details in different diseases.

    So, already in terms of logic they failed, let alone any empirical terms.


    Is it finally the confession that they havn´t found anything after so much time? Or is it even the confession that they havn´t looked at their main subject? (Let me guess, it´s the last one.)
     
    Last edited: May 7, 2020
    Cheshire, Yessica, alktipping and 2 others like this.
  11. Invisible Woman

    Invisible Woman Senior Member (Voting Rights)

    Messages:
    10,280
    I agree with your post above @spinoza577, but

    this is dodgy ground, I think. Psychological and social aspects may make some difference -

    for example - if you live in poorly maintained housing that might have dampness, mould or infestation and can't afford to move somewhere better you might already be weakened and so that affects the severity and course of your illness. A socio-economic factor.

    If you were the lone carer for a vulnerable person, naturally you will feel the burden of that responsibility and so may well not take the same care of yourself while ill & thus affect the course of the illness. A psychological and socio-economic factor.

    It may be that someone who is wealthier and has choices will, when their symptoms improve, allow themselves a break or holiday. A bit like the idea of convalescence. Having this option gives both a psychological and physical (or bio) benefit.

    That's the sneaky thing with the BPSers. There is a grain of truth & some merit in looking at how biopsychosocial factors can affect illness. Like in the covid pandemic when it's worth investigating why some communities are more at risk than others. That's not what the BPSers do though - they simply patient blame and gaslight.
     
  12. rvallee

    rvallee Senior Member (Voting Rights)

    Messages:
    13,605
    Location:
    Canada
    Considering that the government adopted guidelines based on claims by some that knew so much about the role of individual psychological and psychosocial (why even separate those two?!) factors that all other efforts could be stopped and their model, based on affirmation of this knowledge they now admit does not exist, should be official practice, coercively used to deny medical and social support.

    Frankly this is basically an admission of fraud, of experts having made baseless assertions that lead to official policies impacting the lives of vulnerable sick people despite them having no such knowledge at any point and there being no actual evidence offered beyond those very weak claims.

    If there was any accountability in medicine this would be a massive scandal. But the lies perpetrated on us have made it so that nothing done to us matters, even explicit and well-documented harm. Based on those people's assurances that they had superior knowledge of our internal thoughts and cognitions, claims that are as absurd as it was pathetic for anyone to actually greenlight such obvious junk pseudoscience.

    This goes beyond incompetence and malpractice, it's malicious criminal neglect. I have little hopes for justice any time soon but it's so grating that everything that happened since was predicted and desperately warned against, only to be met with comments from the likes of the RACGP (or whatever is their acronym) that if it were up to them it would be 100% psychological and effectively left out of medicine entirely.
     
    bobbler, spinoza577, pteropus and 2 others like this.
  13. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,219
    Location:
    Aotearoa New Zealand
    Wow, so far it is as nasty as I thought it might be:
    It's sort of true. But it is pretty much implied that the reason poor people die earlier is because they have poor health behaviours, experience more psychological stressors and they don't get themselves to the doctor - it's their fault. Whereas someone with a genuinely holistic biopsychosocial view (in its best sense) might point to low incomes causing poor nutrition, time-poor families with both parents working long hours opting for cheap low quality meals from the abundant fast-food outlets in poor areas. They might point to the crowded, low quality housing, high pollution levels, substance abuse as a means to block out a difficult reality, and systemic prejudice affecting access to health care.

    When they say that, you can be pretty sure that they aren't suggesting an increase in the minimum wage.
     
    Last edited: May 8, 2020
  14. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,219
    Location:
    Aotearoa New Zealand
    I didn't know that about the Dubbo study - that's quite a potential selection bias.
     
  15. Hutan

    Hutan Moderator Staff Member

    Messages:
    29,219
    Location:
    Aotearoa New Zealand
    Ok, I think, after googling Hazard Ratios, I have understood this a bit. (But, be warned, I may not have.)

    Participants filled out a huge battery of questionnaires - there was a whole orchard of fruit to pick from so not surprising they found a small cherry. Notably, higher hypochondriasis didn't decrease the chances of reported recovery.

    Below is Table 3. It looks at the relationships between, among other things, parameters and Overall Severity (Table 2 deals with Baseline Severity). Have a look at the right hand column, titled Overall Severity. Three parameters were identified as predictors of overall severity at any point in time during the study:
    • Severity at intake (self-reported)
    • 'Neuroticism' as measured by the 12 yes/no subscale questions of the Eysenck Personality Questionnaire Short form, and
    • 'Locus of control behaviour' (that is, a feeling of individual control over events - low numbers suggest a feeling that you control things, high numbers suggest a feeling that external factors control things)
    Screen Shot 2020-05-08 at 1.04.45 PM.png
    So, now look at the Hazard Ratios for each of the three parameters in the 'Overall Severity' column. I think, from something in the paper, that a Hazard Ratio of 1 here means that a one-standard deviation change in a parameter results in no change to the average 'risk' of being recovered (in terms of overall severity) at a certain time.

    If your Severity at Intake is one standard deviation higher than the average, the 0.29 figure means that you only have 29% of the chance of being recovered that a participant with the average Severity at Intake had, at any point in time during the study. So, just for example, if a person with an average Severity at Intake had a 90% chance of being recovered at 12 months, the person with a Severity at Intake one SD higher would have a chance of just 26% of being recovered. That is major. A higher Severity at Intake also substantially lowered the risk of recovery in terms of fatigue, pain, mood and neurocognitive disturbance.

