… who would fardels bear,
To grunt and sweat under a weary life,
But that the dread of something after death,
The undiscover’d country from whose bourn
No traveller returns, puzzles the will
And makes us rather bear those ills we have
Than fly to others that we know not of?
—William Shakespeare, The Tragedy of Hamlet, Act 3, Scene 1
Engaging in assessment brings risk for families. Risk of stepping into the unknown, into clinging to the problematic-known rather than the dread of what might not be so sure, so therapeutic sensitivity is required. An effective systemic assessment requires openness (retaining an awareness of the very provisional nature of all our ideas – that these are socially constructed rather than arising from absolute ‘truths’), and that the assessment process is embedded within a collaborative relationship, without which therapeutic relevance and trust are unlikely.
Safe uncertainty is the best possible outcome of a systemic assessment. Certainty contains the possibility of closed-down thinking and unsafe conclusions; uncertainty that feels ‘unsafe’ is similarly unhelpful: producing anxiety, disorganisation, and disruption – which usually induces a return to closed-down position-taking, and conclusions of spurious precision.
In systemic work ‘safe uncertainty’ is upheld by a three-legged stool:
1. a ‘right first time’ approach – remembering the importance to the family of resolving the problem for which referral was sought
2. a developed formulation – a systemic shared understanding of the issues uniquely brought by a family to their service-contact
3. a structured assessment process that leads to a clear formulation – a process, capable of replication, by which that understanding was reached i.e. a reliable process and independent of therapist-bias.
Each of these three is explored in turn. A more detailed account of ‘safe uncertainty’ is available elsewhere, likewise accounts of ‘formulation’. All the three ‘legs’ are compatible with key principles of CAPA (Choice and Partnership Approach) in child and adolescent mental health work, as it similarly seeks a ‘right first time’ approach. Given its widespread application in CAMHS, where systemic work is being increasingly adopted, we distinguish between our approach to systemic assessment and how CAPA considers this should occur.
Comparison with the Choice and Partnership Approach in CAMHS: the main difference lies in CAPA’s focus how as a team CAMHS can best function, from ‘front of house’ and the avoidance of bottlenecks (long external or internal waiting times), to ensuring clinician accountability and team flexibility (www.camhsnetwork.co.uk).
The term ‘Choice’ was introduced to encompass both the opportunity provided for newly referred families to choose when to meet a clinician for the first time (which some teams consider an unhelpful imposition on their timetabling), and – surely uncontroversial – that families must a say in what should happen next. However, considerable variation has been observed in the implementation of CAPA, in which lies its main difference from our approach: the breadth of its guidelines for team functioning, and its comparative absence of detail on case-work implementation.
By describing the clinician as an ‘expert facilitator’, and its reliance upon generic ‘core-skills’ for the initial assessment (for that is what the ‘Choice’ meeting is) CAPA clearly seeks to avoid prescriptive approaches that might inadvertently substitute one particular colonising discourse by another. For example, to avoid stating that a developmental enquiry and family-life cycle history MUST be included, replacing the psychiatrist’s previously immutable role for mental state examination, or a psychologist’s for elucidating cognitive schema, etc. The ‘Choice’ meeting could of course be concluded by agreeing that one or more of these four tasks was now necessary (i.e. entry to Partnership with someone who can provide specialist skills).
In consequence, ‘Choice’ appointments vary enormously, from screening and signposting or solution-focused work to detailed assessments, no doubt depending upon the context of the appointment (e.g. community-based or at a specialised clinic) and the complexity of the case. Many clinicians have also interpreted the efficiency-drive of CAPA to assume that CAMHS teams provide no more than an hour for the ‘Choice’ meeting, the large number any team is required to undertaken means that most of its members to contribute (and therefore of necessity a considerable reliance upon relatively junior staff), and that a usual maximum of 5-6 appointments can be allowed for Partnership sessions.
Those responsible for its development (Steve Kingsbury and Ann York) consider these beliefs misinterpret CAPA. These are among the nineteen myths examined and refuted on their website (www.camhsnetwork.co.uk/Intro/capamyths.html).
Integrated Family Therapy: the detailed case-work approach to assessment provided below should result in far less variation in implementation and fewer misinterpretations. The initial encounter with a family would invariably include most of the detailed assessment procedures described below.
Taking a CAMHS context as the example, this should invariably include a developmental enquiry and family life cycle history and some understanding of the parents’ own experience of being parented. And in most contexts
- some understanding of the inner world of the identified patient/client if not of other family members present
- time to explore meaning and cognition (including where some impairment of mental functioning is present)
- and an understanding reached of if and how family functioning co-relates with the problems that presented.
In other words, a formulation – developed with the family and agreed before the meeting is completed. We argue that it can never be achieved in an hour. Nor do we believe it is possible within 90 minutes without considerable skill, knowledge, and experience.
Eyes and ears can’t be distracted by ongoing note-keeping during the meeting, nor does that facilitate the quality of engagement we encourage. With greater experience comes a capacity to remember the details of what occurred, and if an additional 30 minutes is allowed for note-making and letter writing after the meeting closes, a record is then available of all salient details. In secondary care contexts and all other specialist centres, we make no apology for advising front-loading families’ service contact with experienced workers.
The approach is no more elitist or prescriptive than what any of the population would hope was the standard of care exercised by a High Street optician (the difference of course lies in a patient’s profoundly different participation in assessment!).
A ‘right first time’ approach
This acknowledges the importance any family attaches to resolving the problem that had precipitated their referral. Meeting their reasonable expectations that their difficulty may be dealt with there and then rather than being referred on (albeit not always possible), nor lost sight of as other issues emerge during assessment.
It involves a paradox: to respect the central importance to a family of the presenting problem yet escaping its centre of gravity in order to gain a perspective on it. Both family members and the therapist may bring ideas about what’s wrong to the meeting; some pre-conceived ideas will prove ‘unsafe’, providing a false sense of certainty about what underlies their difficulties, the resolution of which is the raison d’etre for their service contact.
The opening few minutes of the initial family consultations is vital, as – unless later corrected – it provides a particular direction that either facilitates or impedes future understanding. The 1985 publication entitled this first phase of a consultation, ‘orientation’. As earlier described, this essential phase allows all involved to learn from each other what expectations are being brought to the meeting, how it will proceed, what it will cover, the affordances and constraints of confidentiality and the sought-for outcomes.
No experienced therapist skips this step, which may take only a few minutes but occasionally requires 50 minutes or more before misunderstandings are resolved and family members become sufficiently at ease; for safe uncertainty at least initially to be secured; and to facilitate open exploration of all salient issues (see below).
A developed formulation
Described as a 4-layer ‘expert’ model in the 1st edition, this allowed a dynamic integration of the many different observations and findings obtained from an assessment of a family in clinical practice.
The 1st layer: constituted data that emerged from direct questioning, about the presenting problem, and how it is embedded in the family, as further described in terms of communication, roles, problem-solving styles, emotional involvement and how this is expressed.
The 2nd layer: integrated data that emerged from therapist-observation, for example recursive behavioural patterns (evident from family member accounts or at interview) and family beliefs.
The 3rd layer: (‘links with the past’) involved inference, drawn from this accumulated data and enriched by the history taken (of life-events and the family life cycle).
The 4th layer: ‘Why now?’
Trainees in the McMaster Model during the 1980’s were encouraged, in both their practice and in their later writing up of sessions, to carefully follow the sequence described above. Social constructionism and systemic methods of enquiry since have blurred the boundaries between layers, and the sequence that should be followed.
Of course now there are instead just different ‘shoulds’. In particular, the value now attached to curiosity-based questions that place any focus of enquiry, and family responses to it, in a relational context.
For example, on a father’s inconsistent parenting: how non-judgemental curiosity facilitated an understanding that Paul – brought up by emotionally distant parents – regains closeness to his wife (that he otherwise struggles to attain) by berating their son for behaviour that he otherwise often seems to condone. The non-judgemental approach had minimised defensive responding, enabling his own and emerging family agreement about a major recursive behavioural pattern.
In the original Integrated Model, social science and the observational studies of non-clinic family functioning previously undertaken by the McMaster group (then in Montreal) had contributed to its 1st layer, entitled ‘surface actions’ (e.g. family roles, emotional involvement, communication, etc.). Very different sources contributed to the 2nd layer, entitled ‘dysfunctional transactional patterns’, many involving psychodynamic concepts. We illustrate the application of this framework to the fictional family of Willy Loman, the central character of Arthur Miller’s Death of a Salesman. As a clinical example of our approach to formulation, we include a case of anorexia nervosa.
We are not alone in emphasising the importance of a comprehensive formulation based on integrating theoretical models. It provides the basis of cognitive-analytic therapy, and for family problem-resolution as Breulin and Pinsof have described, where a training programme based on these principles has been developed at the Family Institute, Northwestern University (Illinois USA). However, we are unaware of any other approach that integrates the four models we describe. Nor do any currently include a neuroscience and epigenetics perspective, despite its undeniable salience to a full understanding of past trauma, and its invariable importance for effective working with anorexia nervosa.
A structured assessment process
As noted elsewhere, this process does not have the formality of a pre-ordained structure, facilitating manualisation. Instead, we describe the broad steps usually undertaken to complete an assessment to arrive at a formulation which makes shared-sense of the difficulties that had prompted the referral.
Initial data-gathering and problem description: once the orientation has been completed, an assessment begins with the presenting problems. For example, how long these had been present, the antecedents and consequences of each, who is most directly involved and who less involved, others who are aware of the problems, and how all this is embedded in current family relationships. This may not vary very much even if quite different family problems had precipitated the referral.
Concrete examples are sought and the relational contexts of each explored (tracked), including who gets drawn in to help or problem-solve. ‘Data saturation’ – a concept drawn from qualitative research methodology – provides an indication of how far curiosity should extend: to move on to other topics when no ‘new’ additional information has emerged, when examples of more of the same would be given (e.g. after summarising what has been said, ‘have I got that right?’ … is that what happens? are other problems just like that? or are there other ones that are a bit different? … can we talk about those …?’).
This demonstrates how two types of enquiry occur during assessment: a ‘search’ mode – when something is explored in greater depth, and a ‘scan’ mode of enquiry – when curiosity is extended to go beyond a particular focus, although perhaps (but not necessarily) using that focus as a ‘jumping off’ point to other aspects of family functioning.
Being able to change topic, to move the session on, is as vital as being able to persist with a line of enquiry to ensure that a comprehensive account of family life is possible within the time constraints of the consultation. The original McMaster model emphasised the understanding of family problem-solving as central to any family therapy assessment. It provides an indication of family roles, beliefs, stories, communication styles, and emotional involvement. In short, assessing the adaptive flexibility that became established after a couple got together and which, through ‘thick and thin’, has continued or altered over the years before referral took place. These same resources will underpin a family’s future problem-solving, so enhancing these and reflecting back on them as intervention concludes is a vital aspect of leaving the family.
Managing time and the balance between problem-definition and constructing hope and future possibilities: the phase of an assessment described above is likely to take between 20-40 minutes in order to provide a thorough initial understanding of a family’s communication, problem-solving, values and beliefs, their closeness, of who else is involved in key aspects of family life and how (relatives’ care, neighbourly concern, a teacher’s interest, etc.). Robin Skynner termed the latter, the ‘minimum sufficient network’ of people whose involvement either helps to make sense of the problems and/or helps resolve them.
As we indicate elsewhere, it is important to recognise and expand ‘problem-saturated’ accounts. So as well as clarifying what the family have said, we ensure that exceptions have been talked about too, seeking indications of where resilience lies or has been found.
In summary, unlike completion of the assessment described below (which may differ considerably between one family and another; clarification of issues becoming increasingly shaped by the unique aspects each family), initial data-gathering is likely to vary rather less.
Problem-clarification: this is likely to increasingly occur during the initial phase of assessment, as therapist’s curiosity becomes more and more led by the unexpected in what is said by a family member. We use the case of Dana as an illustration.
This case had been developed to demonstrate the rather different directions that could be taken to explore family life, After her comment ‘It’s OK being close to your Dad but you want to be close to your Mum as well’, a McMaster family therapist might explore whether the issue Dana had raised was one that others had recognised was a problem for her, were in agreement with its significance, what discussion had taken place. A structural therapist might have regarded her remark as evidence of a teenager triangulated between her parents. A psychodynamic therapist might cross-check between members on how the family had felt on hearing her plea, and wonder how that might correspond to past internal-world situations that were perhaps now being re-enacted. A social constructionist might be quicker on picking up what they thought had been a wish rather than a complaint, opening up family stories around closeness and talking. A constructivist would dwell longer on the family’s positive experiences of closeness, in order to later construct new realities of personal competence and what their best hopes might be. We, using the Integrated model, will be interested in all of these positions but would, whilst internally acknowledging them, privilege one more than another depending on a judgment made of the family style, engagement and hope of outcomes from therapy.
Only 30-40 minutes might remain to complete the assessment as described below. Allowances might be made for any subsequent within-team discussion and family responses; for the additional time required toward the end of the consultation for summarising how together they made sense of the problems that brought the family to therapy; and to agree on the broad outline of a possible way forward.
Building a comprehensive picture of family functioning: by comprehensive we do not imply that the questioning-style of therapist must be very broadly based. For example, returning to the brief case-example of a father’s inconsistent parenting, once what lay underneath Paul’s behaviour was recognised, building a comprehensive picture of his family might privilege several aspects: an exploration of issues related to closeness; differences and similarities of individual needs and how emotional involvement was expressed and conflict dealt with; and focus on family sub-systems, alliances, limit-setting and other boundary issues.
The foundations of this were established within the first part of the consultation, and its aim had been openly stated during the orientation. Any family doubts about going beyond a detailed understanding of the referred problems should again be addressed, as its purpose is to make the best sense of these problems i.e. it’s helpful to fully harness the family’s experience of themselves as individuals and as a family. Furthermore, without that we may inadvertently under-use the resources that each family member could bring to help future problem-resolution.
Thinking on your feet: in practice no consultation can include – as Dana’s case-description did – the equal application of all perspectives in the integrated model. A comprehensive understanding of the family may be sought but not everything said, observed, or experienced can be clarified and agreement reached.
Dana’s case-description had been constructed to allow a formulation that drew upon all perspectives (where none of them were ‘wrong’, all had thickened the description). It was not intended to reflect the wide variety of family problems referred to health and social-care agencies. Few of these consultations are likely to end with similar formulations, irrespective of whichever perspective had been adopted.
So the best use of the time remaining is vital. Hard choices are sometimes necessary. For example, to postpone some enquiries whilst deepening others, selectively drawing upon a few of the approaches that contribute to the integrated approach, always guided by what seems to be ‘the heart of the problem’ presented, rather than by the therapist’s preferred perspective for completing an assessment.
Links with the past: a comprehensive picture should nevertheless invariably include a conversation about the family’s past. This may be prompted by a current issue that seems particularly significant (e.g. fear of anger or longing for closeness), often a useful short-cut to family history. Where such prompts don’t present, simply taking the step of asking about the family history, both inter-generational (using a genogram) and exploring family life cycle can facilitate openings. These then prompt a post-hoc discussion of their possible significance and meaning to the family, which may not be able to be fully concluded in the time remaining within an initial assessment.
One of the advantages of any well-defined perspective is to allow a therapist to more readily recognise, as time begins to run out, how much the picture of the family that has emerged ‘fills out’ the details of what that perspective would ordinarily expect. A therapist is more able to recognise the ‘known unknowns’ i.e. what still remains quite unknown or uncertain and to then consider whether that matters. Then, if so, how important that might be. The very few remaining minutes might be sufficient to explore these in a preliminary way, or if not to at least acknowledge what hasn’t been talked about. Families generally appreciate that acknowledgement and may flag up issues they now recognise are relevant to what had been talked about.
Remaining unknowns: may include unknowns about quite basic aspects of current family functioning as well as of a family’s past experience. For example, influenced by the McMaster model – how practical matters in the house are dealt with, family holidays decided, school decisions made? From a structural perspective – what has not yet been learnt about sub-system strengths, or competencies outside of the family? From a psychodynamic perspective – an insufficient understanding of those anxieties that are more or less readily tolerated or how the family enjoys closeness. From neuroscience and epigenetics – that there had been absence of any clear focus on current mood or that no family history had been sought of any specific mental health difficulties. Such matters can be briefly reflected upon and those of possible significance flagged up for future discussion.
Questioning and Curiosity
As the examples of Dana and of Paul make clear, any exercise of ‘search’ and ‘scan’ mode of questioning beg further questions, on family communication, closeness, shared stories, roles, challenging past events for example. These may not have been openly discussed before within the family, so very often something quite new in the family is occurring, which Dan Siegel calls opening a neurological plane of new possibility for family members. The permission sought in Orientation to allow such curiosity therefore was a vital step, reflecting a respectfulness that is all the more necessary as no assessment can be without the possibility of something in the family changing as a consequence of it. For example, better understanding Dana’s remark or Paul’s inconsistent parenting led directly to marital issues that each of the couples might not have expected and about which going back close to impossible.
Elsewhere we have described our diminishing emphasis upon task-setting in this edition. This has occurred with greater recognition of how much of significance can begin to change simply as a result of a systemic assessment (a detailed example of that is provided elsewhere. This may contribute to the concluding formulation but however compelling early change may seem, it does not obviate the benefits of a formulation that pulls aspects of family functioning together.
Questions as Intervention: any relational question contains the possibility of an intervention, emphasised by Karl Tomm. What seems to be a linear question to one family member (e.g. when did that last happen? who did you tell about it? how did you feel when that happened? what did you think when you heard …? When did you last have a nightmare about …?) influences other family members because – even if none of the above questions were followed up by a similar enquiry of another – each question had been raised within a family context, so it was heard by others. As a result, hearing questions asked and answered prompted thoughts and feelings in others (even if undisclosed). Systemic ‘neutrality’ for family members in therapy is therefore impossible.
Many authors have provided accounts of questions that are manifestly relational (Selvini, Boscolo, Cecchin and Prata, 1978; Karl Tomm 1986; and those earlier described by Cleghorn and Levin at McMaster). A detailed account of those of most direct relevance to the Integrated model is provided elsewhere, so here we simply acknowledge their value in problem-clarification, in widening the enquiry for an increasingly systemic perspective on what had been presented as the problem necessitating referral, and to deepen emerging family stories – allowing meaning to be explored and its significance for their current life together.