Functional Incapacity in CFS

Psychopathology
        Vol. 41, #6, pp. 339-345
 http://content.karger.com/produktedb/produkte.asp?typ=fulltext&file=000152375

Functional incapacity and physical and psychological symptoms: how they
interconnect in Chronic Fatigue Syndrome
-----------------------------------------------------------------------
Stefan Priebe(a,*), Walid K.H. Fakhoury(a), Peter Henningsen(b)
a Unit for Social and Community Psychiatry, Barts' and the London School of
  Medicine, Queen Mary University of London, London, UK;
b Department of Psychosomatic Medicine, University of Munich, Munich, Germany
* Prof. Stefan Priebe, Unit for Social and Community Psychiatry, Newham Centre
  for Mental Health London E13 8SP (UK), Tel. +44 20 7540 4210, Fax +44 20
  7540 2976, E-Mail s.priebe@qmul.ac.uk  

Abstract

Background
It has been argued that perceived functional incapacity might be a primary
characteristic of chronic fatigue syndrome (CFS) and could be explained by
physical symptoms. If so, it could be expected to be closely associated with
physical, but not psychological symptoms. The study tests this hypothesis.

Sampling and Methods:
The sample consisted of 73 patients, with a diagnosis of CFS according to the
Oxford criteria, randomly selected from clinics in the Departments of
Immunology and Psychiatry at St. Bartholomew's Hospital, London. The degree
of fatigue experienced by patients was assessed using the Chalder Fatigue
Questionnaire and a visual analogue scale. Self-rated instruments were used
to measure physical and social functioning, quality of life, and physical and
psychological symptoms.

Results
Principal-component analysis of all scale scores revealed 2 distinct
components, explaining 53% of the total variance. One component was
characterized by psychological symptoms and generic quality of life
indicators, whilst the other component was made up of physical symptoms,
social and physical functioning and indicators of fatigue.

Conclusions
The findings suggest that perceived functional incapacity is a primary characteristic of CFS, which is manifested and/or explained by physical symptoms.

Introduction

Patients suffering from chronic fatigue syndrome (CFS) often report, along
with fatigue, symptoms such as pains in joints and muscles, headaches, sore
throat, lymph node tenderness, cognitive difficulties and sleep problems [1,
2]. CFS is commonly associated with anxiety and depression [3, 4] and usually
follows a chronic and fluctuating course, with an essentially unknown
aetiology.

Studies measuring different aspects of quality of life and functioning [5, 6]
have demonstrated that quality of life is severely disrupted in CFS patients.
In the vast majority of these patients, the functional impairment is severe
and long lasting, with rates of recovery of only 3-19% [6, 7]. Low social and
physical functioning may reflect a perceived incapacity that is central to
the experience and symptomatology of CFS patients. The question is whether or
not this incapacity is a consequence of other symptoms or a primary and
defining feature of CFS [8, 9]. It has indeed been argued that a perceived
incapacity in fulfilling social roles - or the anticipation of such an
incapacity - may be a primary characteristic of CFS, and the symptoms usually
reported might have a function to justify this perceived incapacity [3]. Such
a justification could be understood in the context of prevailing, socially
and culturally determined illness paradigms, and also of the struggle to
recognize CFS as a legitimate disease. However, it is possible that the
illness itself mediates functional incapacity primarily through physical
symptoms. If it were a secondary phenomenon and a consequence of other
symptoms, it could be assumed that the level of perceived incapacity would be
positively associated with the severity of all psychological and physical
symptoms of CFS. As a primary feature, however, it may neither be socially
acceptable nor fit into existing medical classifications. Patients might
explain or articulate their incapacity in terms of more conventional
symptoms, responding in part to prevailing social or medical norms. As
social and physical incapacity appears closer to physical than psychological
symptoms, and physical symptoms may be regarded as socially less stigmatizing
than psychological impairment, one may hypothesize that functional incapacity
should be associated with physical, but not necessarily psychological
symptoms.

The results of Hardt et al. [10] are consistent with this assumption. They
assessed the internal factor pattern of health-related incapacity in CFS
patients from three countries with the General Health Survey (Medical Outcome
Survey) Short Form 36 (SF-36) and found an 'unusual profile'. In the most
common 2-component structure of the SF-36 found in different patient
groups, social functioning and vitality load together with the emotional
subscales on a 'mental' component, whilst physical role fulfilment and
functioning, pain and general health perception load on a 'physical'
component. Social functioning and vitality, a measure of non-fatigue, were
associated with the physical components; emotional well-being and emotional
role functioning were the only subscales that loaded on a different factor
than the other 6 [10].

To our knowledge, there have been no studies so far that try to unravel the
structural relations between different aspects of functional incapacity and
various physical and psychological symptoms in CFS patients. We studied a
sample of secondary/tertiary-care patients with CFS to address the following
question: how are indicators of perceived functional incapacity associated
with physical and psychological symptoms? We hypothesized that the level of
perceived incapacity would be associated with the severity of physical rather
than psychological symptoms.


Methods

Sample
A random sample of patients, with a diagnosis of CFS attending clinics in the
Departments of Immunology and Psychiatry at St. Bartholomew's Hospital,
London, were approached to take part in the study. Patients attending these
clinics were selected using random number tables so that between 2 and 4
patients were approached each week, a number chosen to give a manageable
number of subjects for the study every week. Thirty-seven of 146 patients
approached in the immunology clinic (response rate 25.3%) and 36 of 50
patients (response rate 72%) approached in the psychiatry clinic participated
in the study (response rate 37.2%, i.e. 73/196). Members of the research team
interviewed the patients to obtain informed consent and to ensure they met
Oxford criteria for CFS [1]. Patients were excluded if they were unable to
complete the questionnaires for reasons of either language or severe
disability. Data were collected from medical notes and using a questionnaire
designed specifically for the study.

The patients from the two clinics did not statistically significantly differ
in terms of their sociodemographic characteristics (age, ethnicity,
sexuality, marital status, previous education, employment, number of
children, having close friends or receipt of benefits). They did not
statistically significantly differ in their specific symptoms, disability,
quality of life, psychological distress and previous attendance to mental
health professionals either. Therefore, there was no reason for not treating
them as one sample in the analyses. The comparisons of patient
characteristics between the two settings and the sample characteristics are
described in detail in White et al. [11] and Rakib et al. [12].


Measures
The degree of fatigue experienced by patients was assessed using the Chalder
Fatigue Questionnaire [13] and a visual analogue scale [14]. The Chalder
Fatigue Questionnaire (11-item scale) assesses the degree of physical and
mental fatigue in the last month. The respondent rates each item on a
categorical 0, 0, 1, 1 point scale.

Several instruments were used to measure physical and psychological symptoms
and aspects of functioning and quality of life. The Symptom Checklist 90
Revised (SCL-90-R) [15] has 90 items of which 83 are grouped into 9 primary
symptom dimensions: somatization, obsessive-compulsive behaviour,
interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety,
paranoid ideation and psychoticism. The remaining 7 items are used in the
calculation of global scores. The Spielberger Trait Anxiety Questionnaire
[16] has 20 items that measure trait anxiety (how one generally feels). Items
are rated on a 4-point Likert rating scale, ranging from 1 (almost never) to
4 (almost always). Scores on the Spielberger Trait Anxiety Questionnaire
range from a minimum of 20 to a maximum of 80. The Health Anxiety
Questionnaire [17] is a 21-item scale that measures worry and preoccupation
about health, fear of illness and death, reassurance-seeking behaviour and
interference with life. Items are rated on a 4-point Likert scale, ranging
from 1 (not at all/rarely) to 4 (most of the time), with scores therefore
varying between 21 and 84. The Somatic Discomfort Questionnaire (SDQ) [18]
has 50 items that assess how bothering symptoms have been in the past week.
Each symptom is rated on a 4-point Likert scale, ranging from 1 (not at all
bothersome) to 4 (a lot bothersome). Scores on the SDQ range from 50 to 200.
The Beck Hopelessness Scale [19] has 20 true-false items that measure 3 major
aspects of hopelessness: feelings about the future, loss of motivation and
expectations. Items are rated 1 if true and 2 if false, with scores ranging
from a minimum of 20 to a maximum of 40.

The Medical Outcome Study SF-36 [20] was used as a self-report questionnaire
to assess perceived incapacity based on physical functioning, social
functioning, pain, mental health, role limitations due to physical problems,
role limitations due to emotional problems, vitality and general health
perception. The social and physical functioning subscales were regarded as
indicators of functioning and incapacity in this study.

The Manchester Short Assessment of Quality of Life (MANSA) [21] was used to
assess generic quality of life. The MANSA contains 16 questions of which 4
are termed objective and the rest are ratings of satisfaction with life as a
whole and the following life domains: leisure activities, financial
situation, living situation, personal safety, social relations, mental health
and family relationship. As far as psychometric properties are concerned,
results of the MANSA have been found to have high concordance with results of
the much longer Lancashire Quality of Life Profile [22]. The mean score of
satisfaction ratings was used as an indicator of generic subjective quality
of life.


Analysis
The data were analysed using SPSS for Windows version 11. The hypothesis was
tested in a multivariate analysis to reduce the probability of type I error
and consider all different symptoms and aspects of incapacity in one
analysis. For this, principal-component analysis (PCA) was used. The
Kaiser-Meyer-Olkin measure of sampling adequacy was conducted to ensure there
was adequate and high variability in the data (values greater than 0.5 for
the Kaiser-Meyer-Olkin measure indicate that the data could be subjected to
PCA). Bartlett's test of sphericity was conducted to ensure that the
different instruments were sufficiently correlated. PCA with rotation of
component(s) was conducted, and several rotations were tried in order to
identify the solution that would provide meaningful and distinct components.
Data on the rotated solutions are only presented for the best solution. As
the analysis is exploratory, the best solution was chosen based on the
eigenvalues (explained variances extracted by the factors) and the
meaningfulness of the components extracted. Only cases with valid values on
all variables subjected to PCA were included in the analysis.


Results

Sum Scores of Various Subjective Measures
Table 1 shows the sum scores of the administered instruments.


Appropriateness of the Data
The Kaiser-Meyer-Olkin measure was 0.84, suggesting sufficient and high
variability in the data to conduct PCA. The results of Bartlett's test of
sphericity (approxi- mate chi^2=1,169.6, d.f.=276, p<0.0001) suggested
that the scales were sufficiently correlated to perform PCA.

Tables 2 and 3 show the correlations between the sum scores of various
subjective measures.


Component Solution
When all the scales were subjected to PCA, the analysis resulted in a
solution with 5 components with eigenvalues greater than 1. Component 1 had
an eigenvalue of 9.9 (41.3% of the variance). Component 2 had an eigenvalue
of 2.8 (11.7% of the variance). Components 3, 4 and 5 had eigenvalues of 1.5,
1.1 and 1.0, respectively (15.4% of the variance). However, most of the
scales loaded heavily on components 1 and 2. None of the scales had their
highest loading in either component 3 or 5. The Chalder Fatigue Scale was the
only scale that loaded the highest on component 4 as compared to its loading
on the other components. The other chronic fatigue measure, the Fatigue
Visual Analogue Scale, loaded the highest on component 1. Given that
components 3, 4 and 5 had much lower eigenvalues than components 1 and 2, and
had almost no high loading scales in them, it was decided to choose a
2-component solution.

Table 4 shows the loadings of the various items in the rotated 2-component
solution.

The quartimax rotation was found to provide the most meaningful solution. In
this solution, component 1 comprised 14 variables and had an eigenvalue of
7.9 (32.9% of the variance). It had positive loadings of the SCL-90-R
subscales depression, anxiety, obsessive-compulsive symptoms, psychoticism,
hostility, phobic anxiety, interpersonal sensitivity, paranoid ideation, the
Spielberger Trait Anxiety Questionnaire, Health Anxiety Questionnaire and
Beck Hopelessness Scale, and negative loadings of the SF-36 subscales of
mental health and emotional role fulfilling. Component 2 consisted of 10
scales and had an eigenvalue of 4.8 (20.1% of the variance). Component 2 was
made up of the following scales with positive loadings: the SF-36 subscales
of social functioning, general health, vitality, role ­ physical, and
physical functioning. The Fatigue Visual Analogue Scale, SCL-90-R subscale
somatization, SDQ and SF-36 subscale bodily pain loaded negatively on this
component. The Chalder Fatigue Questionnaire also loaded negatively on this
component, but the loading was less than 0.5. The first component captures
psychological aspects, the second one physical phenomena. The total variance
explained by the 2 components was 53.0%.


Discussion

This study investigated the association between self-ratings of incapacity
and quality of life and various physical and psychological symptoms in CFS
patients. Our study is different from other studies in this area [23, 24]
because it focused on symptom patterns, while taking into account data on
quality of life and aspects of functioning. The PCA yielded a satisfactory
2-component solution that explained 53% of the variance. The 2 components
reflect a physical and a psychological side of CFS, and reveal that generic
subjective quality of life and functional incapacity are not homogeneously
associated with symptoms of CFS.

The perceived incapacity in fulfilling social and physical roles may be best
captured by the subscales of the SF-36 on social and physical functioning.
The scores on these subscales are associated with vitality and inversely with
one of the defining symptoms of CFS, i.e. fatigue (Chalder Fatigue Scale,
Fatigue Visual Analogue Scale). They CFQ are also associated with other
physical symptoms (SDQ, SCL-90-R subscale 'somatization'), but not with
psychological symptoms such as depression (Beck Hopelessness Scale, SCL-90-R
subscale 'depression') and anxiety (Spielberger Trait Anxiety Questionnaire,
SCL-90-R subscale 'anxiety'). These psychological symptoms are linked to a
generic measure of quality of life (MANSA), reflecting satisfaction with life
in general and life domains, and to emotional role functioning and mental
health (SF-36, subscale).

These findings demonstrate that there are clinically meaningful patterns to
be found beyond the general statement that functioning and quality of life
are severely disrupted in CFS patients. Perceived incapacity, not only on the
physical, but also on the social level of functioning, is particularly
associated with fatigue [25], vitality and other physical symptoms, whilst
deficits in generic quality of life have a strong association with
psychological symptoms. The latter association between generic subjective
quality of life and general psychopathological symptoms, particularly mood
symptoms, is in line with results from other studies.

The identified pattern is consistent with the hypothesis that a perceived
incapacity in social and physical functioning is linked to the severity of
physical rather than psychological symptoms [3]. It also corresponds exactly
with the distribution of subscales of the SF-36 between the 2 identified
components, the 'unusual profile' that Hardt et al. [10] described without
interpreting it further.

One interpretation of our findings rests on the assumption that attributive
processes of the patients influence the pattern of reported symptoms and
their association with both functional incapacity as well as generic quality
of life. There is some evidence supporting this assumption. For instance,
Moss-Morris and Petrie [8] examined cognitive behavioural styles that
differentiated depressive disorders and CFS among 73 patients. For different
levels of depressive symptoms, but equivalent levels of energy loss and
disability, they found a highly significant preponderance of psychological
attributions in the pure depressed and of physical attributions in the pure
CFS group. They conclude that CFS and depression can be distinguished by
unique cognitive styles characteristic of each condition, and CFS is a
disorder notable for the patients' emphasis of an organic causal attribution.

The results suggest that CFS patients perceive their general quality of life
more in tune with their current mood than their fatigue, whereas their
functioning is more related to physical symptoms and fatigue, as the defining
symptom of the illness. Thus, functional incapacity might be understood as a
primary characteristic of CFS that is manifested and possibly explained by
physical symptoms. Cultural and social factors can evidently influence the
expression of illness by symptoms. Alternatively however, the nature of this
illness could itself cause functional incapacity primarily through physical
symptoms, with psychological symptoms having a limited impact on incapacity,
although influencing quality of life; this would negate one of our stated
assumptions. If so, this may relate to the nature of the distinctive
experience underlying the use of the word fatigue in CFS. The direction of
causality and the role of attributions have not been assessed in this study.
Clinically, one may conclude that fatigue and physical symptoms need
attention to reduce perceived incapacity and to improve functioning, whilst
psychological and mood symptoms may require treatment to influence subjective
quality of life.

Our study has several methodological limitations. We used PCA which is based
on correlations between variables found in a cross-sectional assessment. The
findings do therefore not allow a conclusion on causal relationships. Also,
the findings of the multivariate analysis depend on the selection and
specificity of the scales that were chosen to capture functional incapacity
and different types of symptoms. Most of the scales we used had been
originally developed and validated for patient groups with disorders other
than CFS. Although we regarded the instruments used as the most appropriate
ones for the study, one might question their validity for the specific
purpose. Moreover, the selection of psychological and physical variables
tested for their associations with key features of CFS and with functioning
and quality of life may be seen as unbalanced, because the majority of
physical symptoms that were assessed was measured on only 2 scales, the
somatization subscale of the SF-36 and the SDQ. However, both subscales
comprise items on physical symptoms like pain and muscular soreness that have
been suggested to be specific for CFS on the one hand and correlate closely
with other non-pain physical symptoms on the other. Thus, it may be argued
that the sum scores of physical complaints used in our study reliably reflect
single scores for more specific symptoms.

In further research, qualitative methods and longitudinal designs may be
required toassess the patients' experiences and their views of symptoms,
functional incapacity and quality of life in more depth and to understand
better the components and also the underlying mechanisms of illness and of
attribution. It should also be possible to identify factors that influence
these over time, with or without interventions.

To conclude, this paper has focused on a significant issue relating to
investigating whether perceived functional incapacity is linked to physical
or psychological symptoms in CFS. The uniqueness of the analyses stems from
the fact that they used a comprehensive set of criteria to unravel the
structural relations between different aspects of functional incapacity and
various physical and psychological symptoms in CFS patients. Our finding that
there are distinct 'physical' and 'psychological' factors inherent in CFS is
new - and specific - to the CFS literature. Although patients were recruited
from two different settings, there were no statistically significant
differences in their sociodemographic, clinical and service use
characteristics, thus reducing significantly the possibility of bias that
could have resulted from combining the two in one analysis. However, the data
were collected in secondary/tertiary care, and the results may not be
generalizable to patients in primary care and community settings. Finally,
the design was cross-sectional; therefore, the correlations do not indicate
causal relationships.


Acknowledgments

The authors would like to thank Prof. Peter White, Prof. Anthony Pinching and
Dr. Barbara Hedge for their contribution to the design, organization and
management of the study, and Dr. Abdur Rakib for his contribution to data
collection.