Clinical scenario

A community-dwelling 45-year-old lady presented to secondary care services (mood, anxiety and personality clinic). Established previous diagnoses included attention-deficit hyperactivity disorder (ADHD), bipolar affective disorder (BPAD) and emotionally unstable personality disorder (EUPD).

At the time of the first interview she presented with multiple difficulties, evidence of minimal response to pharmacological treatments and historical difficulty engaging with psychological services. The most distressing complaint involved a recent episode lasting around 2 months when she had felt elated and more impulsive, with an increased level of activity and risk-taking behaviours. During the episode she had travelled abroad and become involved in a short-term relationship with a much younger man who was financially exploitative and had a history of criminal behaviour. She also reported experimenting with different illegal substances and incurring significant financial loss after she had lent a large sum of money. The experience as a whole had a negative and distressing effect on her life. She reported feelings of guilt and shame, which resulted in her feeling low in mood, but with some variability. On examination, she was objectively euthymic with a reactive and appropriate affect. There was also a hint of cognitive acceleration, which she described as speeding and agitated thoughts with no useful purpose, observed as slight pressure of speech. Her risk fluctuated throughout the months she spent with the team. At times, she demonstrated an ability to think and plan ahead with strong protective factors, but also reporting distressing thoughts of self-harm, which she had not acted on.

Initial diagnostic considerations

Diagnostic features suggestive of a primary mood disorder¹ included elation, increased physical and mental activity, and overspending, all of which appeared to occur during a distinct episode, resulting in a severe or complete disruption of normal life with cognitive acceleration on examination. However, there were also features suggestive of EUPD – presenting in emotional crisis following an episode which could be interpreted as ‘impulsive’, i.e. without consideration of the consequences and possibly due to lack of self-control involving an intense relationship along with suicidal ideation. Overactivity and impaired inattention of early onset, evident in more than one situation, are cardinal features of hyperkinetic disorder (attention deficit) and could have also explained some of the behaviours and symptoms experienced during this episode. It would be a plausible hypothesis that with an established diagnosis of ADHD, changes in mood might have exacerbated symptoms of ADHD, resulting in a similar behaviour.

There were many overlapping features of each condition; in this case including impulsivity, overactivity, disinhibition, mood instability, lack of self-control, relationship difficulties, and thoughts of self-harm and suicide. One of the main difficulties when dealing with such complexity is trying to establish whether each symptom is a core feature of a disorder or the result of a combination of features of different disorders. The consideration of both the aetiological nature and severity are crucial elements of the design of the treatment.

Questions regarding her presentation

• • Is the above psychopathology the result of one psychiatric disorder or a combination of conditions with overlapping features?

• • What should be the focus of the follow-up sessions?

Summary box

• • Diagnostic categorisation is a typical stage of the medical model. Nevertheless, it is important to consider what is helpful to both the patient and the clinician when symptoms and experiences are categorised.

• • What can improve diagnostic specificity with conditions such as EUPD, ADHD and BPAD, which have significant phenotypic overlap? Is it really true that many different diagnostic entities can coexist, or are we rather describing one dynamic cluster of situation-specific, overlapping phenotypic variants?

• • Patients’ symptoms and experiences can be communicated and interpreted differently based on factors such as cognitive state, social and cultural background, current mental state, etc. Can we ever fully understand what the patient is experiencing?

• • How do we use phenomenology to make hypotheses and formulate clinical problems?

• • How do we practically approach complexity and comorbidity in community psychiatry?

The usefulness and limits of diagnostic categories

Criticism of standard psychiatric taxonomy remains vigorously debated and theoretically unresolved.^{Reference Moses2,Reference Nesse and Stein3} Interrater agreement statistics for psychiatric diagnostic clarification remain poor, despite DSM and ICD revisions, and comorbidity rates within psychiatry are high, which could be reflective of this.^{Reference Moses2}

A categorical approach to diagnosis serves a valid purpose. The presence or absence of symptoms helps to formulate diagnoses and informs clinical management.^{Reference Nesse and Stein3,Reference Sedler4} It offers objective, operationalised criteria allowing for observation, analysis, research and effective communication of information, but it also relies heavily on accurate information gathering, interpretation and diagnosis, all of which can be influenced by many factors. An assessment based purely on operationalised criteria neglects many of the subtleties and complexities of psychiatric entities; it also fails to consider overlapping similarities, as well as underlying mechanisms and aetiology.^{Reference Moses2}

