Around one in five new consultations in primary
care are by patients with physical symptoms for which no specific
organic cause is found (medically unexplained symptoms). While many
of these symptoms are transient, over one-third persist and can
cause distress and disability. Here, we discuss strategies for the
management of adults with medically unexplained symptoms.
The approach to managing patients with physical
symptoms with no identifiable cause is complicated by unhelpful and
often controversial terminology and disease concepts. For many
years, patients with unexplained symptoms were said to have a
psychosomatic illness. However, this term is now considered to be
potentially misleading, as it implies that the symptoms necessarily
have a purely psychogenic origin.
An alternative term, “somatisation”, has been
coined to describe a putative process by which some people
experience and communicate psychological distress as physical
symptoms, but this description presumes that psychological problems
are being avoided and that the physical symptoms are “all in the
mind”. To avoid the artificial separation of the mind and body,
terms such as “medically unexplained symptoms” and “functional
somatic symptoms” have been used. We prefer to use the former term
because, unlike “functional somatic symptoms”, it makes no
assumptions about the cause of the symptoms.
Clusters of chronic medically unexplained
symptoms can appear to form symptom syndromes, and so are sometimes
given diagnostic labels such as irritable bowel syndrome, chronic
fatigue syndrome or fibromyalgia, depending on the symptoms and the
medical specialty to which the patient is referred. The results of a
recent observational study suggest that substantial overlap in
symptomatology exists across apparently diverse symptom syndromes.
Such illness can be chronic, recurrent and cause
disability and, for the practitioner, take up time and resources.
When managing patients with unexplained symptoms, an integrated
approach that acknowledges the possible, but often unproven,
contributing roles of biological, psychological and social factors
is required. It is important for doctors to recognise the reality of
patient’s symptoms and, where possible, to provide explanations for
them.
Antidepressants can help with unexplained
symptoms such as chronic pain and poor sleep, even in the absence of
depressive illness. Reattibution training can help doctors to make
the patient feel understood, to explore psychological and social
factors, and to make appropriate and sensible links between symptoms
and any psychological or social problems. Such training may result
in fewer referrals outside the primary care team and appears to be
cost-effective. Cognitive-behavioural therapy can be helpful,
particularly in reducing physical symptoms.
One of main pleasures of medicine is in reaching
a difficult diagnosis. Doctors hope that they approach each case
with no pre-conceptions, gather all the positive signs and symptoms
together without trying to force them into a diagnostic category,
and then attempt a synthesis, which may end up ignoring the odd and
borderline.
This classic approach to clinical reasoning was
challenged by empirical studies performed by experimental
psychologists in the 1970s. This is not how clinicians' mind work -
they do not collect ALL data
first, and then analyse them as impartially as possible.
Hypotheses are continually being exluded or
confirmed as data are collected; it has been estimated that a first
hypothesis is generated within 28 seconds after hearing the
patient's chief complaint and then on average 7 ± 2 hypotheses are
active at any one time.
With time they develop sets of rules (heuristics)
to help us deal with clinical problems in a more effective way. This
"rule-based" reasoning and "skill-based reasoning" (that relates to
patterns of thought and actions that are governed by stored models
of pre-programmed instructions, largely unconsciousness, and
acquired over years) are more cost-effective problem solving
techniques than "knowledge-based" cognition.
It appears that the main difference between the
novice and experienced clinicians lies in the capability of the
latter to move from
knowledge-based to skill-based reasoning. Experts have a much larger
repertoire of skill-based schemata and problem-solving rules than
novices.
The problem is, however, that exclusive use of heuristics may
lead to (diagnostic and treatment) errors. The problem then becomes
to identify those rules that are effective and accurate. One
approach would be that each
specialty identify common rules in its field, and then further to
scrutinize and develop those rules that are least prone to errors
Back to top
The concerns about the cardiac operative outcomes
in Bristol infants could not have been brought about without the
rigorous approach of the Cardio-thoracic surgeons.
They suggest that variation in cardiac surgical
outcomes may be due to
-
random variation
-
appropriateness of treatment
-
unmeasured severity of illness or
-
varying standards of care.
The aim of risk stratification is to measure
severity of illness and relate this to outcome.
The surgeons have developed a set of calculations
known as the modified Parsonnet system. This has been developed on
4,931 non valve surgery patients in the patients from the selected
group between 1993 and 1995 inclusive. The standard Parsonnet system
allows and encourages the inclusion of subjective additions which
tend to increase predicted mortality. The calculations below were
made without these subjective additions.
Most existing scoring systems have been derived
from logistic regression, which is a multivariate statistical
technique that simultaneously estimates the contribution of each
possible variable to the risk of mortality. Handling missing data is
a problem, both in the development and use of such system.
A simpler method is the independent Bayes
technique. Based on tables relating outcomes to single risk factors,
the probability of an adverse outcome can be estimated for a patient
with any combination of risk factors.
The method is based on the repeated use of a
basic formula in probability theory first developed by the Reverend
Thomas Bayes, a non-conformist minister from Tunbridge Wells, which
was published posthumously in 1763. This theorem tells us how the
probability of an event should be revised when additional relevant
evidence is obtained
Back to top
