Coping strategies in chronic fatigue syndrome: outcomes over time

[Dolphin]

Coping strategies in chronic fatigue syndrome: outcomes over time

Susan K. Johnson1,*, Virginia Gil-Rivas1, Karen B. Schmaling2

Article first published online: 30 MAY 2008

DOI: 10.1002/smi.1185

Stress and Health
Volume 24, Issue 4, pages 305312, October 2008

Keywords:chronic fatigue syndrome;coping;neuroticism;prognosis

Abstract*

This paper examined the contribution of coping strategies and neuroticism to longitudinal outcomes in a tertiary care sample of participants with chronic fatigue syndrome (CFS).

Coping strategies and neuroticism were assessed among 93 participants with CFS, and used to predict participants' illness-related outcomes.

Coping by instrumental support was the only strategy associated with clinical outcomes over time: more reliance on instrumental support predicted less vitality over time after accounting for the level of vitality at the index visit.

Neuroticism and coping strategies were not associated with fatigue symptoms, physical functioning or role functioning over time.

How to Cite
Johnson, S. K., Gil-Rivas, V. and Schmaling, K. B. (2008), Coping strategies in chronic fatigue syndrome: outcomes over time. Stress and Health, 24: 305312. doi: 10.1002/smi.1185

* I gave each line its own paragraph

 

[Dolphin]

I think they haven't adjusted for multiple comparisons.

It looks to me that they looked at 64 comparisons and found one that was significant at the threshold of p<0.05. However, unusual things can happen by chance if you look enough times e.g. toss enough coins 10 times, you may find one set where you get 10 heads or 10 tails but you can't be sure that that particular coin is biased - it could simply be due to chance.

They looked at 15 of the 16 coping styles measured by COPE:

Coping. A measure of coping strategies Coping
Operations Preference Enquiry (COPE) (Carver,
Schier, & Weintraub, 1989) was completed at the
index visit. The COPE is a multidimensional 60-
item inventory that yields 15 factors that refl ect
active and avoidant coping strategies. Ratings of
trait-like tendencies to use each strategy are made
on a four-point scale, ranging from I (usually)
dont do this at all (1) to I (usually) do this a lot
(4). The measure has good psychometric properties
with alphas ranging from 0.45 to .92, test
retest reliabilities ranging from 0.46 to 0.86, and
strong evidence of discriminant and convergent
validity. The COPE scales are: active coping, planning,
seeking instrumental social support, seeking
emotional social support, suppression of competing
activities, religion, positive reinterpretation
and growth, restraint coping, resignation/acceptance,
focus on and venting of emotions, denial,
mental disengagement, behavioural disengagement,
alcohol/drug use, and humour. The alcohol/
drug use scale was not used in the analyses based
on clinical evidence suggesting that patients with
CFS are extremely sensitive to the effects of alcohol
and drugs, leading to abstinence.

So one can see that people would generally use more than one method.

and neuroticism:

Neuroticism. The revised NEO-Five Factor
Inventory (FFI) (Costa & McCrae, 1992) was
completed by all participants at the index visit.
This self-report instrument consists of 60 statements,
scored on a scale from strongly disagree
(0) to strongly agree (4). The NEO-FFI measures
the following personality domains: neuroticism,
extroversion, openness to experience, agreeableness
and conscientiousness; only the neuroticism
score was retained for analysis.

They checked whether it had an effect on for outcome measures:
CFS symptom severity (9 symptoms incl. fatigue)
and three of the SF-36 subscales: physical functioning, vitality, and physical role functioning.

Out of these 64 possible cases where one coping strategy/personality trait (neuroticism) could be associated with an outcome measure, there was only one case it was:

more reliance on instrumental support predicted less vitality over time after accounting for the level of vitality at the index visit.

 

[Dolphin]

Discussion section:

Discussion

Seeking instrumental social support (asking for
assistance, information or advice) was a signifi -
cant predictor of worse outcomes as refl ected in
decreased vitality levels among patients with CFS
over an 18-month prospective assessment period,
after adjusting for the contributions of factors
previously identifi ed to be associated with poorer
outcomes. These results are similar to Afari et al.
(2000), who also found social support was higher
in fatigued compared with non-fatigued twins.
Our hypotheses regarding the association between
neuroticisms and avoidant coping strategies and
CFS symptoms were not supported. Specifi cally,
neuroticism did not signifi cantly predict greater
CFS symptomatology or functional status after
accounting for baseline scores and other important
covariates.

The percentage of patients meeting criteria for
a current depressive disorder in the present study
was very low relative to previous studies, but may
be partly attributable to the coding assumptions
applied to the DIS in this study. Previous studies
have not adjusted for the symptom overlap
between psychiatric disorders and CFS symptoms,
potentially overestimating the prevalence of psychiatric
disorders among this population (Henningsen,
Zimmerman, & Sarrel, 2003; Johnson
et al., 1999; Skapinakas, Lewis, & Meltzer, 2000;
Tiersky et al., 2001). If patients seeking services
from a tertiary care centre specializing in CFS are
aware of the research regarding CFS and psychiatric
co-morbidity, they may have attempted to
present themselves in a psychologically positive
manner.

The use of instrumental social support, which
includes seeking information, advice or assistance,
was negatively associated with vitality. On
fi rst blush, this result may seem counterintuitive,
but studies have found membership in a self-help
group to be a robust predictor of poor treatment
response to a graded exercise trial (Bentall, Powell,
Nye, & Edwards, 2002) and poor outcomes over
time (Sharpe et al., 1992). Involvement in selfhelp
groups may represent illness conviction or
attachment to the illness, but self-help groups also
represent an instrumental social support strategy,
and thus, our result is consistent with the fi nding
that too much information/advice seeking may be
counterproductive in CFS. An alternative interpretation
is that those who are more seriously
ill and need more assistance are more likely to
seek it. (they could have expanded on this more - it's buried here in this one sentence)

In summary, the findings in the present study
support a limited role for personality and coping
factors in CFS outcome over time. The tertiary
care clinic sample in this study was mostly unemployed,
had low levels of psychopathology, was
largely white and female, which may have yielded
different and less generalizable results from more
diverse community samples. We did not identify
coping strategies that were associated with better
outcomes over time. However, the converse of the
predictor of poorer outcomes in this study may
suggest an approach to improve outcomes in CFS,
such as reducing information and advice seeking
that may promote a narrow focus on the disease.
Our results are consistent with advocating graded
activity, relaxation training, scheduling of pleasant
activities, and attributional change (Williams,
2003) and positive interpretation of illness
(Hyland, Sodergren, & Lewith, 2006) (I think this is stretching it a lot - they virtually found nothing). As has
been reported previously, only a minority of participants
improved or recovered from CFS, leaving
the challenge of identifying potentially modifi able
predictors of the course of illness in CFS open to
future investigations.

I'd be suspicious of these psychologists in future. It is my impression that some psychologists can be sympathetic and see patients as human beings facing challenges with possible ways to improve things (e.g. Leonard Jason and his team); while others seem to see patients more as generally dysfunctional.
----
The paper doesn't mention at all the literature that people can use different coping strategies and different times within the illness e.g. the first few months after diagnosis might be different from years later. Leonard Jason and his team have looked at that in some papers e.g. on the Fennell scale.

 

[illsince1977]

I wish I could rationally discuss this, but I just can't get past my anger. I have the utmost respect for anyone, like you, Dolphin, who has the wherewithal to actually read these presumably scientific papers in detail. Hats off to you! Kudos! :victory:

I understand that I'm not serving our cause by my avoidance and reflexive condemnation of this nonsense, and that no reasoned rebuttal to this narrow-minded, uncompassionate, wrong-headed view of illness is possible without acquainting myself with the literature, but I'm just not up to it!

Sorry for the rant.

 

[Snow Leopard]

I don't see how reducing social support will improve outcomes.

It would have been nice if they would actually do other tests on the data to explore this. For example, were those who sought social support and didn't improve more likely to have had lower baseline SF-36 scores, or longer illness duration for example.

