Living with
unexplained
somatic symptoms
A combined qualitative and quantitative
study
of life-stories from rehabilitation
clients
Peter Elsass, D.Ms. cand.psych.
Bodil Jensen,
Rikke Mørup, cand.psych.
Marie Høgh Thøgersen, cand.psych
Centre for Humanistic Health Research, Institute of Psychology,
University of Copenhagen
&
Department of Social Medicine, Aalborg Hospital
______
Citation:
Elsass P., Jensen B.,
Morup R., & Thogersen MT (2006) Safe Minds – Living with
unexplained somatic symptoms:
A combined qualitative and quantitative study
of life-stories from rehabilitation
clients.
International Journal
of Psychosocial
Rehabilitation. 11 (1), 11-20
Author responsible for
correspondence:
Peter Elsass
Institute for Psychology
Landemaerket 9
1119 Copenhagen K
DENMARK
Abstract
Objective:
To
compare life-stories written
by rehabilitation patients with medically unspecific complaints
with life-stories written by other patients at the
same rehabilitation centre.
Methods: 136 stories were divided into four groups:
well-defined somatic complaints, unspecific complaints, a combination
of
well-defined and unspecific complaints and a group of exclusively
mental and
psychological complaints. Data-driven categories in the texts were
identified
and used for non-parametrical statistical tests.
Results: Overall, the
life-stories described a high amount of stress in both childhood and
adulthood.
But the patients with unspecific complaints differed from all the other
groups
in that they experienced a significant more stressful work-load. They were the most dissatisfied group and
revealed
the highest wish for pension.
Conclusion:
Although our study has many constraints it
support the hypothesis
that clients with medically unexplained and vague symptoms express a
view of
their daily lives as filled with more burdens and difficulties than do
other
clients in the same rehabilitation setting.
Keywords:
Chronic
fatigue syndrome; Illness
beliefs; Life-stories; Somatoform disorders.
Introduction
The
classification, conceptualization and treatment of patients with
“medically
unexplained (somatic) symptoms” are an ongoing debate. There are
different
positions in the debate where the patient is often depicted as a
marginal,
neglected wasteland between the walled citadels of medicine and
psychiatry (Editorial 2004). In a broader interdisciplinary context, they
present a problem for the integrity of medicine and psychiatry and are
therefore
often conveniently removed or exported to related, disciplines; from
medicine
to psychiatry and further to the social sciences. These
different positions do not exclude each other,
but rather
represent fundamental conflicts between the present professional conceptions of illness (Afari & Buchwald
2003). The scientific debates are, however, characterized by a
lack of
empirical evidence for the patient’s perspective. Although various
descriptions
of patient experiences exist in
the literature on e.g. somatoform disorders and so-called functional
illnesses,
these are usually presented in questionnaires and semi structured
interviews
aimed for clinical assessment. Few
have focused on the patients biographies (Bury 1982) or life stories
(Frank
1993, Williams 1994) and used an
observational or qualitative method of inquiry aimed to elicit and
describe the beliefs of the patients in their own terms (Clements et
al. 1997).
This article
contributes with a
phenomenological study that gives priority to the patient and his /her
subjective position guided by his own formulations. The aim is to
compare
life-stories written by rehabilitation patients with unspecific and
unexplained
medical complaints with the
life-stories written by other patients at the same institution with
well
defined medical complaints, in order to examine the content of their
“first
persons experience”.
The study is a
“natural” study,
where the population of a rehabilitation center, were required to write
down
their life-story as part of a training program.
METHODS
Population and life-stories
In the period 1998- 2002, 628 patients
completed a rehabilitation course at a Danish rehabilitation center (YDUN Rehabilitation Center, Department of Social Medicine, Aalborg Hospital. The clients have been referred with a broad
spectrum of medical and psychiatric diagnoses. During a three month
rehabilitation course, participants were requested to write down their
life-stories as one out of various course activities, and in the first
five
weeks of the course, time was set aside each day for writing these
stories. The
participants were free to develop the form and content of the stories
as they
wished.
In 2003 The Social-Medical Unit, Aalborg Hospital, contacted 610 participants and
asked if they were willing to participate in our research project by
anonymously making their life-stories accessible. 230 participants
approved and
contributed their life-story.
10 stories were disregarded due to
missing pages and lack of basic information, such as gender and age. To
make
the data comparable and suitable for a quantitative analysis 84 stories
were
disregarded because of stories of less than 2000 words, and stories
containing
less than 250 words under general
items of childhood or adulthood.
136 life-stories fulfilled the
criteria, and were considered suitable for the analysis. Of these 136
stories,
58 were written by men, and 78 by women, the average age was 40, and
the
age-range was from 21 to 61 years old. The average story was of 5022
words, the
shortest was 2550 words and the longest was 23750 words.
Quantitative and Qualitative analysis
A
traditional qualitative analysis has not been applied, rather a
systematic
reading and recording of units of meaning derived and grounded from the
stories. In 30 randomly chosen life-stories, units of
meaning in the text have been identified and categorized
guided by our interest in 1. general level of stress in childhood and
adulthood, 2. specific work-related stresses, 3. self perception and
causal
understanding of symptoms, 4. user satisfaction and requirement
of retirement for pension. A scoring manual was
developed on the basis of these data-driven categories. A
further 30 life-stories were scored with the manual, following
which the manual was corrected and completed. All 136 life-stories were
then
coded of the authors RM and MHT according to the final scoring manual
containing 105 individual categories.
Disagreement between the scoring was solved on weekly meetings
and
discussed with the research group.
The number of life-stories
mentioning the categories has
been estimated both as 1) a “dichotomized presentation;” (does a
life-story
mention a category or not), and 2) as a “quantitative presentation,”
(how many
times a life-story mentions one category). The dichotomized
presentation,
could, for instance, be the number of life-stories mentioning “fatigue”
one or
several times. The quantitative
presentation would be the number of life-stories that mention “fatigue”
once,
twice, three times, four times etc.
If the same result for these two ways of presenting the data is
seen, it
was concluded that the length of the life-stories does not have a
significant
influence on the result.
The results will be presented as
single categories in actual numbers and percentages and where
similarities was
found as general categories in median and interquartile range.
Non-parametric statistical analysis has been applied with the use of Kruskal
Wallis test to estimate the statistical significance.
RESULTS
COMPLAINTS
In the
life-stories ten categories of complaints emerged. In the following
these ten
complaints are dichotomized presented as
a percentage of the total amount of life-stories containing
the category of complaint, no matter how many
times the individual patient has mentioned the
category.
62
% General pain
56
% Back-pain
40 % Fatigue
38
% Anxiety
25 % Depression
23 % Drug Abuse
16 % Psychiatric diagnostic
symptoms
16 % ”Worn down”
8 % Suicide attempts
56 % Other mental problems
Types of
complaints
It was feasible
to put the individual
categories together in the
following four groups:
- Well-defined compliants:
Complaints are mentioned in
relation to clear somatic etiology, such as back pain
after a car accident or due to medical illness. 19% (18 female, 8 male)
- A combination of well-defined and
unspecific complaints: The
complaints refer to a medical diagnosis combined with additional
complaints of unexplained medical symptoms. 22% (21 female, 9 male)
- Unspecific complaints: The compliants
are vauge and medically unspecific, and mentioned in relation
to fatigue, stress and burn-out. The complaints can be confined such as
back-pain or headache, however they are mentioned without relation to a
specficic etiology.
- 43%
(52 female, 6 male)Mental/
psychological complaints: Exclusively
mental and psychological complaints
such as anxiety and depression. 16% (15 female,
7 male).
The total
population has a predominance of
females over males, for the group with exclusively vague and
unspecific
complaints the gender difference is significantly
different from the other groups (p< 0.045).
GENERAL LIFE
STRESSES
Stress
categories
In the
life-stories there are 45 categories
for experiences of stress. In the following dichotomized presentation
the 15
most frequently mentioned categories illustrates
the total number of life-stories containing this category.
Childhood:
59 % other problems
with parents
46 % other undefined burdens
37 % high level of responsibility
and workload
31 % experiences of being bullying
29 % learning difficulties
27 % loneliness
26 % parent abuse
21 % physically ill parent
21 % economic problems
18 % parent divorce
16 % parent with drug abuse
14 % loss of one or both parents
13 % mentally ill parent
12 % sexual abuse
11 % placement outside home
Adulthood:
59
% high level of workload and stress
51 % unemployment, 18% more than
three years
50 % divorce, one or more times
44 % being fired from job
43 % economic problems
40 % loss of parent or sibling
31 % drug abusing spouse
27 % loneliness
26 % abusive spouse
14 % loss of spouse
9 % mentally ill spouse
9 % children placed outside home
8 % physically ill spouse
6 % bullied at work
5 % sexual abuse outside marriage
General categories of
stress
The individual
categories have been combined in three general categories as follows:
1: Stressful experiences in childhood: consisting of 11
categories concerning family relations, 4 categories concerning social
problems
and 6 categories concerning other problems.
2: Stressful experiences in adulthood: consisting of 15
categories concerning family relations, 4 categories concerning social
relations, and 4 categories concerning other problems.
3: Stressful experiences in both childhood and
adulthood: consisting of all of the 45 categories concerning stressful
expereinces.
The life-stories are characterized by many
descriptions of stressful experiences. In total, 89% describe stressful
experiences in childhood, 39% mention at least four categories and in
most
cases stressful experiences such as incest, suicide attempts, drug
abuse, and
abuse. 94% describe stressful experiences in adulthood. 29% mention at
least
four of these categories.
Tabel 1.
Stress and type of
complaint
The number of times a
patient mentions one or more of
the categories in the three general stress categories in the
median and interquartile range.
|
|
1.
Well-defined complaints
(N = 26) |
2. Combination of
well-defined and vague complaints
(N= 30) |
3. Unspecific complaints
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
| 1: Stressful experiences
in childhood |
3
(0.75 - 11.25) |
7
(4.00 - 10.25) |
8
(4.00 - 13.25) |
8.5
(5.75 - 14.00)
|
| 2: Stressful experiences
in adulthood |
7
(3.00 - 16.25) |
6.5
(4.75 - 15.00) |
10
(7.00 - 17.00) |
9
(4.75 - 15.25) |
| 3: Sum of Stressful
experiences |
12
(4.00 - 29.50) |
17
(7.00 - 22.50) |
19
(11.75 - 28.25) |
19.5
(13.00 - 28.50) |
Many stressful
experiences are mentioned by
all four groups of patients. Although the groups with
unspecific complaints and with mental
/psychological complaints, do not differ significantly from each other,
compared to the two other groups, their mention of stressful
experiences is
significantly higher, particularly in childhood.
Overall comparisons
were
made between the four groups using the Kruskal Wallis test. Where the
overall
effect was found to be significant all pairwise comparisons between the
four
groups were examined, also using the Kruskal Wallis test
Dichotomized
estimates: General
category 1: groups 1,2,3 and 4 (p < 0.01 ), 1 and 3 (p < 0.01), 1
and 4 (p < 0.001). General category
3: groups1 and 3 (p < 0.02 ).
Quantity estimates: General
category 1: groups 1,2,3 and 4 (p < 0.02), 1 and 3 (p < 0.01), 1
and 2 (p
< 0.05) 1 and 4 (p < 0.01). General category 2: groups 2 and 3 (p
<
0.05). General category 3: groups 1 and 3 (p < 0.03), 1 and 4 (p
< 0.05)
WORK-LOAD
70%
of the
patients mention that they have been stressed by having
had a lot of responsibility and a heavy workload. The
three categories; high level of responsibility in childhood, heavy
work-burden
in adulthood, and an experience of being burnt
out, is combined in a general category called “heavy work-load”
Tabel 2.
Work-load
and type of complaint
The number
of times one or more of the categories are mentioned, is illustrated in
median
and interquartile range:
|
|
1. Well-defined
complaints
(N = 26) |
2. Combination of
well-defined and vague complaints
(N= 30)
|
3.
Unspecific complaints
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
| Heavy work
load |
1.5
(0.00 - 4.00) |
1.5
(0.00 – 4.25) |
2.0
(1.00 - 4.00) |
1.0
(0.00 - 3.00) |
In the above general
category for heavy workload, the patient groups have been examined for
all
combinations. 68% of those patients who describe themselves as being “burnt-out” belong in the group of
unspecific complaints. The patients with unspecific complaints differ
from the
other groups by scoring higher on the heavy workload general category,
examined
both by dichotomized numbers, and by quantitative
numbers. The group with mental / psychological complaints, differs from the other groups by
mentioning a significant lower number of categories than the patients
with
unspecific complaints.
Dichotomized
estimate: General category 4: groups 1,2,3
and 4 (p < 0.03), 2 and 3 (p <
0.02), 3 and 4 (p < 0.02).
Quantity estimate: Groups
3 and 4 (p < 0.03).
CAUSAL
UNDERSTANDING AND COHERENT LIFE-STORY
Causal understanding
was defined as those points in
the life-stories where two different categories are connected to one
another.
In all four groups:
40% correlate two psychological
categories
19%
correlate experience of stress with actual mental condition
16% correlate physical
complaints with actual mental condition
These three categories for correlation have
been combined in a general category, labelled “causal understanding.”
Tabel 3.
Causal understanding
and type of complaint
The number of times
the patients mention one or more of the categories for causal
understanding is
given by the median and interquartile
ranges for the general category and by the
amount and percentages for the individual categories.
| |
1.
Well-defined complaints
(N = 26) |
2. Combination of well-defined and vague complaints
(N= 30)
|
3.
Unspecific complaints
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
| Causal
understanding |
1
(0 – 3) |
2
(0 – 3) |
1
(0 – 3) |
1
(0 – 3) |
| Correlation
between stressful experience and actual physical condition |
4
15%
|
4
13%
|
11
19%
|
8
36%
|
| Correlation
between physical complaint and actual mental condition |
4
15% |
4
13% |
11
19% |
3
13% |
| Other types
of psychological insight
|
13
50% |
8
26% |
22
37% |
14
64%
|
| Wish for
psychological help
|
9
35% |
8
27% |
14
24% |
10
46%
|
The
patient group with mental/psychological complaints (4), reveal the
highest
incidence of causal understanding while the groups 2 and 3, with
unspecific
complaints, reveal the lowest
Dichotomized
estimate:
Groups 1, 2, 3 and 4 (p < 0.05), 3 and 4 (p < 0.04), 2 and 4 (p
<
0.008)
The form
of the life-stories was examined. If coherence between the individual
categories was found in more than half of the life-story, the story has
been
categorized as coherent.
Tabel 4.
Coherence
of life-story and type of complaint
Amount and
percentage of patients that reveal a coherent life-story:
|
|
1.
Well-defined complaints
(N = 26) |
2. Combination of
well-defined and vague complaints
(N= 30)
|
3.
Unspecific complaints
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
| Coherent life-story
|
9
34% |
7
23% |
9
15% |
12
55% |
The group with
unspecific complaints, reveal the least coherent, and most fragmented
life-stories, whereas the group with mental problems have the most
coherent
life-stories.
Dichotomized estimate: groups
1,2,3 and 4 (p < 0.005 ), 3 and 4 (p < 0.001), 2 and 4 (p <
0.04)
SELF PERCEPTION
Tabel 5.
Negative self
perception and type of complaint
Amount shown as a percentage that
reveals a negative self-perception
| |
1.
Well-defined complaints
(N = 26) |
2. Combination of
well-defined and vague complaints
(N= 30)
|
3.
Unspecific complaints
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
Negative
Self- perception
|
7
27% |
16
53% |
32
55% |
17
77%
|
53 % of all life-stories contain a category with a
negative
self-perception. The group with specific complaints presents the least
negative
self- perception, whereas the group with mental / psychological
problems presents
the most negative self-perception. The group with unspecific complaints
has a
more negative self- perception than the group with well-defined
complaints.
Dichotomized estimate:
groups 1,2,3 and 4 (p < 0.006), 1 and 3 (p < 0.02), 1 and 2 (p
<
0.05), 1 and 4 (p < 0.001)
USER SATISFACTION AND
REQUIREMENT FOR RETIREMENT AND PENSION
59 % of the patients
have described their dissatisfaction with sentences like: ”not very
helpful”,
”did not work at all” in relation to the following categories:
General Practitioner 14%
Health
System 36%
Psychologist 5%
Social services 33%
Work
3%
Friends and Family 6%
Rehabilitation centre 16%, fellow
participants 1%
Tabel 6.
Dissatisfaction and
type of complaint
These nine categories
have been brought together in a general category, labelled “patient dissatisfaction”. The
number of times the patients mention one
or more of the categories is in the median and interquartile
range.
| |
1. Well-defined
complaints.
(N = 26) |
2. Combination of
well-defined and vague complaints
(N= 30)
|
3.
Unspecific complaints: The new disease
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
Patient
Dissatisfaction |
0.0
(0.00 – 1.00) |
1.5
(0.00 - 2.00) |
1.0
(1.00 - 2.00) |
1.5
(0.00 - 2.00) |
Patient groups 2 and
3, are significantly more dissatisfied than the other two groups.
Dichotomized estimate: groups
1, 2 and 3 (p < 0.03), 1 and 2 (p < 0.01)
Tabel 7.
Requirement of retirement for
pension and
type of complaint
Numbers as a percentage
that express a wish for early retirement.
| |
1.
Well-defined complaints.
(N = 26) |
2. Combination of
well-defined and vague complaints
(N= 30)
|
3.
Unspecific complaints
(N=58) |
4.
Mental / Psychological complaints
(N=22) |
Wish
for pension |
7
27% |
13
43% |
34
59% |
4
18% |
56% would like retirement
for pension.
Among the group with mental complaints, the wish for pension/early retirement is the lowest, and
among the group with unspecific complaints the wish is the highest.
Dichotomized estimate:
Groups 3 and 4 (p < 0.01), 1 and 4 (p < 0.04)
DISCUSSION
Resume
Overall, the
life-stories from the participants at the Rehabilitation Centre were
characterized by experiences of stress in both childhood and adulthood.
The groups with unspecific complaints and
with mental/psychological complaints did not differ significantly from
each
other as regard to amount of perceived stress. Compared to the two
other groups, however, they mentioned
significantly
more stressful experiences, particularly in childhood. Concerning
stressful
work related experiences, the patients with unspecific complaints
differed from
the other groups in that they experienced a significant higher level of
work-load.
The group with unspecific complaints
wrote the must incoherent life-stories, whereas the group with mental /
psychological problems, wrote the most coherent stories. The group with
unspecific complaints revealed a more negative self-perception compared
with
the group with well-defined complaints,
however, not as negative as the group with mental /
psychological
problems. The latter was the
least likely to ask for retirement and
pension. The group with unspecific complaints revealed the
highest wish
for pension and was the most
dissatisfied group.
These results should be taken with a grain of
salt. The material is not necessarily representative for the big and
varied
group of patients with “unexplained medical complaints”. We have named
our
study a “natural study” because it reflects the situation on a
rehabilitation
centre with patients of many illnesses. Our intention was to give a
more
detailed picture of the patients own perception of their life and to
examine if
they themselves perceive the etiology of their illness.
The quantitative and
qualitative
analysis of these
self-presentations is not compared with the patients’
“objective” medical and social situation, due
to phenomenological position to convey and recognize the patient
experience as
valid in itself.
Although
the life stories are “first persons” self-presentations they do reflect
the context
in which they are written. A story always has a sender and a receiver
even it
is hidden in an internal dialogue of the writer. And although the life
stories were
written for the patients itself and not intended to be read by others,
they did
have the potential of being discussed with the professionals at the
rehabilitation
centre and therefore did have a message.
A traditional qualitative analysis
has not been applied as e.g. described in the method of grounded
theories, but
has consisted of a systematic reading and recording of units
of meaning with the intention to examine hypothesis
already given. The large number of life stories gave us an upportunity
to
combine the method of ”inductive narrative analyses” with a
quantitative
method. Even it is not consistent with evidence based medicine the
combination
of qualitative and quantitative methods are recommended suited to test
medical
hypothesis (Pope & Mays 1995).
With these reservations our study do give an empirical
supplement to the
battleground between professionals and laypersons viewpoints on the
understanding of the clientele of unexplained medical complaints.
To be a patient at a
rehabilitation
centre is to be situated in a context where the recognition of the
validity of
the symptoms is a more or less hidden agenda for them. Attributions
about the
causes of an illness or its symptoms are important in determining a
patient’s
response to the illness (Sensky et al. 1996). Patients with functional
illnesses as chronic fatigue syndrome often attribute their illness to
physical
causes and minimize psychological or personal contributions (Powell et
al.
1990, Schweitzer et al. 1993, Butler et al. 2001). Our study supports
that patients
situated at a rehabilitation centre do give stress and
workload high priority in their perception of
apprehension of their illness rather than physical and medical causes.
We hope that this might be
recognized in the debate where patients with unexplained medical
symptoms do
have the not very respectable concept of “new age disease” and where
the focus
sometimes are more on minor characteristics of somatoforme patients as
e.g.
their dissatisfaction, low self esteem and wish for pension and
retirement
rather on their burdensome general life experiences.
Acknowledgement:
We thank
dr.
Finn Jacobsen for initiation the study and dr. Tom Teasedale for
helping us
with the statistical analysis.
References
1. Afari N, Buchwald D. Chronic
fatigue syndrome. A review. Am. J. Psychiatry 2003; 160; 221– 236.
2. Bury M. Chronic illness as biographical description.
Sociology of Health & Illness 1982; 4; 167–182.
3. Butler JA, Chalder T, Wessely S: Causal attributions for
somatic sensations in patients with chronic fatigue syndrome and their
partners. Psychol Med 2001; 31:97–105
4. Clements A, Sharpe M, Simkin S, Borill J, Hawton K. Chronic
fatigue syndrome: A qualitiative investigation of patients’ beliefs
about the illness. J Psychosom Res 1997; 42; 615-624.
5. Editorial. J Psychosom Research 2004; 56:
6. Frank A. The rhetoric of self-change: Illness experience as
narrative. The Sociological Quaterly 1993; 34; 39–52.
7. Pope C, Mays N. Reaching the parts other methods cannot
reach: an introduction to qualitative methods in health and health
services research. BMJ 1995; 311; 42-45.
8. Powell R, Dolan R, Wessely S: Attributions and self-esteem in
depression and chronic fatigue syndrome. J Psychosom Res 1990;
34:665–673
9. Sensky T, MacLeod AK, Rigby MF: Causal attributions about
common somatic sensations among frequent general practice attenders.
Psychol Med 1996; 26:641–646
10. Schweitzer R, Robertson DL, Kelly B, Whiting J: Illness
behavior of patients with chronic fatigue syndrome. J Psychosom Res
1993; 38:41–49
11. Williams G. The genesis of chronic illness: Narrative
reconstruction. Sociology of Health & Illness 1994; 6; 175 - 2000
