Clinical Formulation

Copyright © 1998 Elsevier Science Ltd. All rights reserved.

6.01
Clinical Formulation
GILLIAN BUTLER
University of Oxford, Warneford Hospital, UK
6.01.1 INTRODUCTION

1

6.01.2 DEFINITIONS: WHAT IS A FORMULATION?

2

6.01.2.1
6.01.2.2
6.01.2.3
6.01.2.4
6.01.2.5
6.01.2.6

Main Principles
Formulation and Diagnosis: Assumptions
Formulation and Diagnosis: Controversial Issues
The Difference Between a Formulation and a Model
Types of Formulation
Levels of Formulation

2
4
5
6
7
7

6.01.3 PURPOSES: WHAT A FORMULATION IS FOR
6.01.3.1
6.01.3.2
6.01.3.3
6.01.3.4
6.01.3.5
6.01.3.6

8

Understanding: The Overall Picture or Map
Prioritizing Issues and Problems
Planning and Selecting Intervention Strategies
Predicting Responses and Difficulties
Determining Criteria for Successful Outcome
Thinking About Lack of Progress

9
9
10
10
11
11

6.01.4 METHODS: HOW TO CONSTRUCT A FORMULATION

12

6.01.4.1
6.01.4.2
6.01.4.3
6.01.4.4
6.01.4.5

Sources of Information
Putting the Information Together
Key Factors and Basic Elements
Issue of Completeness
Conceptualizing Processes of Change

12
14
17
18
19

6.01.5 ACCURACY: HOW TO TELL IF A FORMULATION IS RIGHT
6.01.5.1 Criteria of Accuracy
6.01.5.2 Questions for Research

20
20
20

6.01.6 USING THE FORMULATION: PRACTICAL ISSUES
6.01.6.1 The Value of Organizing and Clarifying
6.01.6.2 Developing an Internal Supervisor
6.01.6.3 Communicating a Formulation

21
21
21
22

6.01.7 CONCLUDING DISCUSSION

22

6.01.8 REFERENCES

23

thereby enabling the patient to regain his morale.
(Frank, 1986)

6.01.1 INTRODUCTION
Patients come to psychotherapy because they are
demoralized by the menacing meanings of their
symptoms. The psychotherapist collaborates with
the patient in formulating a plausible story that
makes the meanings of the symptoms more benign
and provides procedures for combatting them,

Although not all therapists would be happy
with the idea that they are ªformulating a
plausible story,º the process of clinical formulation remains the lynch pin that holds theory and
practice together. This is agreed by proponents
1

2

Clinical Formulation

of most major therapeutic traditions: for example, behavior therapy (Turkat & Maisto, 1985;
Wolpe & Turkat, 1985), psychodynamic therapy
(Barber & Crits-Christoph, 1993; Perry, Cooper,
& Michels, 1987; Silberschatz, Fretter, & Curtis,
1986), family therapy (Minuchin, 1974), cognitive therapy (Freeman, 1992; Persons, 1989,
1993), cognitive analytic therapy (Ryle, 1978,
1990), and interpersonal therapy (Klerman,
Weissman, Rounsaville, & Chevron, 1984).
The attempt to construct and use a clinical
formulation is central to the work of therapy.
Various methods for systematizing the processes
involved have recently been proposed (Horowitz, 1989; Luborsky & Crits-Christoph, 1990)
and, thinking specifically about the issues involved in psychotherapy integration, Goldfried
(1995) has put forward a case for developing a
common language for case formulation that is
independent of theoretical orientation. Personal
discussions of many kinds may be more or less
valued and helpful to someone experiencing a
difficulty, including the informal advice traded
between friends, but one of the major differences
between informal discussions and responsible
clinical practice is that they do not make use of
the process of formulation. The attempt to
formulate a case, so as to apply an appropriately
chosen method of intervention in the light of a
particular theory, is one of the activities that
makes therapists, as opposed to friends, accountable for their practice.
This chapter discusses issues concerning
clinical formulation that are relevant to therapists from different theoretical backgrounds.
However, the illustrations of the general points
made will largely be drawn from the author's
own experience and will therefore reflect the
author's original cognitive-behavioral training,
together with a more recent interest in exploring
possibilities for integration between different
kinds of psychotherapy.
6.01.2 DEFINITIONS: WHAT IS A
FORMULATION?
6.01.2.1 Main Principles
A formulation is the tool used by clinicians to
relate theory to practice. Clinicians use theoretical as well as practical knowledge to guide
their thinking about the problems and difficulties presented by the people who come to them
for help, and this combination of ideas helps
them decide how best to help those people.
However, although the theories are relatively
simple and clearÐadmittedly to varying
degreesÐthe information brought to treatment,
and gathered during the process of assessment, is
always complex and often unclear. The process

of marrying theory and practice is therefore
fraught with difficulty. As well as having
different reasons for requesting psychological
help, people vary in their ability to describe or
name their difficulties, in their histories and
relationships with their families, friends and
colleagues, in their ability to relate to a therapist,
degree of psychological-mindedness, and emotional expressiveness. As well as having different
theories, training, and clinical experience, therapists vary in the ways in which they understand,
communicate with, and relate to their patients.
Therapists bring with them to therapy specific
skills, expertise, and information, and also their
individual personalities and inclinations. The
process of formulation is influenced by all these
disparate factors, and this makes learning how to
formulate a case with the necessary objectivity,
clarity, and attention to the individual to guide a
successful treatment one of the most fascinating,
rewarding, and difficult tasks faced by clinicians.
The assumption that many clinicians of
different orientations probably share about
the psychological difficulties of others is this:
at some level it all makes sense. Even though our
understanding of the processes involved, and
particularly of their inter-relationships, is incomplete, this assumption was given a simple,
and relatively uncontroversial, diagrammatic
form by Padesky and Mooney (1990). The
difficulties that people describe to their therapists have four inter-related aspects (cognitive,
affective, behavioral, and physiological), and
change in any one of these variables affects all of
the others, as shown by the bidirectional arrows
in Figure 1. So, taking anxiolytic medication can
make one feel calmer, think about problems
more constructively, and do some of the things
that previously seemed too difficult or overwhelming. Feeling more cheerful can lighten
ones step, help one to feel more optimistic, and
relate more productively to others. Changing
ones perspectiveÐor way of thinkingÐcan
provide the sort of new outlook that helps to
dissipate distress, reduce tension, and encourage
constructive activity, and so on. The four ways in
which aspects of psychological life are conventionally categorized reflect the internal workings
and psychological state of a person at a
particular point in time. This person is at the
same time relating to the external world through
a personal social, political, and historical
context. The factors that determine this context,
and fashioned it to be the way that it now is, are
not easy for psychological therapists to know
about: hypotheses for explaining and understanding the way they interact with each of the
four types of phenomena have been made. The
overall configuration is the source of the
narrative, or story, that a person brings to

Definitions: What is a Formulation?
therapy. If we understood the rules governing
the relationships between all these factors we
would, no doubt, be better able to help our
patients.
The business of therapy, to a large extent,
involves intervening to facilitate change in (at
least) one of the four main aspects of psychological life shown in Figure 1, and different
kinds of therapy attend differently to these
different aspects, entering the process of change
through different gateways. The intention,
however, is much the sameÐto help people
solve the problem or problems that they bring to
therapy. Pharmacological and traditional behavioral therapies provide perhaps the clearest
examples as the methods that they use, and the
formulations upon which these methods are
based, can be isolated relatively easily. Cognitive therapies, which adopt both cognitive and
behavioral methods, operate on at least two
levels. They may concentrate on identifying and
reexamining particular thoughts, thereby changing feelings and behavior, and/or they may
focus on underlying meanings and beliefs and
adopt more sophisticated and complex methods
of intervention, often related to those used in
more dynamic and experiential traditions.
Experiential therapies make specialized use of
the medium provided by the feelings and
thoughts arising in the present context of
therapy, and work with these to facilitate a
dynamic process of change. In order to do this,
it becomes essential to think about, and to
formulate, what happens in the relationship

The
environment:
personal,social,
historical
context,
etc.

3

between the two people involved in therapyÐ
methods which were originally described and
understood by proponents of the various
psychodynamic schools of therapy. Interpersonal therapy and systems therapy also formulate
problems in terms of relationships between the
person requesting help and others around them,
and use this understanding to help people
change as they wish. All of these methods
initiate the process of change in different ways,
determined by the way in which they understand, or formulate, the problem presented, and
it is this understanding that determines what
therapists doÐwhat steps they take to alleviate
the problem.
The point is that the way in which a
formulation is constructed will be influenced
by the point at which a therapist enters, and
attempts to influence, this dynamic relationship
between these main aspects of psychological
life. Some general points are important:
(i) each aspect influences all of the others, so
none of the therapies has the exclusive aim of
changing one factor. Rather, by focusing the
process of change in one place, the aim is to
bring about the change that the patient
desiresÐusually to ªfeel better,º in all the
relevant respects.
(ii) The main medium of therapy is
languageÐwhat one person says or suggests,
to another. To this extent, the cognitive, implicational context within which therapies take
place provides the basis for the way in which the
presenting problems will be formulated.

Cognition

Affect

Physiology

Behavior

Figure 1

Inter-relationships between aspects of functioning (Padesky & Mooney, 1990).

4

Clinical Formulation

(iii) Understanding of other people, and
hence the ability accurately to formulate their
problems, develops within the context of the
relationship between them, mediated by factors
such as trust and acceptance as well as by
language.
(iv) Our understanding of the ways in which
the aspects of psychological life are integrated is
partial. At this point in time, psychology is an
imperfect but developing science. The implication of this is that formulations for the purpose
of therapy have to be speculative.
Formulations can best be understood as
hypotheses to be tested, and the most obvious,
if not the most logical, test of a formulation will
be the response to the selected interventions.
This is not to say that an expected change
following a specific intervention proves that the
formulation on which it was based is accurate.
Unfortunately, the reasons why change occurs
are far more complex and difficult to discern
than this. However, the formulation used in this
way is perhaps the main tool that the therapist
has from which to draw such conclusions in the
individual case. Thus, working in an openminded way with a formulation provides a
means of contributing as a therapist to the
scientific endeavor involved in finding out
which are the best, most effective, and most
efficient, methods of treatment.
Although a formulation provides the link
between theory and practice, it does so at a
different level of generality. A theory is the
source of general explanations and general
hypotheses, whereas a formulation is specific to
the person to whom it applies, and therefore is
the source of more specific explanations and
hypotheses. The specificity of the formulation is
the source of ideas about the selection of specific
interventions and about how to adapt them for
use with a particular person. It is for these
reasons that Wolpe and Turkat (1985) describe
a formulation as a theoretically guided way of
structuring the information concerning a patient's problem. It reflects the product of taking
an individual approach to clinical phenomena
and combining this with knowledge of relevant
theories, scientific principles, and research
findings. It involves imposing an explanatory
system upon the material presented, and raises
questions concerning the degree to which this
explanatory system should reflect every aspect
of a problem. One view is that it should reflect
everything, including a patient's past development, characteristic ways of behaving and
forming relationships, emotions, beliefs, assumptions, attitudes, self-evaluations, expectations, attributions, appraisals, and so on. In
practice, the degree of elaboration required
depends upon the purpose for which the

formulation is made. At this point it is probably
sufficient to enunciate one of the principles that
will run through this chapterÐthat of parsimony. In principle, it is always better, and more
useful, to keep the formulation as simple as
possible. The temptation to elaborate a formulation is strong, especially when dealing with
complex cases. However, the simpler and clearer
it is the more readily will its implications be seen
and the easier it will be to use.
Theoretically speaking, the principles that
guide the practice of formulation are derived
from the way in which the concept is defined.
The three main ones to be proposed here are:
(i) A formulation should be based on a
theory, reflecting an attempt to put the theory
into practice.
(ii) A formulation should be hypothetical in
nature, so that it can be modified by information gained during the course of treatment.
(iii) A formulation should be as parsimonious as possible.
6.01.2.2 Formulation and Diagnosis:
Assumptions
In psychological practice there appears to be
a common assumption that only those patients
who participate in research trials have simple
diagnoses, for example, of the kinds defined in
the various versions of the DSM. Diagnostic
systems are useful for ensuring that the
populations studied in different places are
similar in the relevant respects, and they are
useful for insurance purposes, but from the
point of view of the therapist they have
limitations in that they rarely provide specific
implications for treatment. Besides, unselected
samples of patients often do not have single,
clear problemsÐindeed informally they are
commonly said to ªfulfill criteria for an average
of 2.3 diagnoses.º A formulation, however, is
designed precisely to fit the individual and is
intended to help therapists to derive
theoretically-based hypotheses about factors
that contribute to causing and maintaining their
specific problemsÐto explain as well as to
describe. Therefore, the argument runs, diagnoses are less useful than formulations, from
which specific treatment implications can be
derived, and they may be less necessary than
formulations. For example, one depressed
person's sense of failure may be triggered by
an inability to live up to exacting standards and
another person's by an inability to form close
relationships (for any number of reasons, which
may be discovered during therapy and included
in the formulation). Only having the diagnosis
tells the therapist nothing about this difference,

Definitions: What is a Formulation?
and ignoring the difference will reduce the
chances of achieving a satisfactory outcome.
This argument has much to recommend it to
the therapist, especially as diagnoses are largely
atheoretical descriptions, and therapists can use
their theoretical knowledge to construct formulations that are clinically useful. This does
not mean thinking anew with each patient, but
keeping in close touch with theoretical and
clinical research so that, for example, empirically validated treatments can be selected when
the diagnosis suggests they would be appropriate, and individual formulations then used to
specify details of their application. Learning to
work with a formulation instead of relying on a
diagnosis also has advantages when the problems presented are unusually rare or complex
and do not fit readily into a diagnostic system,
or when the system does not succeed in ªcarving
nature at the joints,º and the demarcation
between one diagnosis and another is difficult to
establish. Of course there are difficulties with
this point of view. Seen from the patient's
perspective, over-reliance on the process of
formulation may involve a degree of risk.
What if the theory is wrong? Or if the therapist
is unclear about it? Or susceptible to bias? Or
unable to come up with an adequate formulation? Or attempts to combine one theory with
another without understanding sufficiently well
the implications of doing soÐas when borrowing from experiential or dynamic ideas when
doing cognitive therapy for instance? This risk
can be reduced by formalizing the requirements
of responsible clinical practiceÐby providing
adequate training and supervision, by clarifying
ethical guidelines, and by defining criteria for
professional accountability, including the expectation that practitioners will keep in touch
with the literature relevant to their practice.
Ultimately though, the mysterious faculty of
clinical judgment has also to be brought into
play. Without thisÐwhatever it isÐclinicians
may well run into difficulties, both making and
using formulations.
The implication of this argument is primarily
that, much of the time, formulations are more
useful than diagnoses, provided that therapists
are well versed in the theories they are using, and
that diagnoses, which can after all convey a
large amount of information in a few words,
may help to streamline the process of assessment, and may guide decisions about treatment
in relatively straightforward cases. For example, knowing someone is socially phobic directs
attention towards a fear of being humiliated or
embarrassed, and knowing the diagnosis is of
bulimia nervosa focuses attention on overconcern with shape and weight (among other
things). Underlying problems of self-conscious-

5

ness and poor self-esteem may be relevant in
both cases, so the assessment which provides an
adequate basis for a formulation, and for a
specific treatment plan, must cover more than
the criteria for inclusion and exclusion that
determine whether or not someone qualifies for
a diagnosis.
6.01.2.3 Formulation and Diagnosis:
Controversial Issues
The assumption behind the argument presented above is that a treatment plan based on a
formulation will have a better chance of success
than one based on a diagnosis. However, there is
considerable debate about this issue, and some
recent research suggests that the assumption
could be false. Schulte, KuÈnzel, Pepping, and
Schulte-Bahrenberg (1992) and Schulte (1997)
found that patients with phobias, assigned to a
standardized treatment (exposure in vivo) on the
basis of their diagnoses, responded at least as
well as, and possibly better than, patients whose
treatments had been selected on the basis of
individual problem analyses. With this finding
in mind, Wilson (1996, 1997) summarized the
arguments for using manual-based, empiricallyvalidated treatments, also selected on the basis
of diagnoses, and argued that there are inherent
limitations involved in basing treatment on
idiographic case formulation. As he points out,
making formulations involves making judgments and judgments are fallible. They are
demonstrably susceptible to bias and using
them introduces an additional source of error. It
would be better, he argues, to adopt an actuarial
approach to assessment and treatment as this is
more likely to result in a superior outcome than
using clinical judgment, at least when treatment
manuals are available.
The issue is complex (Beutler, Williams,
Wakefield & Entwistle 1995; Hayes, Follette,
Risley, Dawes & Grady, 1995; Norcross,
Alford, & DeMichele, 1992; Seligman, 1995;
Stricker & Trierweiler, 1995), and differences
will not be settled here. Nevertheless, it is useful
to clarify the basis of the disagreement, as two
issues are frequently confounded. The first
concerns the failure of practicing clinicians to
adopt standardized practices and the second
concerns the dangers of over-reliance on
individual formulations. Those who argue
against the use of formulations seem to forget
that it is the job of practicing clinicians to bridge
the gap between science and practice, and in
doing so to balance the requirements of
recommended procedures with clinical flexibility. A formulation, as defined above, is intended
to facilitate this processÐto assist the clinician

6

Clinical Formulation

in adapting the procedure to the particular
circumstances. When treatments so adapted are
reported to be less effective than expected, then
many factors in addition to formulation could
contribute to this finding. These include the
quality, integrity, structure, and delivery of the
treatment, the accuracy with which the effects of
treatment can be measured, and the relevance of
the measures used to the outcome desired by the
patient. Proponents of the view that treatments
can be selected on the basis of diagnoses alone
seem to assume that case formulation is
idiographic, in the sense that making one is
unconstrained by theoretical ideas and using it
to select interventions is independent of the
findings of clinical research.
Neither of these points is accepted here.
Instead it is argued that individual case
formulation is always relevant, even when
applying a manual-based treatment (examples
will be found below). It is also argued that
formulations have to be rooted in theory to be
useful, and that using clinical judgment is not
providing a licence for subjectivity, but recognizing that at least some of the time clinicians will
not be able to follow the rule book, even when
there is one. Then they have to use their
judgment. In doing so, they can appeal to many
sources of understanding, including theories
about psychological dysfunction, and their
knowledge of the relevant literature. As Stricker
and Trierweiler (1995, p. 997) put it ªit is likely
that the practitioner always will be required
to go beyond firm and available scientific
knowledgeºÐless so when treating phobias
than when treating a complex of depression and
anxiety in someone with a dependent personality type, and not without keeping in touch
with scientific advanceÐbut individual judgment and case formulation remain indispensable clinical tools. Using these tools does not
exempt the practitioner from being aware of the
pitfalls of basing decisions about treatment on
anecdotal case material, intuition, or subjective
impression. On the contrary, working with a
formulation that can be explained to others
provides a check on the use of too much speculation and too many far-fetched inferences.
Therapists need to speak about their patients'
problems in many settings and contexts, and to
do so can make use of any of the available
systemsÐlabels, diagnoses, descriptions, and
formulations. Labels (e.g., manipulative, hysterical, narcissistic, personality disordered) are
efficient but can bring assumptions with them
(and in these examples, assumptions that may
not be to the advantage of the person being
labeled). Diagnoses reflect agreed systems of
categorization and for the most part are based
on particular kinds of descriptions rather than

on theories. They may or may not be subject to
the same disadvantages as labels. Formulations
differ in that they bring together the products of
theoretical knowledge and clinical judgment.
Their theoretical basis reflects ideas about the
factors that cause and maintain problems, and
that precipitate or prolong particular episodes
of distress. This theoretical basis provides a
framework for the type of personal, individual
formulation on which precise decisions about
treatment can be based. Their advantages and
disadvantages are discussed further below.
6.01.2.4 The Difference Between a Formulation
and a Model
Models are ways of conceptualizing particular disorders (e.g., the cognitive hypotheses of
obsessive-compulsive disorder and of health
anxiety described by Salkovskis (1996), or of
formulating particular patterns of functioning
(e.g., the role±relationship models developed by
Horowitz, Eells, Singer and Salovey (1995) or
the functional analytic causal model of Haynes,
Uchigakiuchi, Meyer, Orimoto, and Blaine
(1993). Models, as understood here, are constructed from a particular perspective, so there
are separate cognitive models of panic disorder
(Clark, 1988) and social phobia (Clark & Wells,
1995), and the psychopharmacological or
interpersonal psychotherapy models of panic
disorder differ from the cognitive model. These
differences are valuable in that they stimulate
useful research, as well as the development of
sets of coherent treatment strategies. Using the
cognitive model of panic disorder as an
example, this would suggest that catastrophic
misinterpretation of bodily symptoms plays a
crucial role in triggering panic attacks, and that
understanding this will help people who suffer
from panic disorder to identify the symptoms
that trigger their panics. They will then be in a
position to think again about the meaning of
these symptoms, and to reinterpret them in
terms of (harmless but distressing) panic rather
than of real, impending catastrophe. In order to
facilitate the therapeutic process, the model has
to be translated into a conceptualization (or
formulation), and structured systems for doing
this can be developed, as in this case has been
done by Dattilio (1994). So the model provides
guidelines for an individual formulation which
encourages a new explanationÐthe leap in my
heart could be a response to the coffee I have
just drunk, or a normal arrhythmia that I notice
more readily than I used to because it frightens
me, and not a sign of imminent cardiac crisis.
Although a model has implications for
treatment, it differs from a formulation in that

Definitions: What is a Formulation?
it operates at a different level of generality, and
has a different content. So, the way in which a
formulation applies to particular people will
depend upon their personal history and circumstances. One person's panic may be triggered by
leaps in the heart and another's by losing
concentration when being spoken to (and a
third may find that memories of traumatic
incidents, flashbacks, or nightmares precipitate
panic, possibly because they trigger associated
sensations that then trigger the panic attacks).
There will in practice always be exceptions to
the rule, cases in which, for example, no
sensational trigger can be identified. Then the
clinician may be best advised to base the
formulation on a higher level theory rather
than on the specific modelÐin this case on the
general theory that cognitions, including meanings, are closely related to feelings and behavior,
and that changing one is likely to change the
others. Thus the formulation illustrates, in ways
that are clinically relevant, how the model
applies, and does not apply, to the case. It assists
the therapist in looking for particular theoretical constructs or processes (catastrophic
misinterpretations in this example), and also
in making a judgment about the degree to which
the case is typical.
Atypical cases arise when patients have more
than one difficultyÐsocial anxiety as well as
panic disorder for exampleÐor when they have
especially complex or rare problems such as
panic attacks in the context of avoidant or
borderline personality disorder. Then, conceptually speaking, it may be more useful to draw
on more than one model to construct a single
formulation, or to look for models with a higher
order of generality. Writing about psychodynamic formulation, Perry et al. (1987) point out
that overlapping models of mental functioning
may emphasize different aspects of development and psychopathology. They distinguish
ego-psychological, self-psychological, and object relations models, and make the important
point that a certain amount of trial and error
may be needed in constructing a formulation
that explains the presenting data: ªthe absence
of a meta-model to explain all data makes this
trial and error unavoidableº (p. 546). What
clinicians are looking for in a formulation is a
way of explaining and understanding the
relationship between a patients' inner lives
and their outer lives that is the product of their
personal history, explains present difficulties,
and guides future therapy. Their sources in this
search include knowledge of diagnostic systems,
of relevant theoretical models, and of outcome
research, as well as information about the
individual caseÐotherwise they would have to
reinvent the wheel each time.

7

6.01.2.5 Types of Formulation
Typically, different therapeutic schools are
thought to use different types of formulation. In
general, behavioral and cognitive therapies
make use of more mechanistic formulations,
based on theories about learning and detailed
functional analysis (Hayes & Follette, 1992), or
on theories about processes such as the
supposedly circular relationships between
thoughts and feelings, and more dynamic
therapies employ more narrative-based formulations, placing current problems in the context
of a developmental history. Some systemic and
experiential approaches to therapy adopt a
third, essentially dynamic, approach, claiming
that formulations have constantly to be reformed in the present, as therapy focuses on
moment-to-moment events (Goldman &
Greenberg, 1997). They also point out that
the process of formulation can be dangerous
and limiting when it makes use of preset
categories and ideas. A constantly changing
situation then appears to be fixed, and
opportunities for change may be obscured
(Eells, 1996; Rosenbaum, 1996). However, the
process of formulation is still thought to be
essential, and its main purpose is still to look for
patterns and links that assist in understanding,
and to provide ideas about how to bring about
change. So, distinctions can be applied too
rigidly. The developmental history of a problem
or a person, or the narrative, is always relevant
(Nicholson, 1995; White, 1989), although it may
be understood in different ways, and so are ideas
about the mechanisms that precipitate an
episode of distress or perpetuate a problem.
Overt differences between types of formulation
are therefore relatively unimportant to an
understanding of the term, and of the functions
that the activity of formulating a case performs
for the therapist.

6.01.2.6 Levels of Formulation
When making a formulation, it is necessary to
think at many different levels, and the number
of levels postulated obviously varies with the
theory being applied. Taking an example from
cognitive-behavioural therapy (CBT) to illustrate the point, at the most superficial level, or
the level of ªovert difficultiesº (Persons, 1989),
the main task is to define the problems and the
ways in which they are maintained, usually in
terms of vicious cycles. Someone who feels
depressed may withdraw from company, think
about being all alone, and become increasingly
depressed. Even such a simple formulation
suggests a focus for interventionÐworking to

8

Clinical Formulation

reduce the withdrawal. Thinking about the
factors that precipitated the depression adds
another level to the formulation. The person
might have become depressed when their job
required them to move to a new place, when
they got divorced, or when their children left
homeÐfactors that would demand different
types of adaptation, to be promoted by the
therapist in different ways. Stressors are
additive, so many factors may be involved,
and an apparently minor stressor may be the
straw that broke the camel's back (and relatively
irrelevant to the formulation), or it may reflect a
particular personal vulnerability. Factors that
predispose someone to become depressed,
biological as well as psychological factors,
add a further level, and the way in which these
are understood, and formulated, will again
influence the selection of interventions. At the
most profound level of all, assuming that ªat
some level it all makes sense,º the formulation is
supposedly capable of reflecting the meaning of
structures through which people interpret and
think about, remember and recount, their
experiences, and theoretical assumptions about
the origin of these things.
This is the standard way in which psychiatric
formulations have traditionally been madeÐin
terms mainly of predisposing, precipitating, and
perpetuating factors. However, there are yet
other levels to consider, reflecting social,
cultural, and historical factors. Social assumptions (ªmen should not show their feelingsº or
ªwomen are bad organizersº) influence the
views of therapists as well as patients, and
cultural assumptions may or may not be shared
between therapist and patient. Some cultures,
for example, do not share the common Western
therapeutic goal of autonomy, especially for
women. Others assume that a relationship
between a professional person and their client
is one involving activityÐor authoritative
pronouncementsÐon the one side, and
passivityÐor receptivityÐon the other. In
addition, different hierarchies of values can
interfere profoundly with the therapeutic process. An example in our culture is when someone
thinks it more important to avoid giving offence
than it is to tell the truth. Although it is never
possible to stand outside all of these factors,
making a formulation helps therapists to think
about them, to identify them clearly, and to
become aware of their potential influence on the
interpretation of other people's circumstances.
It can help therapists to ensure that the ways in
which they understand problems and select
interventions are not influenced by unwanted
biases. Seen in this way, a formulation assists
therapists in achieving a relatively objective
stance.

Formulations are always made from a
particular perspectiveÐin the author's case
made (usually) from a cognitive-behavioral
perspective, and from that of a White woman
of a certain age, living in Britain now, whose
ways of thinking have been formed by her own
learning and experience. A formulation is
neither about fitting information about a
patient to a predetermined formula, whether
that formula be derived from a general theory or
from a more specific model, nor is it a personal
judgment, though both things are relevant. It is
about developing the kind of understanding of
another person, their circumstances and their
difficulties, that enables a therapist to apply the
theoretical knowledge acquired during training
to help that person. There is no single right way
of making a formulation. The general aim is to
map the territory so that one can then explore
the possibilities for change, and not to let these
be influenced by factors that are irrelevant to, or
unwanted by, the person who is receiving help.
6.01.3 PURPOSES: WHAT A
FORMULATION IS FOR
One common view of the purpose of
formulation is that it is for explaining the past,
making sense of the present, and suggesting
what to modify in order to influence the future.
It can also be an important means of communicating understanding, either to the patient or
to another professional, whether in the role of
supervisor or colleague. However, its prime
purpose is to help therapists to apply the theory
they have learned to their practice (a comprehensive account of different approaches to
formulation is given by Eells (1997).
In practice, there are many answers to the
question ªWhat is a formulation for?º The main
functions of formulation are listed in Table 1.
The main point is that making formulations is
an essential, and not an optional, element of the
therapeutic process. Formulations do not have
to be 100% accurate or complete in order to be
useful precisely because they provide a source of
testable hypotheses. They can be changed when
they turn out to be wrongÐand nothing is lost
by using a partial or partially mistaken
formulation which can be improved and
corrected as the process of therapy continues,
and reveals the initial mistake. They guide
questioning, and open the therapist's mind to
the kind of understanding from which effective
treatment strategies can be derived, applied, and
evaluated. Therefore, the author would argue,
that therapists should work with a formulation
in mind right from the start. Ideas about people
and their problems cannot be kept at bay or

Purposes: What a Formulation is for
excluded, even when first meeting them or
reading a referral letter about them. One way of
trying to ensure that this information is openly
received and accurately assessed is to engage
immediately in the process of formulationÐin
applying both theoretical and clinical knowledge to the particular case. Just as when first
visiting a new place, a rough sketch map may set
one on the right road at first, but will need
expanding and revising if it is to guide more
detailed exploration.
6.01.3.1 Understanding: The Overall Picture or
Map
A formulation ªprovides the map of the
territory and once you have that you can use
whatever vehicle you are most comfortable
withº (Beck, 1991). Formulations, just like
maps, provide an overall view (often in
diagrammatic, conventional form) of something that it is not possible to see directly all at
onceÐthe wood as well as the trees. They
indicate which are the important features, their
size and shape, and the way in which they relate
to each other. Mapping the territory is clearly
the product of accurate assessment (see Section
6.01.4), and formulating enables therapists to
make and to justify such statements as ªthis lack
of energy is part of the depression,º or ªin this
case the anxiety seems to be primary and the
depression secondary.º Similarly, formulations
can indicate where information is missing and
prompt appropriate questions: where did this
low self-esteem come from? Why does it become
apparent in the context of close personal
relationships but not at work?
6.01.3.2 Prioritizing Issues and Problems
An overall formulation helps to differentiate
what is essential from what is secondary in a
general sense. It also helps in a more particular
way to decide which issues or problems should be
prioritized. Someone who believes that they
cannot change is unlikely to remain engaged in

9

therapy unless they can see the point of it.
Creating hope, or the context for a developing
relationshipÐsomething with a future, in which
change is inherent and undeniableÐthen becomes a priority. Likewise, an initial assessment
may indicate primarily that inability to trust
people will make it hard to disclose distressing
material, and building trust within the therapeutic relationship is necessary before a more
detailed and accurate formulation can be made.
It is probably not unusual for patients and
therapists to start the process of therapy with
somewhat different priorities. Usually this
problem can be overcome during assessment
and those early stages of therapy during which
goals become clear or are specifically agreed. But
sometimes different priorities persist, and then
the process of re-formulating can help to solve
the problem. For example, an anxious and
hypochondriacal patient who was worried,
among other things, about seeing ªfloatersº in
his visual field, started to respond well to
treatment that was formulated in terms of his
underlying sense of vulnerability. The formulation reflected the way in which his various
concerns made him feel threatened, and think
that he was at risk for being unable to handle a
number of initially rather vaguely specified
distressing eventualities. However, although
his confidence increased, his distress about the
floaters did not diminish. If anything it increased, in tune, it must be acknowledged, with
the therapist's frustration when discussing this
issue became his main priority. Focusing the
work of one session on the meaning or understanding of this problem revealed (for reasons
which later became clear) that visual anomalies
for this person felt, in his words, ªlike a
bereavement.º Formulating this aspect of the
problem in terms of loss rather than in terms of
vulnerability changed the focus of treatment,
which then became more productive. This
example also illustrates how characteristics of
the process of therapy can contribute to ideas
about the formulation, especially in those cases
in which change is not proceeding as well as
might otherwise be expected.

Table 1 Summary of the purposes of formulation.
Clarifying hypotheses and questions
Understanding; providing an overall picture or map
Prioritizing issues and problems
Planning treatment strategies
Selecting specific interventions
Predicting responses to strategies and interventions; predicting difficulties
Determining criteria for successful outcome
Thinking about lack of progress; trouble shooting
Overcoming bias

10

Clinical Formulation

6.01.3.3 Planning and Selecting Intervention
Strategies
Once a hypothesis about how the presenting
problem can be understood has been formulated, the most important functions of a
formulation are in planning a treatment
strategy and selecting appropriate methods of
intervention. Persons (1989) provides some
lucid examples: someone who avoids exercise
because they are bad at time management,
scheduling, or self-organization has a different
problem to overcome, and needs to acquire
different skills from the person who avoids
exercise because they are embarrassed about
their appearance; insomnia that is associated
with the fear of letting go may require different
interventions from insomnia that results from
overcommitment.
The way in which a problem is formulated
thus determines what should be done about it
(Blackburn & Twaddle, 1996; Butler & Low,
1994; Eells, 1997). If avoidance maintains the
problem, then facing the fear is likely to reduce
it, and in individual cases the formulation helps
to specify idiosyncratic aspects of the avoidance (the spider phobic who will not walk
under trees; the social phobic who is more
fearful of silence than of conversation). The
general vicious cycle model is common to
bothÐand indeed, a standardized method of
treatment of proven effectiveness, exposure in
vivo, is readily available. The individual
formulation is still necessary because it
specifies exactly what steps to encourage the
person to take.
Planning overall strategies is just as important a product of formulation as the selection of
specific methods of intervention, but is a more
complex task, and requires of the therapist more
than one level of understanding. The way in
which depression or anxiety is understood may
suggest, for instance, that it would be helpful to
increase levels of activity before discussing
thoughts associated with depression; or to build
up a repertoire of coping skills before facing
fears. Many such imprecations are based on
clinical judgment (or clinical intuition) as much
as on theoretical or experimental work, and in
these cases it is especially important that they
should be made clear by means of a formulation. For example, it is often said that when
working with people who have suffered abusive
experiences in childhood, one should help them
to develop a variety of support systems, ways of
dealing with intense feelings or suicidal impulses
and of creating around themselves a sense of
safety, before exploring memories of early
traumatic experiences, and the meanings of
such events, in depth.

Clearly, this overall strategy reveals assumptions about how the effects of these events can
be understood, about the effects of talking
about them, and the interventions usedÐ
assumptions which formulations clarify, and
which are potentially amenable to research, but
which will differ according to the therapist's
theoretical orientation. A secondary purpose of
clarifying the formulation and its function in
selecting strategies and interventions is to
facilitate evaluation of interventions.
6.01.3.4 Predicting Responses and Difficulties
Because a formulation reflects theoretical
assumptions, it helps therapists make two kinds
of predictions that are essential in therapy: to
predict the effect of the intervention, assuming it
is successfully applied, and to predict the
stumbling blocks and difficulties that will be
encountered during therapy. An anxious person
treated during a clinical research trial (Butler,
Fennell, Robson, & Gelder, 1991) held the belief
that ªall my ideas are bound to be wrong.º She
became more confident as she learned to identify
her ideas, to act upon them, and consciously to
evaluate the consequences of doing so. Her
formulation enabled us to predict first that she
would feel especially vulnerable and be likely to
overgeneralize and catastrophize the consequences when she made mistakes, and second,
that she was likely to find it especially difficult to
apply the new strategy when relating to her
partner, but easier to build up the necessary skills
(and courage), and to increase her confidence, in
the context of other relationships (including
ours). Treatment in this case was guided by the
requirements of a treatment manual, and the
example illustrates the important role played by
clinical formulation in the application of
standardized treatments.
It is probably true to say that interpersonal
difficulties are one of the most common sources
both of patients' problems and of problems
encountered during psychological therapy; for
example, an ability to form superficial relationships without being able to sustain deeper
friendships, or veering between passivity and
aggression when interacting closely with others.
Such difficulties also play their part within the
therapeutic relationship, and they are much
more easily dealt with if the processes involved
have been understood in terms of the theory
being used, and problems predicted in advance.
Formulating helps people to recognize such
patterns, to develop hypotheses about their
origins, functions and effects, and to think
about whether and how to engage in a process of
change.

Purposes: What a Formulation is for
6.01.3.5 Determining Criteria for Successful
Outcome
Theoretically a formulation provides the
basis for hypotheses about what needs to
change for someone to feel better, or the goals
of therapy in the broad sense of the term. This is
obvious when a theoretical model for the
condition being treated is available, but the
point applies more generally as well. The
present version of the cognitive model of social
phobia (Clark & Wells, 1995), for example,
suggests that self-awareness, or self-focused
attention, plays a central role in the disorder. In
outline, when in a socially frightening situation
a social phobic feels self-conscious, notices
symptoms of anxiety and tries to keep safe. An
individual formulation based on this model
would specify the way in which this actually
happens. For example, when speaking to others
(e.g., colleagues during a lunch break), Marie
became aware of the sound of her own voice, felt
anxious, flustered, hot, and shaky, and found it
hard to listen to what was being said. She
thought other people must be able to see how
nervous she felt and tried to fade into the
background as quickly as possible (keeping
herself safe by avoiding eye contact, saying
little, speaking in a quiet voice). Both general
and specific goals for change can be derived
from thinking along these lines. In simple terms,
if Marie can focus her attention outside herself,
and listen without self-criticism to those around
her, if she can reverse the safety behaviors (make
appropriate eye contact, speak more audibly,
move around freely), she will break the cycle
and start to feel less anxious. The general
criteria for change are reflected in the three
elements of the model specified here, the selfawareness, safety behaviors, and symptoms of
anxiety, and specific ones reflect the individual
ways in which these factors are manifested in the
case of Marie.
Of course this might not be the whole story.
Marie's social anxiety may be based on a belief
in her own unworthiness relative to others, and
reflect an unhappy history of family relationships. Such formulations again indicate criteria
for changeÐa sense of worthiness or the ability
to form more satisfying relationships in the
present. The difficulty here is that more
abstract and general phenomena are harder
to identify, define with any precision, and
measure than more superficial and specific
ones. Criteria for change are therefore more
easily derived from formulations at lower than
at higher levels of abstraction, and indeed the
more specific the formulation the easier it will
be to be clear about what exactly needs to
change.

11

6.01.3.6 Thinking About Lack of Progress
There are many possible reasons for lack of
progress in therapy, including working without
making a formulation. The first line of defence
when this happens must be to formulate or to
reformulate the problem. The way in which this
is done will have specific implications for the
next steps in therapy. For example, if the
problem is a long-standing, chronic one, it may
be that much practice is needed and that it is
unrealistic to expect faster change, in which case
it may be important to think about how to keep
the momentum of change goingÐabout how to
maintain hope and create the energy for change
when doing so is difficult. If the original
formulation was inaccurate or incomplete, the
failure to change may suggest that different
strategies and interventions are needed. When
lack of progress leads to frustration, and the
reactions of both the patient and the therapist
interfere with subsequent progress, including
these factors in the reformulation can reveal
ways of overcoming them. Blocks in treatment
are nearly always informative and formulation
skills should be used to identify their specific
nature.
Often this is complex and involves making
hypotheses about past events, the exact nature
of which can never be known. Possible
formulations in these circumstances, often
derived from a combination of observation
and understanding of the apparent effects of the
past on the present, can suggest which avenues
to explore so as to make further progress. For
example, a patient who provided a cold and
dispassionate account of a childhood in which
she was neglected, often frightened and sometimes threatened with physical abuse, appeared
to have developed a variety of ways of
controlling both the experience and the expression of her emotions. Many, but by no means
all, of these ways were dysfunctional. A possible
formulation of this case suggests that improvement will remain blocked unless or until she
becomes able to experience and express the
relevant feelings. Doing this is likely in the first
instance, to precipitate periods of distress, and
the precise implications for therapy to be
derived from it will depend on both the skill
and the emotional sensitivity of the therapist as
well as on a willingness to adapt the formulation
according to what happens.
Drawing these points together, it is clear that
formulations cannot be treated as a matter of
last resort, only to be constructed and worked
on when the going gets difficult, when dealing
with chronic problems, when treatment has
apparently gone on too long, or when preparing
to report to someone else. Formulations do not

12

Clinical Formulation

provide the answers to questions, but a rich
source of questions and ideas of potential
therapeutic value. They should not become
the tool for applying a preconceived theoretical
plan to someone for whom the plan does not fit,
nor should they focus exclusively on someone's
problems and difficulties. Accurate formulation
takes account of a person's strengths as well as
failures, talents and potential, as well as shortcomings and failures.
If formulations can be so useful it is surprising
that so little attention has been devoted to them
both within training programs and in the
literature. One reason for this may be that
formulations were supposed to follow logically
from the processes of assessment and functional
analysis, and additional skills were not often
specified. A more important one is probably that
formulating is difficult. As already indicated, in
practice it involves exercising clinical judgment
as well as the ability to relate theory to practice.
Also, until recently, there was less communication between people with different theoretical
backgrounds, and fewer challenges to think
about alternative methods of formulating specific cases. So, the next important question is
ªHow do you construct a formulation?º
6.01.4 METHODS: HOW TO CONSTRUCT
A FORMULATION
The main reason for considering the purposes
of formulation before thinking about how it
should be done is that there is no single correct
methodÐhow you do it is in general determined
by understanding the purposes that it serves,
and in particular by the theoretical orientation
of the therapist. The end product should enable
the therapist to relate theory to practice in a way
that can direct and inform the process of
therapy, and the methods used vary enormously. For the student this is both confusing
and liberating, as it demands creativity and the
ability to deal with abstractions as well as the
more mundane skills primarily involved in
assessment. Assessment is a necessary step in
the development of a formulation, but it is not a
sufficient condition for it. Unfortunately, it is
possible to assess, in the data collection sense,
without developing a formulation.
6.01.4.1 Sources of Information
An account of presenting problems, informed
by knowledge of psychological processes and
diagnostic systems, provides a common starting
point, and assessment covers all of the four
aspects of functioning illustrated in Figure 1 and
their determinants: cognition (thoughts, as-

sumptions, attitudes, beliefs, images, etc.);
affect, behavior, and physiological sensation;
the present context for the ways in which these
things are manifested; and an account of their
background and associated developmental history. It also draws on information gathered
during the process of referral, such as a summary
of the problems as understood by the referrer, of
the reasons for requesting help and of responses
to treatment received so far, and on the
impressions and observations made during the
first encounter with the therapist when the
processes of mutual interaction are set in
motion.
Therapists use many skills in helping them to
understand this material: theoretical knowledge; products of academic learning and
professional training; and clinical judgment.
The process of encapsulating this understanding
in a formulation, which at first takes time and
becomes quicker with practice, is facilitated by
adopting a questioning stance. The aim would
be to be able adequately to answer three of the
key questions that patients ask: Why me? Why
now? What keeps it going? and in doing so it
helps to draw on a further set of questions
central to the process of formulation, which
therapists can pose either to themselves or to
their patients: How do you understand that (or
make sense of it)? What do you think is going
on? How does this all fit together? What might
be the missing links? What does that mean
about you now? Is there a pattern here?
Formulations are useful in helping people to
think again about their difficulties, and see them
in a new (e.g., clearer, more realistic, or more
illuminating) light, and the process of assessment potentially reveals the patient's present
point of view. In order to develop an understanding of such personal and unique phenomena, it is particularly useful to pay attention to
the ways in which people react to their
experiences. Their comments provide a rich
source of such informationЪI have to keep
controlº or ªI need to know I am succeedingº
are remarks that suggest hypotheses about the
self and about underlying processes and
mechanisms. Ideas expressed about others, such
as ªshe'll be miserable aloneº may fit with
assumptions that precipitate or maintain presenting problems. General comments of the
kind ªyou have to conform or you can't get onº
reveal attitudes that may (or may not) dominate
within the real world in which the person lives.
Expectations about the future, including those
about the process and outcome of therapy, are
also revealing: ªI won't be able to do what is
needed,º ªThere are some things I would rather
not talk about,º ªI'm relying on you to make me
better.º In order to formulate, it is important to

Methods: How to Construct a Formulation
understand the personal significance of experiences as well as their phenomenology. These
comments illustrate well how the processes of
formulation and assessment meet, and indeed
may overlap. Therapists assess to find out about
problems and their context, and they formulate
differently according to what they think their
findings mean.
Patients' comments may need clarifying
during assessment if they are to inform the
process of formulation maximally. Statements
that are apparently clear to the person expressing them may not be clear to the therapist, or
may reveal ambiguities and contradictions, as
when angrily saying ªI'm not capable as a
parentº (when sadness sounds more likely and,
superficially, more appropriate), or when
commenting wryly that ªI felt sorry for myself,º
without elaborating on what that means. One of
the most useful sources of information for
formulation comes from the mutual reactions of
the patient and therapist to each otherÐ
information that is used differently in different
types of therapy, and which is understood using
different theoretical systems, of varying degrees
of sophistication, but which is always relevant.
The processes of assessment and formulation
therefore go hand in hand, and inform each
other, but they remain different processes. Ideas
about how to understand (conceptualize or
formulate) what is being said, about its personal
meaning and implications for theorized psychological structures and processes, guide
questions and observations. When formulating
as well as when assessing, the information
gathered changes and shapes these ideas as
hypotheses are formed, revised, and (theoretically) refined. So, making a formulation is not a
one-off activity that defines a fixed state, but the
reflection of a dynamic process, and the
resulting system of understanding develops
and changes over time. This is why the process
of formulation should start at the same time as
the process of assessmentÐjust as the process of
finding ones way around a new place starts with
the first encounter with itÐand may be on paper
rather than in person.
Two points that follow from this line of
argument help to determine how a formulation
is made. First, if therapists are always formulating as well as assessing, then their
questions and statements should be guided by
conceptual hypotheses. They should always be
able to answer the question ªWhy did you ask
that then?º The answer should not just be
phrased in terms of curiosity or information
gathering, but should relate to a hypothesis
about how to understand the minutiae of the
case. The patient's response to the therapist's
comment or question is then maximally in-

13

formative. This may sound unrealistically
demanding, as if every sentence the therapist
utters should be shaped by the developing
formulationÐindeed, it is intended as a rule of
thumb rather than as a categorical imperative.
However, it is less unrealistic than it might seem.
The initial question in the therapist's mind
could be quite a simple one, for example: Is the
withdrawal described by this person associated
with feelings of depression and sadness or is it a
kind of avoidance motivated by fear? Will
attentive listening help this person feel sufficiently comfortable to disclose significant
material? Are my questions too specific and
intrusive at this stage? Answers to these
questions could of course lead to more complex
ones: Is this person's reticence a product of
experiences that have destroyed trust? Does it
reflect a preference for an autonomous style of
relating to others? Is it a product of inexperience
and lack of practice or opportunity in talking
about intimate and personal matters? Is this way
of interacting culturally unfamiliar to them?
The second point is that the process of
therapy should not be artificially separated into
discrete stages of assessment, formulation, and
treatment (or intervention). It is not that these
processes cannot be distinguished, or that one
or other of them may not predominate at a
particular time, but that they cannot in practice
be wholly separated from each other. Thus, one
of the hardest tasks therapists have to learn is
how to bear all three of them in mind at
onceÐhow to gather information, think about
it in theoretical/structural terms, and remain
aware of the various ways in which they are
likely to exert an influence, so as to enhance the
potential for productive change, rather than
limit or delay it.
The many sources of information available to
therapists when starting to develop a formulation, assuming an adequate process of assessment has been set in motion, are summarized in
Table 2. This list includes both direct and
indirect sources of information, information
from standardized questionnaires, and from
initial interventions such as self-monitoring and
homework assignments (when these are used).
The purpose of this summary, in the context of
the preceding discussion, is not to overwhelm
therapists with long and exhaustive lists of
material to be gathered, items to consider,
processes to complete, and so on, but to illustrate
that there is an enormously rich source of
relevant material potentially available, and the
process of formulation can draw on any of it,
beginning anywhere. The process of formulation
is essentially one of abstraction and it works by
relating observable phenomena to hypothetical
underlying processes and mechanisms. It is not

14

Clinical Formulation

necessary to observe everything before making a
guess at what lies underneath. An (informed)
guess may either indicate the need for more
assessment or it can short circuit the process.
Because formulating is a dynamic process, and
depends on the ability of the therapist to retain
an open mind, the process can productively start
to serve the functions listed in the previous
section straight away. Therapists can focus their
minds on the process of formulation by asking
more formal questions: How can I understand
the information I have been given in terms that
make theoretical sense? What implications does
that understanding have for what to do next?
What difficulties will I have, working with this
person? What difficulties will they have (working in this way) with me? What use will this
person be able to make of treatment? Answers
help to determine how to intervene and to predict
what will or will not happen as a consequence.
6.01.4.2 Putting the Information Together
Given that a formulation provides connecting
links between theory and practice, the precise
form that it takes will be partly determined by
the theoretical approach of the person making
it. Nevertheless, some general points apply, and
these are illustrated here using the cognitivebehavioral approach.
First, initial formulations can provide crosssectional understanding of an aspect of the
presenting problem. The most obvious example
is probably that of a vicious cycle which
summarizes the way in which a particular,
readily accessible, symptom pattern is thought
to be maintained. It is used here to illustrate the
way in which a formulation helps to specify
processes, links, and mechanisms. In this case
the focus is on certain kinds of links. Other
cross-sectional formulations might focus on

other patterns, for instance in interpersonal
functioning, sequences of behaviors and their
consequences; thoughts, feelings, attitudes, and
beliefs; dilemmas and traps. In this example
(Figure 2), a woman living through a stressful
period described feeling tired much of the time
and being unable to relax. Asked about what
goes through her mind when trying to switch
off, she described a stream of worries, most of
which were rather vague and hard to specify in
detail. The worry disturbed her sleep pattern,
which exacerbated the tiredness. A cycle, which
symbolizes how one thing leads to another, can
easily be illustrated diagrammatically, and it has
obvious implications. Breaking the links will
help to solve the problem, and this can be done
in various ways, such as learning to relax,
identifying and dealing with the worries, or
taking hypnotic medication. The assumption
behind the formulation so far is that the
problem will subside if the process that
maintains it is interrupted, and the intervention
selected could be determined by the preferences,
understanding, or skill of either of the parties
involved.
However, a formulation essentially relates
theory to practice. Applying the cognitive
model to this case would suggest that a close
relationship between thoughts and feelings is
likely to be of central importance. There are at
least three ways in which this initial formulation, in its hypothetical and simplified form, can
help the cognitive therapist to focus on factors
that theoretically are likely to be relevant. It
identifies worry as an important cognitivemaintaining factor, it reflects an overall understanding of the problem, suggesting that the
symptom pattern is recognizable, understandable, and changeableÐattitudes which may
differ strikingly from those the patient starts
withÐand it poses questions about the context
of the problem. Nothing has been specified

Table 2 The main sources of information for use in formulation.
Examples of direct information
Reports of present phenomena: cognitive, affective, behavioral, and physiological
The context: historical background and development, real life problems
Reactions, comments, and expectations, about the self, others, therapy, events, etc.
Interactions within therapy: ability to relate, tenor of relationships
Observations of body position, movement, facial expression, eye contact, etc.
The outcome of interventions such as self-monitoring, homework assignments,
behavioral experiments, etc.
Products of questionnaires, tests, standardized interviews, systematic observation, etc.
Examples of indirect information
Knowledge about diagnosis: DSM
Referral information: summaries, previous treatment, opinions
Knowledge of cultural norms (of the therapist and of the patient)
The socioeconomic and political context

Methods: How to Construct a Formulation

15

Under stress

Feel tired and
unable to relax

Can’t sleep
well

Figure 2

Worries keep
coming to mind

Example of a simple cross-sectional formulation: basis for a more complex formulation.

about why this is, for this person, a stressful
period. Theory-driven questions help to develop
more hypotheses: What does it mean about her
habitual response to stress? How does she
construe her present situation? What does her
reaction to it mean to herÐabout herself, about
other people, and/or about the world in which
she finds herself? So, the initial formulation
triggers further inquiry, and starting from a
simple cross-sectional map can lead to more
sophisticated levels of understanding, and to
more complex formulations, as well as being
practically useful. The precise way in which this
happens will be determined by the theory being
used.
Cross-sectional formulations can also provide an outline summary of the way in which
complex underlying factors are understood, or
of the way in which aspects of a problem are
linked. Three statements made by an unemployed, unconfident young man with a wide
range of social, interpersonal, and affective
problems were used as the starting point for the
initial formulation illustrated in Figure 3: ªIf I

always please others they'll never find out about
me,º ªI'll be OK if I stick to doing easy things,º
and ªPeople will reject you if you don't toe the
line.º In this diagram, three aspects of his
problem are represented in different ways. First
there is a rather shapeless ªthought bubbleº at
the top in which hypotheses about underlying
cognitive structures, beliefs, attitudes, and rules
about himself have been put into words: ªI'm
incompetentº; ªI have to do what others askº;
ªI'm thick (stupid) . . . º These actual words
were his responses to specific (theory-driven)
questions, and they illustrate how the process of
formulation interacts with that of assessment,
and depends on the ability to abstract and to
generalize. The broken line is labeled a
ªprotective wallº because it represents the idea
that the three statements listedÐstarting points
for a more detailed formulationÐreflect behaviors that serve a function. Reacting in these
ways protects him from having to confront (the
hypothetical) underlying beliefs and attitudes,
and prevents others from discovering them,
both of which would be painful experiences for

16

Clinical Formulation

him. However, these protective reactions cause
problems, not specified here but referred to in
the box in Figure 3. This formulation contributed to the process of developing a shared
understanding of some complex problems, and
it was used to explain how change would
probably involve working at all three levels. It
also has implications for decisions about
general aspects of therapy. For example, it
suggests that at times this will be a distressing
process that will demand sensitivity and a good
sense of timing from the therapist.
Cross-sectional formulations potentially reflect ideas about psychological processes and

mechanisms as well as about the relevance and
relative importance of different facets of a
problem. Longitudinal formulations reflect
assumptions about etiology as well. They are
used in most kinds of therapy, and are readily
illustrated in the case of CBT. The basis for
using this theoretical model in clinical practice
has been summarized in the form of a template
(Table 3) which can be used to illustrate how
theoretical understanding can be translated into
practice. This shows that, theoretically, experience, both early in life and subsequently, gives
rise to a set of beliefs and assumptions about the
world, about other people, and about the self.

I’m incompetent
I have to do what others ask
I’m thick

Protective wall:
“If I always please others they’ll never find out”
“I’ll be OK if I stick to doing easy things”
“People will reject you if you don’t toe the line”

Me with my
problems

Figure 3

Example of a cross-sectional formulation.

Methods: How to Construct a Formulation
These beliefs are seen as a product of the ways in
which earlier events have been perceived,
understood, and remembered. They can be
functional or dysfunctional, actively influential
or latent at any particular time, and relatively
easy or hard to identify and to recognize. A
critical incident (see also below) is an event that
fits with a beliefÐbeing rejected for someone
who believes they are not socially acceptable, or
being let down for someone who believes that
other people are unreliable or untrustworthy.
Critical incidents activate the relevant beliefs
and assumptions, and thus produce negative
automatic thoughts (NATs). Then a variety of
interacting cognitive, affective, behavioral, and
physiological reactions follows. At this level the
problem is theoretically maintained by cyclical
processes of the kind summarized in the crosssectional vicious cycle described above.
Clearly a template such as this can be used to
structure information about a patient, and this
will have implications for what the therapist
does. For instance, if it appears that dysfunctional beliefs play a small part in the presenting
problem, or are well balanced by a set of positive
beliefs, the theory (and the formulation derived
from it) suggest that the work should focus
predominantly on the level of maintaining
factors. Another type of implication might
reflect the degree of verification available for the
theory. For example, psychologists do not yet
know which are the most effective ways of
changing beliefs (the cognitive frameworks with
which people approach the world). One common strategy is therefore to begin working at the
level of the NATs and to evaluate the degree of
Table 3

17

belief change that follows. The processes of
change may, or may not, be set in motion by
work at this level. If not, then another
hypothesis might be that one of the many
processes now available for changing beliefs
should be adopted as well as or instead of. This
example is not meant to explain how to do CBT,
but to illustrate how the internal map provided
by a theoretical understanding relates to a
specific formulation, and how therapists can use
such maps as guides even when there is
incomplete evidence for the theories upon
which they are based. Doing so enables them
to explain what they have been doing, and it
enables others to decide whether their actions
were skilful, appropriate, and so on.
6.01.4.3 Key Factors and Basic Elements
This example also illustrates that when
learning how to construct a formulation, it
can be helpful to think in terms of key factors.
Critical incidents provide a good example of
these as they reflect the way in which hypothetical underlying mechanisms are manifested,
and link these with observations about present
phenomena. Critical incidents are ªcriticalº
because they provoke a high degree of affect,
often in excess of what might otherwise be
expected (an over-reaction, such as becoming
enraged if kept waiting for 10 minutes); they are
easy to notice and remember, and are of special
significance for the person who experiences
them. Examining them potentially reveals other
elements of the CBT template: underlying

Template for a longitudinal formulation using cognitive-behavior therapy.
Experience (early or otherwise)
;
Beliefs, about the self, the world, and others, which are expressed in
categorical statements: I am . . . ; the world is . . . ; others are . . .
;

Assumptions derived from beliefs, which can be expressed in
conditional statements: If I . . . then . . . ; One should . . . otherwise . . .
;
Critical incidents
;
Activated beliefs and assumptions
;
Negative automatic thoughts (NATs)
; :
Cognitive, behavioral, affective, and physiological reactions

18

Clinical Formulation

beliefs, preferred coping mechanisms, maintenance cycles, and so on. Focusing on critical
incidents is thus theoretically helpful when
stuck in constructing a cognitive formulation.
This is not to say that all cognitive therapists
think about them, or base their formulations
upon them. An alternative method might
involve working from a problem list, weighting
the problems for importance, and going on to
abstract and understand the connecting themes
and links in ways that fit with the theory. The
point is that within a particular method of
working there are many ways of constructing a
formulation, but it can be helpful to keep those
factors in mind which play a central part in the
theory, or in revealing the manifestations of
important theoretical constructs whether these
are core beliefs, core interpersonal schemata, or
core conflicts. To repeat, there is no single
correct method.
Use of the word ªcoreº suggests that
formulations may be thought to have certain
basic elements, and that unless these are
identified the formulation will, in Perry et al.'s
words, ªlack an integrative coherence.º When
writing about psychodynamic formulation and
about central conflicts, Perry et al. (1987, p. 546)
say ªThe aim is to find a small number of
pervasive issues that run through the course of
the patient's illness and can be traced back
through his or her personal history, and then to
explain how the patient's attempts to resolve
these central conflicts have been both
maladaptive . . . and adaptive.º The overall
intention is clearly closely similar across
different therapeutic orientations, as is the
general approach: first, apply a particular,
theory-driven model; if that does not in practice
fit the particular case, explore further using
questions and trial and error in the (scientific)
search for a formulation that fits better.
This process might be facilitated if there was
agreement over which were the basic elements of
a formulation and an atheoretical way of linking
them together. One way of doing this has been
developed by Goldfried and his collaborators.
This transtheoretical coding system ªwas developed as a common language for use in
conducting comparative process research across
orientationsº (Goldfried, 1995, p. 222). It
specifies which are the relevant components
of functioning (e.g., self-observation, self-evaluation, intention, emotion, and action) and the
types of links that can be made between them
(vicious cycles, patterns, contradictions). These
can be manifested both in intra- and interpersonal contexts, involving other people or
not, over a particular time frame. One advantage of this type of formulation, the coding
system of therapeutic focus (CSTF), is that it

indicates what the problem is, and where to
intervene, but (being atheoretical) cannot
indicate how to do so. It cannot therefore
provide specific implications for treatment, but
it does provide a common language, and using
this it is potentially easier to find out precisely
how theories differ when put into practice.
6.01.4.4 Issue of Completeness
The formulations illustrated so far have been
kept simple for the sake of clarity, to emphasize
the point made at the beginning about the
principle of parsimony, and because they
demonstrate the point that it is never too soon
to start formulating. They are examples of
initial hypotheses. As treatment progresses they
would be likely to become more complex and
also to take more account of a person's
developmental history and the supposed underlying mechanisms.
This raises an important issue for discussion.
Many people assume that formulating is a
difficult and lengthy process, the aim of which is
to encompass, systematize, and explain all
relevant factors about a particular case. This
view can lead therapists either to bypass the
process of formulation and start treatment
straight away, or to delay the start of treatment
until they have got the picture right. Both of
these reactions cause problems: bypassing the
process makes it hard to move beyond the stage
of trial and error; interventions are selected in
the absence of a coherent underlying strategy.
This seems to be successful when the patient
responds well (as many patients do initially), but
it leaves both parties feeling confused and
unable to understand what has happened when
half a dozen sessions later progress is halted and
setbacks are encountered. It is rather like trying
to stop a car rattling by cleaning and adjusting
those parts of the engine that are most
accessible. Delaying the start of treatment is
another false economy, for many reasons: the
initial momentum provided by a fresh start and
a new encounter may be lost; the impact of being
listened to, heard, and understood by someone
new may be dissipated; and the goodwill,
advice, and new ideas derived from interacting
with a trained therapist may not be harnessed in
a way that is either helpful or informative (or
both). So, opportunities to test hypotheses may
be lost.
In an ideal world therapists, believing that ªat
some level it all makes senseº, would be able to
use their formulations to make sense of the
material presented in a particular case. But at
present complete formulations, like complete
theories, are not possible. A person cannot be

Methods: How to Construct a Formulation
summarized in a diagram. But some of their
problems and patterns of behaving can be
understood in theoretical terms and this understanding can be represented in a way that helps
to guide treatment. The complexity and accuracy with which this is done varies according to
the stage of treatment.
The emphasis on completeness that is often
found in discussions about conceptualization
may be a consequence of the historical associations between medical practice, psychiatry, and
psychology, and the common use of the word
ªtreatmentº to refer to the actions of people
trained in those professions when they are trying
to help others. The assumption is that it could be
dangerous to miss something serious or to apply
the wrong treatment. So, a complete understanding is supposedly an essential (or important) prerequisite for deciding how to intervene.
The situation is different in psychotherapy (or
psychological therapy), first because the psychological influence of one person on another
cannot be withheld (as can a medical treatment),
and then applied when ready, in a self-contained
package. Various (partially unspecified) factors
are always operating, and in psychotherapy the
ways in which these function will to some degree
be influenced by the theoretical views and
assumptions of the therapist (as well as by their
personal characteristics). The business of formulating can direct this process, clarify what is
intended, and make the way in which theories
are being applied accessible. Formulations do
not have to be complete to perform this function
but the method of working with them does have
to be in place. Second, when dealing with
psychological matters, the process of formulation is overtly interactive. Patients' comments
and reactions contribute to the process; their
opinions are relevant, and these may change
over the course of treatment.
6.01.4.5 Conceptualizing Processes of Change
Therapists seek understanding of the way in
which change takes place as well as of the way in
which problems arise and persist, and they may
also formulate this understanding in theoretical
terms. Conceptualizing the processes of change
is thus another way of relating theory to
practice, and formulations may be technical,
phrased in terms that are derived from the
particular theory being used, or metaphorical.
Technical formulations might explain how
changing reinforcement patterns would change
behaviors, how change in one person will
prompt the system around them to adjust, or
how changing patterns of defensiveness might
change opinions of the self. Examples of

19

metaphorical formulations are provided in this
section to illustrate how metaphors can encapsulate information about complex processes
that may be hard to specify otherwise. Some
examples are well known and their use has
become quite conventional, such as ªa journey
of a thousand miles begins with the first step.º
Others are created in a particular therapy
context. For example, a manager of an
engineering company, whose habitual rigidity
was exacerbated by various (personal and
industrial) crises, saw himself as ªhanging on
for dear life,º and being unable to contemplate
change. He was asked to think about how to
build a building to withstand an earthquake.
Thinking about this enabled him to reconceptualize change as a way of developing the
combination of flexibility and rigidity needed to
provide stability in difficult times. Another
relatively simple way of representing and
summarizing a complex process of change
was spontaneously developed by a woman with
longstanding problems involving low selfesteem and lack of confidence. She saw herself
as ªwobblyº and at risk of falling, as if trying to
sit on a two-legged stool. The process of change
for her was like ªputting down the third leg.º In
practice this meant many things that contributed to a sense of stability: developing new skills
and abilities, thinking about herself in new
ways, and making more respectful and open
relationships with those around her.
People often use metaphorical language to
communicate their experience of distress.
Indeed, it might be more accurate to say that
it is difficult to describe such experiences
without using metaphorÐpeople explain to
clinicians how they feel broken, trapped, fenced
in, cast adrift, close to the edge, messed up, out
of reach, cut off, high, low, and so on. Perhaps
the most common methaphors describe life as a
journey and ourselves as traveling through
different kinds of emotional weather. Patients'
understanding (or personal formulation) of the
processes involved is also reflected in the words
used to describe their experiences: ªI've hidden
myself away . . . built a protective wall around
me . . . had to harden my shell . . . can't see my
way out of the tunnel . . . waited to be rescued.º
It is hardly surprising that the processes
involved in therapy are similarly described.
Someone who came to understand the stultifying and self-destructive effects of overt compliance with the wishes of those around her,
despite her own inclinations, and the relationship of this pattern of behavior to the fear and
anger for which she was requesting help, said
that she felt as if she had spent her whole life
trying to grow flowers in her garden and cutting
off the buds before they could flower. She saw

20

Clinical Formulation

therapy as a process that would help her to
allow the flowers in her garden to bloom.
Undoubtledly, the process of developing a
shared understanding is a complex one, and
the more abstract the material considered, the
more difficult this process will be. Although a
metaphor is not a formulation, and it may
reflect only part of what is involved, using one
can help to fulfill some of the purposes of
formulation that were described above, and it
can do so with a startling degree of economy
and emotional sensitivity because it operates at
more than one level.
These examples have been chosen because
they illustrate a point not so far emphasized
about formulation, that it is a way of
summarizing meanings, and of negotiating for
shared ways of understanding them and communicating about them. When these are complex it can be helpful to use metaphor, and of
course this applies generally, not just when
formulating processes of change. A formulation
provides a source of common language, and
when this is available it can then be used to
relate a theoretical framework, at a high level of
abstraction, to practice, so as to facilitate the
process of change.
6.01.5 ACCURACY: HOW TO TELL IF A
FORMULATION IS RIGHT
Formulations can never be shown to be right
as they are hypotheses not statements of facts.
The evidence may support them or it may not,
and they should be judged according to
probabilities rather than on an absolute scale
of rightness. Like other scientific hypotheses,
formulations can only be shown, conclusively,
to be wrong. Nevertheless, practical guidelines
are useful, and a number of attempts have
recently been made to evaluate their inter-rater
reliability and predictive validity (Barber &
Crits-Christoph, 1993; Horowitz & Eells, 1993;
Persons, Mooney, & Padesky, 1995).
6.01.5.1 Criteria of Accuracy
A summary of the kinds of practical guidelines that might provide clinicians with criteria
of accuracy is given in Table 4. Unfortunately,
the fact that a formulation makes good internal
sense (provides a plausible narrative for instance) is not a guarantee of its accuracy, which
should therefore be tested out in practice. It goes
without saying perhaps that a formulation
which is simple, clear, and easy to understand,
and therefore easy to explain, is more readily
testable than one which is overly complex. One
which is more specific and low level will have

clearer implications than one which is phrased
in more general, abstract, and high-level terms.
Presenting the formulation to someone else, or
putting it onto paper, is therefore a useful and
revealing exercise.
6.01.5.2 Questions for Research
It would probably be fair to say that, of the
many questions that could be asked, few have
been studied and none have been conclusively
answered. Persons, Padesky, and Mooney
(1996) found only moderately good inter-rater
reliability of cognitive-behavioral formulations
when tapes of initial therapy sessions were rated
by a large group of therapists who had been
trained in CBT, and who varied in their level of
experience. Surprisingly perhaps, agreement
was better with respect to underlying mechanisms than in listing patients' overt problems.
Barber and Crits-Christoph (1993) found, when
reviewing the psychodynamic literature, that
when clinicians based their formulations on
preset categories, formulations were more
reliable, and in addition the predictions of the
psychotherapy process and outcome were
better. Both these findings fit with the view
that the more clearly specified the activity (as in
CBT and interpersonal psychotherapy, or when
using clearly defined conceptual categories), the
less room there is for wide-ranging, speculative
inferences, and the more agreement there is both
about particular case formulations and about
their utility.
As discussed above, there has been some
suggestion that making overall decisions about
treatment purely on the basis of a diagnosis may
be at least as useful as basing them on an
idiographic formulation. However, a diagnosis
only enables therapists to make general decisions about which set of interventions to
employ; for example, to use exposure in vivo
to help someone with a simple phobia, or those
techniques that will assist in resolving a role
dispute in a case of depression treated with
interpersonal psychotherapy. In both cases the
actual steps used will still depend on the way in
which the individual case is formulated (Markowitz & Swartz, 1997). The question as to
whether treatment that is based on a formulation is more successful than treatment that is not
is more complex than at first appears. Most
clinicians bring their theoretical knowledge to
bear in the way that they understand, and
communicate understanding about, a case.
They use covert formulations, which may not
be made overtly communicable even though
they inform and direct the process of treatment.
This happens because, once therapists are

Using the Formulation: Practical Issues

21

Table 4 Ten tests of a formulation.
1. Does it make theoretical sense?
2. Does it fit with the evidence? (symptoms, problems, reactions to experiences)
3. Does it account for predisposing, precipitating, and perpetuating factors? (both
overall and with respect to episodes of difficulty)
4. Do others think it fits? (the patient, supervisors, colleagues)
5. Can it be used to make predictions? (about difficulties, aspects of the therapeutic
relationship, etc.)
6. Can you work out how to test these predictions? (to select interventions, to
anticipate responses and reactions to therapy)
7. Does the past history fit (with respect to the person's strengths as well as
weaknesses)
8. Does treatment based on the formulation progress as would be expected,
theoretically?
9. Can it be used to identify future sources of risk or difficulties for this person?
10. Are there important factors that are left unexplained?

thoroughly familiar with the theoretical background to their work, and with the process of
map-making, the activity of formulation cannot
be wholly suspended. Once able to recognize
signs of core beliefs or core conflicts, for
example, such theoretically meaningful constructs cannot suddenly be rendered invisible
again. Formulation skills may still need sharpening, and there is certainly a need for more
and better training (Sperry, Gudeman, Blackwell, & Faulkner, 1992), especially now that
clinicians appear increasingly likely to incorporate ideas from theoretical orientations other
than their main one into their work (Messer,
1996b). The effects of working with (or without)
a formulation will remain hard to evaluate. The
more important question, in practical terms, is
whether or not a particular way of seeing things
is put to good use, successfully to do the things
that a formulation is for. The struggle is to find a
way of seeing things that helps. Although the
assumption that ªat some level it all makes
senseº still underpins much clinical work, it is
not necessary to believe that there is such a thing
as a ªcorrectº formulation. As Messer (1996a,
p. 136) says, ªAn alternative outlook is that
there is no one version of truth possible because
we largely construct our realities, which inevitably leads to multiple perspectives on that
reality. Wearing different glasses provides
different views of the world.º
6.01.6 USING THE FORMULATION:
PRACTICAL ISSUES
A formulation does not have to be correct,
but it does have to be useful. The purposes of
formulation are discussed in Section 6.01.3.
Here, three practical factors that influence
whether a particular formulation succeeds in
fulfilling its purposes are mentioned briefly.

6.01.6.1 The Value of Organizing and Clarifying
Formulating is a way of classifying information, putting it into (conceptual) boxes, and
drawing links between them. It organizes
information, treatment strategies, and the choice
of interventions, and it also clarifies understanding of a case, and therefore the meaning of
what is observed. This process has some less
obvious advantages as well as the obvious ones.
In particular, it helps therapists to see problems
and difficulties as understandable, and this
influences their attitudes and expectations. For
example, hostile or passive±aggressive behaviors
frequently create frustrations and difficulties
for therapists, especially when they persist
despite all their best efforts. Organizing and
formulating the information helps therapists to
see these as characteristic and predictable
difficulties for which they can plan appropriate
strategies.

6.01.6.2 Developing an Internal Supervisor
The process of formulation provides therapists with an opportunity to achieve on their
own many things that otherwise they would
achieve through supervision. It prompts them to
reflect about their work with individual cases,
and to rethink when progress seems blocked. It
helps them to become aware of their own
assumptions and beliefs, and to look out for
ways in which these may cause problems, such
as making it hard for them to notice, understand, or work with particular issues. It helps
them to work well with unusual cases or with
types of problems that they have not previously
encountered. In doing so it helps to build
confidence. Formulation is no substitute for
supervision but, used well, it complements and
extends itÐprovided that the formulation does

22

Clinical Formulation

not become a fixed way of seeing things that
obscures the significance of information that
does not fit.
6.01.6.3 Communicating a Formulation
Some obvious principles can be derived from
the preceding arguments: the simpler the
formulation, the easier it will be to communicate; it should be presented as a hypothesis,
not as fact; and initial guesses are worth
checking out as they can indicate whether a
particular way of seeing things is likely to be
productive.
To some degree a formulation is a matter of
judgment. It is based on clinical judgment as
well as on knowledge and facts. As judgments
about people are bound to reflect some of the
attitudes and assumptions of the person who
makes them, the question arises as to what
should be done with those judgments. Who
should be told about them? Are there people
who should not be told, or circumstances in
which they should not be disclosed?
Answers to these questions are partly determined by practitioners' ethical guidelines and
procedures for professional accountability.
They also depend partly on the theoretical
orientation of the therapist. In cognitive
analytic therapy, interpersonal psychotherapy,
CBT, and in some forms of short-term
psychodynamic psychotherapy, therapists
make their formulations explicit, and have
therefore considered carefully how and when
this should be done (Beck, 1995; Beck, Freeman, & associates, 1990; Butler & Booth, 1991;
Markowitz & Swartz, 1997; Ryle, 1995). The
method used is immensely variable, using
imagery, metaphor, diagram, or verbal explanation, presented in person or in a letter. There
is room here for creative thinking, and sensitive
adaptation of communication skills, though it
may help to specify some general principles.
Being on the receiving end of a formulation
can feel like being weighed up, evaluated, or
judgedÐlike being ªseen throughº or
ªrumbledº rather than understood. This is less
likely if the formulation is presented questioningly and collaboratively, at a time when
therapists are clear that patients are able
honestly to give feedback, and while thinking
about how to facilitate the process of feeding
back reactions in a way that is not just
superficial or polite. It is important to focus
on strengths as well as weaknesses, and to draw
out implications for change, otherwise patients
with chronic problems may conclude that ªthis
is the way that they are,º and become hopeless
about change. The language used should be

simple and jargon free. It may help to give a
small amount of information at a time and to be
ready to repeat explanations, or introduce
technical terms, as necessary. Therapists often
underestimate how much patients can themselves contribute to the process of formulation,
for instance, by elaborating details, filling in
missing links, or providing contradictory information that shows how the formulation can
usefully be adjusted.
Formulation thus goes hand-in-hand with
reformulation, and it is this, as Rosenbaum
(1996) points out, that stops it becoming a way
of ªfitting something to a known formula.º
6.01.7 CONCLUDING DISCUSSION
Formulations reflect the way in which
therapists make sense of someone else's predicament. They reflect the assumptions brought
to bear when thinking about it, the theories
learned, and the meaning made of it. However,
making sense is not the only thing that they do.
All therapists are aware that sometimes (albeit
rarely) providing a formulation can be sufficient
to bring about change. Such cases show that
formulations
do
more
than
supply
understandingÐthey enable someone to see
things differently, to reformulate, or to find a
new meaning. A business executive whose whole
career was threatened by an episode of severe
stress and anxiety was suddenly able to see
himself as engaged in a genuine struggle. It was
then legitimate, in his view, to experience
reactions indicative of both fight and flight.
His symptoms became acceptable, diminished
immediately, and he remained well over the
following six months. Of course this could be
understood in many ways: as a healthy
consequence of a reformulation, as a miracle
cure, or as a flight into health. So therapists are
also in a predicament. Most of the time only
some of the facts are available to them, whether
these are about someone's past life, their
internal experience or their present relationships, and the facts that are available are
consistent with a wide range of plausible
interpretations. Different mechanisms can be
inferred from the same event, as in the example
above, or from the same overt problemsÐthe
bather's hand movements could signify waving
or drowning. Equally, the same mechanisms
could be inferred from different problemsÐa
fear of abandonment could underlie both
hostile and dependent behavior. The skills of
functional analysis may help to advance the
process of formulation here. To end where we
began, Frank (1986, p. 343) said that ªthe best
hope of bringing conceptual order into the field

References
of psychotherapy may lie in thinking of all
psychotherapeutic enterprises as lying in the
realm of meaningsº . . . thinking, feeling, and
behavior are . . . ªresponses to the meanings of
events as much as to the events themselves.º Our
assumptions and knowledge about the ways in
which these meanings are stored, represented,
and recalled, and about the degree to which they
can be brought into awareness, will therefore
greatly influence the meaning we give to our
formulations and the uses we make of them.
Therapy can be understood in many waysÐas
managing anticipated transferences, countertransferences, and resistances; as seeking new
perspectives and using these to restructure a
belief system; as a process of constantly meeting
and adjusting to what is happening each
moment; or as a way of influencing the
contingencies that relate behaviors to their
antecedents and consequences. In all of them,
the process of formulation serves similar functions. It is useful because it helps to determine
what we, as therapists, do and enables us to
understand and to explain that better.
6.01.8 REFERENCES
Barber, J. P., & Crits-Christoph, P. (1993). Advances in
measures of psychodynamic formulations. Journal of
Consulting and Clinical Psychology, 61, 574±585.
Beck, A. T. (1991). Workshop on cognitive therapy of
personality disorders. Brighton, UK: Royal College of
Psychiatrists.
Beck, A. T., Freeman, A., & associates (1990). Cognitive
therapy of personality disorders (chap. 4). New York:
Guilford Press.
Beck, J. S. (1995). Cognitive therapy: basics and beyond.
New York: Guilford Press.
Beutler, L. E., Williams, R. E., Wakefield, P. J., &
Entwistle, S. R. (1995). Bridging scientist and practitioner perspectives in clinical psychology. American
Psychologist, 50, 984±994.
Blackburn, I-M., & Twaddle, V. (1996). Cognitive therapy
in action: A practitioner's casebook. London: Souvenir
Press.
Butler, G., & Booth, R. (1991). Developing psychological
treatments for generalized anxiety disorder. In R. M.
Rapee & D. H. Barlow (Eds.), Chronic anxiety and
generalized anxiety disorder (pp. 187±209). New York:
Guilford Press.
Butler, G., Fennell, M., Robson, P., & Gelder, M. (1991).
A comparison of behavior therapy and cognitive
behavior therapy in the treatment of generalised anxiety
disorder. Journal of Consulting and Clinical Psychology
59, 167±175.
Butler, G., & Low, J. (1994). Brief psychotherapy. In M.
Pokorny & P. Clarkson (Eds.), A handbook of psychotherapy (pp. 208±224). London: Routledge.
Clark, D. M. (1988). A cognitive model of panic attacks. In
S. Rachman & J. D. Maser (Eds.), Panic: Psychological
perspectives (pp. 71±90). Hillsdale, NJ: Erlbaum.
Clark, D. M., & Wells, A. (1995). A cognitive model of
social phobia. In R. G. Heimberg, M. K. Liebowitz, D.
A. Hope, & F. R. Schneier (Eds.), Social phobia:
Diagnosis, assessment and treatment (pp. 69±93). New
York: Guilford Press.
Dattilio, F. M. (1994). SAEB: A method of conceptualisa-

23

tion in the treatment of panic attacks. Cognitive and
Behavioral Practice, 1, 179±191.
Eells, T. D. (1996). Commentary on three case formulations of Jim. Journal of Psychotherapy Integration, 6,
119±126.
Eells, T. D. (1997). Handbook of psychotherapy case
formulation. New York: Guilford Press.
Frank, J. (1986). PsychotherapyÐthe transformation of
meanings: Discussion paper. Journal of the Royal Society
of Medicine, 79, 341±346.
Freeman, A. (1992). The development of treatment
conceptualisations in cognitive therapy. In A. Freeman
& F. Dattilio (Eds.), Comprehensive caseboook of
cognitive therapy (pp. 13±23). New York: Plenum.
Goldfried, M. R. (1995). Toward a common language for
case formulation. Journal of Psychotherapy Integration,
5, 221±244.
Goldman, R., & Greenberg, L. (1997). Case formulation in
process±experiential therapy. In T. Eells (Ed.), Handbook of psychotherapy case formulation. New York:
Guilford Press.
Hayes, S. C., & Follette, W. C. (1992). Can functional
analysis provide a substitute for syndromal classification? Behavioral Assessment, 14, 345±365.
Hayes, S. C., Follette, V. M., Risley, T., Dawes, R. D., &
Grady, K. (1995). Scientific standards of psychological
practice. Reno, NV: Context Press.
Haynes, S. N., Uchigakiuchi, P., Meyer, K., Orimoto, L.,
& Blaine, D. (1993). Functional analytic causal models
and the design of treatment programs: Concepts and
clinical applications with childhood behavior problems.
European Journal of Psychological Assessment, 9,
189±205.
Horowitz, M. J. (1989). Relationship schema formulation:
Role-relationship models and intrapsychic conflict.
Psychiatry, 5, 260±274.
Horowitz, M. J., & Eells, T. D. (1993). Case formulations
using role-relationship model configurations: A reliablity
study. Psychotherapy Research, 3, 57±68.
Horowitz, M. J., Eells, T., Singer, J., & Salovey P. (1995).
Role-relationship models for case formulation. Archives
of General Psychiatry, 52, 625±632.
Klerman, G. L., Weissman, M. M., Rounsaville, B. J., &
Chevron, E. S. (1984). Interpersonal psychotherapy of
depression. New York: Basic Books.
Luborsky, L., & Crits-Christoph, P. (1990). Understanding
transference: The CCRT method. New York: Basic
Books.
Markowitz, J. C., & Swartz, H. A. (1997). Case formulation in interpersonal psychotherapy of depression. In T.
Eells (Ed.), Handbook of psychotherapy case formulation.
New York: Guilford Press.
Messer, S. B. (1996a). Concluding comments: Special
section: Case formulation. Journal of Psychotherapy
Integration, 6, 135±137.
Messer, S. B. (1996b). Introduction to special section: Case
formulation. Journal of Psychotherapy Integration, 6,
81±83.
Minuchin, S. (1974). Families and family therapy. Cambridge, MA: Harvard University Press.
Nicholson, S. (1995). The narrative danceÐa practice map
for White's therapy. Australian and New Zealand Journal
of Family Therapy, 16, 23±28.
Norcross, C. A., Alford, B. A., & DeMichele, J. T. (1992).
The future of psychotherapy: Delphi data and concluding observations. Psychotherapy, 29, 150±158.
Padesky, C. A., & Mooney, K. A. (1990). Presenting the
cognitive model to clients. International Cognitive
Therapy Newsletter, 6, 13±14.
Perry, S., Cooper, A. M., & Michels, R. (1987). The
psychodynamic formulation: Its purpose, structure and
clinical application. American Journal of Psychiatry, 144,
543±550.

24

Clinical Formulation

Persons, J. B. (1989). Cognitive therapy in practice: A case
formulation approach. New York: Norton.
Persons, J. B. (1993). Case conceptualisation in cognitive
behavior therapy. In K. T. Kuelwein & H. Rosen (Eds.),
Cognitive therapy in action. San Francisco: Jossey-Bass.
Persons, J. B., Mooney, K. A., & Padesky, C. A. (1995).
Interrater reliability of cognitive-behavioral case formulation. Cognitive Therapy and Research, 19, 21±34.
Rosenbaum, R. (1996). Form, formlessness and formulation. Journal of Psychotherapy Integration, 6, 107±118.
Ryle, A. (1978). A common language for the psychotherapies. British Journal of Psychotherapy, 132, 585±594.
Ryle, A. (1990). Cognitive-analytic therapy: Active participation in change. Chichester, UK: Wiley.
Ryle, A. (Ed.) (1995). Cognitive analytic therapy: Developments in theory and practice. Chichester, UK: Wiley.
Salkovskis, P. M. (1996). The cognitive approach to
anxiety: threat beliefs, safety-seeking behavior, and the
special case of health anxiety and obsessions. In P. M.
Salkovskis (Ed.), Frontiers of cognitive therapy. New
York: Guilford Press.
Schulte, D. (1997). Behavioural analysis: Does it matter?
Behavioural and Cognitive Psychotherapy, 25, 231±249.
Schulte, D., KuÈnzel, R., Pepping, G. & Schulte-Bahrenberg, T. (1992). Tailor made versus standardised therapy
of phobic patients. Advances in Behaviour Research and
Therapy, 14, 67±92.
Seligman, M. E. P. (1995). The effectiveness of psychother-

apy. American Psychologist, 50, 965±974.
Silbertschatz, G., Fretter, P. B., & Curtis, J. T. (1986). How
do interpretations influence the process of psychotherapy? Journal of Consulting and Clinical Psychology, 54,
646±652.
Sperry, L., Gudeman, J. E., Blackwell, B., & Faulkner, L.
R. (1992). Psychiatric case formulations. Washington,
DC: American Psychiatric Press.
Stricker, G., & Trierweiler, S. J. (1995). The local scientist;
A bridge between science and practice. American
Psychologist, 50, 995±1002.
Turkat, I. D., & Maisto, S. A. (1985). Personality
disorders: Application of the experimental method to
the formulation and modification of personality disorders. In D. H. Barlow (Ed.), Clinical handbook of
psychosocial disorders: A step by step treatment manual.
New York: Guilford Press.
White, M. (1989). The externalizing of the problem and the
re-authoring of lives and relationships. Adelaide, Australia: Dulwich Centre Newsletter, Summer, 5±28.
Wilson, G. T. (1996). Manual-based treatments: The
clinical application of research findings. Behaviour
Research and Therapy, 34, 295±314.
Wilson, G. T. (1997). Treatment manuals in clinical
practice. Behaviour Research and Therapy, 35, 205±210.
Wolpe, J., & Turkat, I. D. (1985). Behavioral formulation
of clinical cases. In I. D. Turkat (Ed.), Behavioral case
formulation (pp. 5±36). New York: Plenum.