    A one standard deviation change in measured 'Neuroticism' results in a much smaller change in the chance of being recovered. The Hazard Ratio is 0.97. Obviously that is quite close to 1, so for all of the beat-up and discussion in the abstract, I don't think it is having much impact on the likelihood of being recovered. If your measured 'Neuroticism' score is one standard deviation higher than average, the 0.97 means that you have a 97% of the chance of being recovered that a participant with the average 'Neuroticism' score had, at any point in time. So, for example, if a person with an average Neuroticism score has a 90% chance of being recovered at 12 months, then a person with a neuroticism score one standard deviation higher would have an 87% chance of being recovered.

    One of the questions in the Neuroticism survey was 'I feel lonely - yes/no'. I'm prepared to believe that someone who is alone, or feels alone and unsupported, might take a little bit longer to report recovery. Maybe they have to get up and chop the wood in winter because no one else is going to do it and that leaves them in pain. Maybe no-one is making them nutritious meals. So sure, some of the things making up that neuroticism score might have a bit of an effect on recovery - but curing people of their so-called neuroticism is, if I'm understanding things right, no where even close to being a silver bullet in protecting people from ME/CFS (or PVFS).

    And they threw in that Locus of Control parameter in there. It has a Hazard ratio of 1.02, so is also very close to having no effect on the chance of recovery. There's probably something interesting to be said about that, but I'm out of steam, and - like I said - its impact is, I think close to zero.

    Very happy to have someone correct what I have written. But, if it holds up, there is nothing here. 'Neuroticism' as a factor in recovery is virtually irrelevant, as is everything except baseline severity. And the authors have not clearly reported their results.
     
    Last edited: May 8, 2020
  16. Hoopoe

    Hoopoe Senior Member (Voting Rights)

    Messages:
    5,417
    Thanks for looking more closely at this @Hutan.

    Maybe I can say it. Believing that you can control the situation has almost zero relationship to the actual outcome.

    This is in extreme contrast to the general belief of society that we patients are able to achieve a recovery if only we did this or that or tried harder.
     
    Last edited: May 8, 2020
    bobbler, Ariel, Simone and 10 others like this.
  17. Sean

    Sean Moderator Staff Member

    Messages:
    8,013
    Location:
    Australia
    Objective metrics for severity and recovery are required...

    They got that much right, at least.

    Nailed it.
     
    bobbler, sea, Hutan and 1 other person like this.
  18. Forbin

    Forbin Senior Member (Voting Rights)

    Messages:
    1,581
    Location:
    USA
    "Emotional lability" (mood swings) is one of the symptoms that Ramsey noted, so the illness could actually be causing symptoms that look like neuroticism. Add to that the fact that the symptoms actually cause a loss of function and control and I question whether rational patient concern over those consequences might not be wrongly attributed to neuroticism. As @Invisible Woman said, there's also no reason to invoke neuroticism to explain the natural anxiety brought on by the tangible economic and social jeopardy that the disease can put you in.

    The only thing I can think of as to why you would want to "control for time since onset" would be that they (hopefully) realize that prolonged illness could lead to a level of despair which could easily be confused for neuroticism. They should want to assess your psychological state as soon as possible after onset (since that's the earliest they'd see you), but, even then, it may be too late to distinguish some sort of psychological predisposition from the effects of the disease itself.

    The only sure way to study this would be to do a prospective psychological categorization of a huge cohort of physically well people and then see if those few who come down with a severe infection fare differently if they'd shown prior signs of neuroticism.

    Of course, the problem with going that route is that it would be... um... hard.
     
    Last edited: May 8, 2020
    Simone, janice, sea and 8 others like this.
  19. Invisible Woman

    Invisible Woman Senior Member (Voting Rights)

    Messages:
    10,280
    Neuroticism and the sense of loss of control is generally seen as a bad thing in terms of recovery. I can see some reasons why but isn't there also a flip side for some?

    For those feeling a loss of control complying strictly with a treatment regime, assuming it isn't making you worse like GET for ME patients, could give some a sense of control with the added motivation of regaining control of other aspects of your life.

    Those who are neurotic might also tend to pay much greater attention to the fine details of their treatment & less likely to make mistakes such as taking their meds at the wrong time.

    So if neuroticism & feeling a loss of control might cause issues in recovery for some, they also might increase the chance of recovery for others.

    So I can see why neuroticism & control might not make much difference overall to recovery as described by @Hutan's finding.
     
  20. rvallee

    rvallee Senior Member (Voting Rights)

    Messages:
    13,605
    Location:
    Canada
    Wow. I get the feeling that if we were to look in their desks, for at least a few of those people we will find a skull, calipers and books by Malthus and Charles Murray.

    The ugly personal politics are oozing, believers in the just-world fallacy. This has nothing to do with science, just people who feel so damn superior and want to "find" justifications for their natural superiorities among different populations.

    This is just about the worst perversion of what a competent BPS model should be, it's exactly and perfectly backwards. These people are nuts.
     
    Ariel and Amw66 like this.

Share This Page