Or in terms if information-seeking behaviour, perhaps this is associated with patients who are more realistic/less biased in the way they answer self report questionnaires (eg less blind optimism). As a consequence, hypotheses about information seeking behaviour and outcomes need to be tested against more robust measures, such as employment status or actometer measurements at long term followups and controlled for in terms of baseline measurements.

 

[5150]

people who "seek social support" are those who are ill to the point of their daily functioning being affected; that's generally when someone looks for help, when their lives take a nosedive due to the progressive, cascade of symptoms. And yes, when we are so ill as to seek social support, such as by using on-line forums, the outcome is worse than than those who aren't sick enough to yet look for help. Why? because _there is no cure_ and we only get worse... notwithstanding the occasional case of reported temporary improvement(although at some point these are followed by relapse).

but what the heck is so revealing about this? it's only common sense! for psych-babblers to present this as significant is twisted. they want to keep their piece of the $$ action.

my blood pressure goes up re this kind of study. grrrrr

 

[Enid]

Grrrrr - too - 5150 !.

 

[WillowJ]

Seeking instrumental social support (asking for
assistance, information or advice) was a signifi -
cant predictor of worse outcomes as refl ected in
decreased vitality levels among patients with CFS
over an 18-month prospective assessment period,
after adjusting for the contributions of factors
previously identifi ed to be associated with poorer
outcomes.

Only after they made statistical adjustments, did they find anything significant (that they cared to report)... that's what they are saying here, yes?

And what factors? Attribution to a physical cause? "Somatization" (or worse degree of ME/CFS-type symptoms, such as weak limbs and difficulty breathing)? It could hardly be membership in a support group since that is what they are testing.

So, they negate the differences between those who say, "I have a physical disease, probably related to a virus--it all started with a case of the flu that just wouldn't go away" and those who say, "I have a psychiatric-classified disease" and those who say "I have an emotional problem or burnout".

They negate the differences between those who say "I have weak legs and am unable to do things" and those who say "I feel sleepy, tired, or not motivated".

Only then do they try to figure out whether seeking assistance is related to worse outcomes.

Confounding factors, anyone? :b

 

[Dolphin]

Here's how they defined somatic attribution:

Attribution for CFS. Participants rated two
items from Wessely and Powells (1989) five-item
attribution scale (My illness is a physical one,
My illness is psychological in nature) on a 1
(not at all) to five-point (extremely so) scale,
the latter item was reverse keyed, and the
responses were added. For the resulting score
(range = 210), higher responses indicated more
endorsement of somatic attributions and less
endorsement of psychological attributions.

 

[WillowJ]

so somatization and attribution to a physical cause is one and the same in this particular study. Thanks for the info.

Still, they do statistical gymnastics to negate the difference between those who say, "I have a physical disease with physical symptoms" (such as weak legs, difficulty breathing) and those who say, "I have a psychiatric-classified disease" or "I have an emotional problem or burnout".

Then and only then, do they test coping mechanisms.

The differences they have controlled for, are confounding factors.

 

[Snow Leopard]

so somatization and attribution to a physical cause is one and the same in this particular study.

No, somatization was not discussed in this study at all. That measurement came under the title "Attribution for CFS."

Here is a quote from the "Statistical Analysis" section:

The analyses were designed to address the contribution of neuroticism and coping strategies to CFS symptoms and functioning (i.e. physical functioning, physical role functioning, and vitality) over time after adjusting for baseline levels of functioning and potential predictors of these outcomes reported previously (i.e. age, sex, somatic attribution). The models were built by examining the contribution of the key variables of interest and better established predictors, and include only those variables that contributed signifi cantly (p < 0.05) to the outcome. All of the coefficients presented are unstandardized. Wald ?2 statistics are presented to indicate the amount of variance in the outcome explained by the models.

 

[Sean]

I don't see how reducing social support will improve outcomes.

Yes, I would love to see robust evidence that reducing social support (ie increasing social adversity) improves general outcomes, in any group of people.