The difficulty with categorisation in psychiatry is trying to draw clear boundaries between disorders with significant heterogeneity (symptom profiles can differ significantly among patients diagnosed with the same psychiatric disorder).^{Reference Moses2} There may well also be transdiagnostic experiences that cut across diagnoses, or psychopathology that falls between existing categories. Adopting a longitudinal clinical approach allows us to observe unfolding factors associated with many conditions and the interplay between these. The phenomenological method can also be used to clarify autobiographical accounts and generate hypotheses with a focus on formulation.^{Reference Ratcliffe5,Reference Seising6}

We have to accept that psychiatric nosology does not always map onto people all that neatly. In fact, it has been shown that important features of commonly diagnosed psychiatric disorders, such as major depression, are not always captured by DSM criteria.^{Reference Kendler7} This is understandable given the complex nature of psychiatry and medicine on the whole, and a more realistic expectation of the role of categorisation may be needed.^{Reference Nesse and Stein3} A complementary approach, with an understanding of the usefulness of diagnostic categories, as well as a longitudinal approach and phenomenology as clinical tools, is necessary.

The literature on the shared features and comorbidity of BPAD, EUPD and ADHD

Research points to significant overlap between personality disorders, mood disorders, and developmental disorders such as ADHD.^{Reference Magill8} EUPD, BPAD and ADHD share features of impulsivity, mood lability and relationship problems. Clinicians may be reluctant to diagnose EUPD because it is perceived as ‘untreatable’.^{Reference Paris9} Longitudinal studies have found that EUPD diagnoses (despite being pervasive by definition) are not necessarily stable, and evidence-based treatment is available.^{Reference Gunderson10} The boundary between bipolar, particularly type 2, and EUPD is particularly porous: EUPD is often diagnosed alongside bipolar type 2. Zimmerman and Morgan^{Reference Zimmerman and Morgan11} argue that overdiagnosis of bipolar is often due to misrecognition of EUPD symptoms. Regarding the comorbidity of ADHD and BPAD, Youngstrom and others^{Reference Youngstrom, Arnold and Frazier12} demonstrate that distinguishing between ‘true’ and ‘false’ comorbidity is a complex and shifting process. EUPD symptoms can thus be conceptualised as a phenotypic variant of BPAD, related to developmental factors.^{Reference Perugi, Angst, Azorin, Bowden, Vieta and Young13}

Recent research indicates an association with similar biological mechanisms across these disorders. For example, there is evidence of hypothalamic pituitary axis dysfunction and childhood psychosocial adversity in ADHD; the same risk factors are also associated with BPAD and EUPD.^{Reference Kulacaoglu, Solmaz, Ardic, Akin and Kose14,Reference Richard-Lepouriel, Kung, Hasler, Bellivier, Prada and Gard15} There is evidence of genetic vulnerabilities to a range of mental and developmental disorders, including ADHD and BPAD.^{16–Reference Kolla, Meyer, Sanches and Charbonneau18} Childhood difficulties and early traumatic life experiences have also been described in a number of psychiatric conditions including EUPD, BPAD, major unipolar depression and ADHD, suggesting common underlying aetiological processes.^{Reference Carr, Martins, Stingel, Lemgruber and Juruena19}

The current research on these psychiatric conditions draws attention to shared aetiological pathways and suggests validated risk factors, some of which are potentially modifiable through available or novel treatments. The gap in the literature is the lack of systematic efforts to link specific risk and causal factors with the phenomenological level. This would help to further discriminate between these entities and hopefully enable the design of clinically meaningful and useful interventions.

Assessment of mental symptoms in psychiatry

The phenomenological approach is an indispensable clinical tool that can help with assessment and interpretation of mental symptoms. It allows a detailed enquiry into their patterns of emergence, their unfolding over time and their relationship with aspects of personality. In contrast with a symptom checklist approach, phenomenology favours a dialectical approach which is semi-structured and supports the patient's narration of their personal experience. The aim is not a merely descriptive account of the patient's experiences (descriptive psychopathology); it can also offer an explanation through seeking to reveal two things: (a) the complementarity of different aspects of the patient's experiences and expressions; and (b) the temporal interdependence of distinct phases in the development of the patient's lived world.^{Reference Gallagher and Schmicking20}

Take the example of affective lability or mood instability. This can take various descriptive forms, as the surrounding affect or mood can include valence, intensity, frequency of shift, rapidity of rise-times and return to baseline, reactivity to psychosocial cues and the extent to which there is overdramatic expression.^{Reference Broome, Saunders, Harrison and Marwaha21} Following a phenomenological approach (complementarity and development over time), it can also be linked with various other types of experiences, such as perception, bodily feelings, self-awareness, motivation and/or those brought about by social-cognitive processes. In EUPD, symptoms such as mood instability tend to present as temporary occurrences contingent upon situational stressors, e.g. perceived rejection and abandonment^{Reference Miskewicz, Fleeson, Arnold, Law, Mneimne and Furr22} and negative affect.^{Reference Law, Fleeson, Arnold and Furr23} Whereas there is overlap with BPAD in the emotion sector, experiences and expressions in EUPD also contain failures in self-appraisal, introspection and empathy (hypermentalisation).^{Reference Gunderson, Fruzzetti, Unruh and Choi-Kain24} Regarding depression, emerging studies suggest that there are similar self and social cognition features, but they rather take the form of hypomentalisation.^{Reference Bateman and Fonagy25} It can also be argued that in EUPD these are rather enduring trait states and probably more pervasive than in other disorders. People with depression can either have a stable low mood or also exhibit mood instability or irritability.^{Reference Bowen, Peters, Marwaha, Baetz and Balbuena26} The subjective experience of emotions in EUPD can have characteristics such as fusion of emotions and self and/or undifferentiated emotional experience, regularly taking the form of dissociation. It is by paying further attention to the unfolding of personal experiences that we can distinguish between depression, especially mixed affective states, and/or EUPD. In depression and mania, mood states are closely associated with cognitive states: a self-denigrated assessment that ‘I am worthless’ in depression, and an inflated sense that ‘I am exceptional’ in mania.^{Reference Radden27} There has not been much work on phenomenology of mood instability in ADHD and generally not much theorising about mood in clinical psychiatry.

Psychiatry and complexity

The phenomenological method is compatible with neuroscience (see work on aberrant salience and formation of delusions^{Reference Kapur28}) and can enrich both clinical practice and empirical enquiry. However, one of its main limitations is that it is a laborious, indirect and fallible inferential process that cannot give a conclusive account of psychopathological kinds unless it is combined with other levels of interpretation (risk factors, longitudinal observation, diagnosis, response to treatment, etc.).^{Reference Oulis29} Psychiatry is generally characterised by complexity. A framework for understanding complex psychopathology is necessary in order to organise clinical information, plan treatment and select interventions based on both evidence and values.

A pluralistic approach, understanding the limits of diagnosis and complementing it with longitudinal observations, and using phenomenology as a method can enrich clinical practice. On a more theoretical level, which also concerns the way research can be integrated into practical psychiatry, Kendler^{Reference Kendler30} has suggested the following framework, which comprises three main guiding principles: (a) psychiatry should aim to clarify risk factors regardless of level, including those assessed using imaginative understanding and phenomenology, with careful attention to causal inference; (b) research will require tracing causal pathways across levels downward to biological neuroscience and upward to social factors, thereby elucidating the important cross-level interactions; and (c) clinicians and researchers should aim to design ways of tracing the effects of these causal pathways back up into the mental realm, moving from the level of causal explanation to that of clinical understanding.

Practical management

The management of complex patients with multiple diagnoses and/or comorbidities normally takes place in a multidisciplinary secondary care setting. Assessment should be patient centred, focusing on the nature of the presentation, the patient's values and any immediate risks, what help is being sought and whether this is available. This will inform diagnostic formulation, hypotheses and clinical investigations, and an agreed treatment plan.

Repetitive information gathering has many advantages in observing symptoms and experience over time, which is useful in conditions such as BPAD and EUPD. We do, however, need to be mindful that some patients may question the purpose of this. A clear explanation of the process of the longitudinal, in-depth assessment can help build alliance and thus improve engagement. The clinicians involved are also supported to maintain consistency (number, frequency and type of follow-ups are discussed in multidisciplinary team meetings) but also to be flexible and able to adjust to the patient's needs. This type of input is provided by the short-term community mental health team (provision of care for a maximum of 3 months), which specialises in lengthy assessments, formulations and brief interventions. The short-term team has interface links with various other teams (home treatment team, substance misuse team, long-term teams and a range of specialist services).

One of the main diagnostic/assessment tasks is to find ways to test the primacy and/or severity of a cluster of BPAD, EUPD or ADHD symptoms. For a trainee, it can be difficult to be confident in making correct diagnoses in complex cases. It is therefore important to seek help and advice where necessary, which can be through informal discussions with colleagues and multidisciplinary meetings. This provides alternative perspectives on a case as well as varied professional expertise. If direct contact with a patient is needed, clinical supervisors can offer support with this.

Tools that can help test diagnostic hypotheses in relation to EUPD and BPAD include the Structured Clinical Interview for DSM-5, a semi-structured interview guide that can be used to screen for personality disorders, and the Mood Disorder Questionnaire for BPAD. Questionnaires to assess severity of either EUPD or ADHD can also be useful. The conceptualisation of personality disorders is particularly important for community mental health teams. The model that we use is the integrative framework proposed by John Livesley,^{Reference Livesley31} which differentiates between clinical assessment of traits and domains of psychopathology such as core self and interpersonal problems, as well as specific symptoms and failures of the regulation/modulation systems. A mood diary is a practical, simple tool that can be used to help with tracking changes in mood but also engages patients and involves them in their care. It also encourages their self-assessment and introspection, which can be further probed by semi-structured clinical interviews. Other ways of testing clinical hypotheses include longitudinal observation, although this is not always practical owing to limitations such as large clinical caseloads and pressures to reduce these. Trainees may find it indispensable to explore techniques for eliciting information in interviews, discuss relevant research findings and address risk factors in supervision and within the team to enrich exploration of relevant information that can contribute to formulation.

In this case, the patient had established diagnoses of ADHD, EUPD and BPAD. As we improved our knowledge of the patient and begin to develop our case formulation based on the methods explained above, the diagnosis of EUPD was questioned and later discarded, which highlights some of the issues discussed in this paper. Defining personality disorder can be a daunting task. In community psychiatry, we tend to distinguish personality disorder from the far more common personality dysfunction. This is based on an assessment of severity, as well as exploration of markers that can be used to differentiate an organised from a disorganised personality system (unstable sense of self, chronic interpersonal dysfunction). ADHD and BPAD can also affect relationships with the self and others. In these cases, there can be benefit from psychotherapy, which might help the person to understand their illness and relationships, although the exact type of psychotherapy offered may be influenced by whether they have been formally diagnosed with a disorder or dysfunction.

It was felt that we needed to identify a helpful mode of pharmacological therapy that would tackle the patient's BPAD symptoms in order to provide some stability and allow exploration and addressing of other factors (cognitive symptoms, self-efficacy, relationships, etc.). This would also help to test the hypothesis relevant to comorbid, untreated ADHD: that cognitive acceleration reduces if the mood is stabilised. In this case, the patient had already been trialled on mood stabilisers, but these were discontinued owing to side-effects. For a trainee, this can be a difficult situation to be faced with when first- and second-line treatment options have already been ruled out. It is always worth keeping an open mind when exploring past pharmacological intervention as to why treatments didn't work. This may allow re-visitation of failed treatments in a clinical context, e.g. if mania had been triggered by certain antidepressants, supporting a diagnosis of BPAD.

Finally, a long-term view is necessary to capture the evolving nature of BPAD, to help disentangle and address comorbidity (psychiatric and physical) and to improve outcomes. Furthermore, a longitudinal approach is essential to elucidate the underlying biology of BPAD from high-risk individuals and prodromal presentations to established cases.^{Reference Cross, Hermens, Scott, Ottavio, McGorry and Hickie32} In this case, the short-term team provided eight sessions, which helped to provide a diagnostic formulation and treat acute mixed affective symptoms. We then decided to refer to a long-term team, as we thought a longitudinal approach would be necessary owing to the complexity and the chronicity of some of the issues discussed here.

The patient herself reflects thus on the experience: