• Welcome to Phoenix Rising!

    Created in 2008, Phoenix Rising is the largest and oldest forum dedicated to furthering the understanding of and finding treatments for complex chronic illnesses such as chronic fatigue syndrome (ME/CFS), fibromyalgia (FM), long COVID, postural orthostatic tachycardia syndrome (POTS), mast cell activation syndrome (MCAS), and allied diseases.

    To become a member, simply click the Register button at the top right.

Cortisol output in adolescents with chronic fatigue syndrome: Pilot study on the comparison with hea


Senior Member
This is from King's College. Now they're claiming that CBT can normalize hypocortisolim.

This study examined cortisol in adolescents with chronic fatigue syndrome (CFS) compared to healthy adolescents and changes in cortisol after cognitive behavioural guided self-help treatment. Exploratory analyses investigated the association between cortisol output and psychological variables.

Salivary cortisol was measured upon awakening, at 15, 30, 45 and 60 mins afterwards and at 12 noon, 4pm and 8pm, in adolescents with CFS and healthy controls (HC). Groups were matched for age, gender, menarche status, menstrual cycle and awakening time. Twenty-four adolescents with CFS provided saliva samples six months after treatment. The main outcome measure was total salivary output over the day, calculated by area under the curve (AUC). The salivary awakening response was also assessed.

Cortisol output over the day was significantly lower in the CFS group (n = 46) than in healthy controls (n = 33). Within the CFS group, lower daily cortisol output was associated with higher self-reported perfectionist striving and prosocial behaviour. There were no significant group differences in the awakening response (n = 47 CFS versus n = 34 HC). After treatment, adolescents with CFS (n = 21) showed a significant increase in daily cortisol output, up to normal levels.

The reduced daily cortisol output in adolescents with CFS is in line with adult findings. Associations between reduced cortisol output and two psychological variables - perfectionism and prosocial behaviour - are consistent with cognitive behavioural models of chronic fatigue syndrome. The mild hypocortisolism is reversible; cortisol output had returned to healthy adolescent levels by six months after cognitive behavioural guided self-help treatment.



Senior Member
I think these results can be explained by the lack of a treatment control group, and these patients having mild fatigue that resolves on its own after a while. CBT is once again given credit for people getting better on their own.


Senior Member
The sample sizes are quite small and they don't quote values for associations and significance so hard to conclude anything from the abstract. What is worrying about the result is they start with a CFS group of 46 or perhaps 47 but quote a significant increase on a group of 21.

The main thing it made me wonder is how much cortisol levels vary naturally and hence could there results be a result of natural variation or say additional stress of going back to a hospital for blood tests where you have had sessions saying 'you are not ill - you must push yourself'.


The problem in the UK will be that our poor children will receive 'treatment' resulting from this pseudoscience.

Snow Leopard

South Australia
My cortisol would go up too if forced to sit through six months of psychobabble.

My old hypothesis of the observed hypocortisolism is that most patients actually suffer less day to day stressors than the average person and thus normalised to a lower level of cortisol and isn't really related to the disease at all.

Of course my hypothesis wasn't very popular here or elsewhere for that matter...


Senior Member
It is well known when you get more active your cortisol will be higher, especially when you have lack of energy. This doesn't say anything about the illness. When they meassure the cortisol level after the CBT it will be low again.


Places I'd rather be.
My old hypothesis of the observed hypocortisolism is that most patients actually suffer less day to day stressors than the average person and thus normalised to a lower level of cortisol and isn't really related to the disease at all.

Of course my hypothesis wasn't very popular here or elsewhere for that matter...

I think there may be some merit to that hypothesis, depending on how 'stressor' is defined. There are many biological systems involved in exertion and activity. Obviously some are affected severely enough to cause disability. Others may not be affected at all and suffer from the usual consequences of disuse (regardless of the reasons for disuse).
Last edited:


Senior Member
More than 50% were lost:

Participants were 49 adolescents with CFS, and 36 healthy adolescents
who were selected to match for age and gender. The CFS group
was composed of patients attending a specialist CFS Unit in London
who met CDC [12] and Oxford [36] criteria for CFS. Assessment at the
CFS clinic included the exclusion of depression or other psychiatric
problems as the primary diagnosis. They had all been assessed by paediatricians who
had conducted appropriate tests to exclude other possible
diagnoses causing their fatigue. Self-reported mean duration of CFSwas
25.3 months (SD= 16.1). Saliva samples were taken prior to treatment.
The healthy adolescents were recruited via local schools and were required
to have no history of CFS and no current medical condition likely
to cause excessive fatigue. Of the CFS sample, 24 returned posttreatment
questionnaires and cortisol samples; of the remainder,
three did not complete treatment and the others did not return usable
post-treatment cortisol samples or questionnaires or both. Independent
t-tests indicated that there were no significant differences between participants
for whom post-treatment questionnaire and cortisol data were
or were not available, in terms of age, duration of CFS, baseline fatigue
(Chalder fatigue scale; [4]) or any of the pre-treatment main cortisol
measures (ts b 1.9).


Senior Member
Eight correlations were looked at for "area under the curve day" - only two were significant.

Associations between daily cortisol output (AUCday) and other characteristics
As the above results had indicated AUCday to be significantly reduced in the CFS
group, exploratory investigations using Spearman's correlations were conducted to examine
whether this cortisol measure was associated with clinical characteristics within this
group. There was a significant association between AUCday and Self-Oriented Perfectionist
Striving (Spearman's rho (rs) = −0.298, p = 0.050) and SDQ Prosocial Behaviour
(rs = −0.425, p = 0.003). AUCday was not significantly associated with Socially
Prescribed Perfectionism (rs = 0.08), Self-Oriented Perfectionism Critical (rs = −0.10),
SDQ Difficulties (rs = 1.0), Chalder Fatigue Scale (rs= 0.06), Birleson Depression Inventory
(rs = 0.05) or Spence Children's Anxiety Scale (rs = −0.10). There was no association
between AUCday and duration of CFS (rs = −0.09) or gender differences in
AUCday (Z =0.85, p = 0.468). For the healthy control group, there were no significant
correlations between AUCday and questionnaire scores; the largest value for rs was 0.29
(p = 0.120) for the association for Self-Oriented Perfectionist Striving, which is in
the opposite direction than in the CFS group.


Senior Member
Main results:

Change in clinical outcomes after treatment

There was a significant improvement in school attendance, the primary clinical outcome,
from 24% to 49% (Z(24)=−2.5, p= 0.012).

The reduction on the Chalder Fatigue
Scale (mean=24 (SD=5) to mean=21 (SD=21)) did not reach significance (t(23)=
1.5, p = 0.14).

There was a significant reduction on the Spence Children's Anxiety Scale
(mean=22 (SD=17) to mean=17 (SD14); t(21)=2.1, p= 0.005).

The Birleson Depression
Scale (excluding Q7 about energy) also showed a significant decrease (mean=
11 (SD = 6) to mean = 8 (SD= 5); t(23) = 2.1, p = 0.044).

There were no significant
changes on the SDQ scales (Zs b 2.0).

Info on SDQ:
Strengths and Difficulties Questionnaire (SDQ; [13]), which provides subscale scores for Difficulties (emotional and behavioural) and Prosocial Behaviour (e.g., “I try to be nice to other people”). This latter scale was used as a way of assessing high standards for personal conduct


Senior Member
Associations between changes in cortisol output and clinical measures
Exploratory correlational analyses were undertaken to investigate possible associations
between change in AUCday and clinical variables. There were no significant associations
between changes in AUCday and changes between pre-treatment and 6-month
follow-up on percentage school attendance, Chalder Fatigue Scale, Birleson Depression
Inventory, Spence Children's Anxiety Scale, SDQ Difficulties or SDQ Prosocial scales
(all correlation coefficients b 0.36).
With no improvement on the Chalder Fatigue Scale, it sounds like the return to school is an arbitrary part of their "therapy", rather than an actual sign of improvement. I doubt they'll be able to maintain that, and also doubt that they're getting any value out of school while they're still that ill.


Senior Member
This is the first study to report that adolescents with CFS have significantly
lower saliva cortisol secretion over the course of the day than age
and gender-matched healthy controls. This is in line with research in
adults with CFS [16]. Lower daily cortisol output was associated with
two psychological characteristics in the CFS participants: perfectionist
striving and prosocial behaviour (kind, considerate, helpful and sharing
behaviours). As far as the authors are aware, this is the first demonstration
of associations between daily cortisol output and these characteristics,
in young people or adults, in relation to CFS or other populations. In
adults in the general population, there is some evidence that perfectionism
may be associated with higher cortisol responses to acute stressors
in the laboratory (e.g., [43]), but it is possible that chronic stress may
lead to a longer-term blunting of the cortisol response [23].

Stress increases cortisol generally but we'll claim low stress indicates stress in people with CFS. :confused:

In the
present study, the association between mild hypocortisolism and
perfectionism or prosocial behaviour may reflect a chronic stress response
related to attempts to keep up high standards of performance
or personal conduct, in line with cognitive behavioural approaches to
CFS (e.g., [20,40]).
I'm not sure that the CBT model predicts prosocial behaviours i.e. "kind, considerate, helpful and sharing behaviours".

Not sure it predicts perfectionist behaviours either.

Also, the perfectionist scale scores didn't change during the trial suggesting CBT doesn't influence such scores.
Last edited: