The pitfall of fragmentation

Because managing healthcare systems is beyond doubt a complex task, a tendency toward reductionism may be seen as a way of understanding and managing very complex realities in a manner that feels controllable.^{Reference Plsek and Greenhalgh19,Reference Dooris, Farrier and Froggett20} However, over time, reductionist approaches such as functional models often appear to have failed to consider the unintended consequence of creating a fragmented system.²¹ As contemporary mental healthcare systems keep placing significant emphasis on short-term care, brief interventions and specific outcomes, an increasing number of people may be at risk of falling through the system's cracks by not fitting the complex criteria for specialist services and interventions. This may paradoxically result in higher rates of hospital admission and community interventions, and, ultimately, greater economic burden for healthcare systems and governments.

A number of psychiatric services have also been traditionally developed around specific conditions or disorders, rather than the very people accessing them. As a result, individuals are often required to describe their circumstances, including sensitive issues, multiple times across different services – sometimes only to be signposted elsewhere or sent back to original referrer. Unsurprisingly, these fragmented processes generate a significant delay before effective help is delivered, and, especially during a crisis, may exacerbate low mood and anxiety and perpetuate pre-existing issues because they fail to consider patients’ attachment styles and needs.

Initially developed by John Bowlby^{Reference Bowlby22} and later expanded upon by Mary Ainsworth,^{Reference Ainsworth23} attachment theory explores the profound emotional bonds formed between individuals and their caregivers. Its fundamental assumption is that children's early interactions with primary caregivers shape their beliefs about themselves, others and relationships. This process results in the development of different attachment styles – categorised as secure, anxious–ambivalent, avoidant and disorganised – which can have profound effects on lifelong patterns of behaviour and interpersonal relationships.^{Reference Bowlby22,Reference Ainsworth23}

An example of the effect of fragmentation on attachment styles was offered by Adshead,^{Reference Adshead24} who highlighted that the way current services are designed may promote insecure and ambivalent attachment in people, particularly because of repeated transfers and referrals across specialist teams – often leaving little to no agency to patients themselves.^{Reference Adshead24} This is especially relevant when working with serious mental illness (SMI), since a report published by the Schizophrenia Commission highlighted the fragmentation of services as a potential perpetuating factor for psychosis, as well as the reason why patients can feel ‘as if on a factory production line’ rather than properly looked after.²⁵ These observations were often placed within the context of a broken and demoralised system with lack of person-centred approaches, family members’ involvement and, once again, continuity of care.

On the other hand, the same report praised the approach of other teams working with SMI, such early intervention in psychosis (EIP), which promoted active engagement of patients and their social networks, and encouraged independence and informed choices – all factors which may be seen as attachment-informed. The recent National Institute for Health and Care Excellence guidelines for EIP in the UK²⁶ stressed the importance of such factors, recommending that care coordinators establish a therapeutic relationship with patients and their family, ideally from the first point of contact to the final discharge.

In another example related to SMI, continuity of care (and thus secure therapeutic attachment) was also a major focus of the development of the Trieste model of mental healthcare, which has been hailed by the World Health Organization as one of the most progressive in the world.^{Reference Portacolone, Segal and Mezzina27,28} Spearheaded by Italian psychiatrist and neurologist Franco Basaglia (1924–1980) – the principal proponent of the eponymous law that dismantled asylums and reformed the psychiatric system in Italy – this model saw lack of continuity of care in people with SMI as contributing to chronicity and relapses, with a great emphasis on offering flexibility to maintain a relationship with the patient and to avoid de-socialisation and abandonment.^{Reference Portacolone, Segal and Mezzina27,Reference Mezzina and Vidoni29,Reference Mezzina30}

The negative impact of fragmentation in terms of clinical outcomes is also no mystery. When considering SMI, the lack of attachment-informed factors such as continuous care and secure therapeutic relationships have been linked with poorer outcomes in people with schizophrenia.^{Reference MacDonald, Adamis and Craig31} Perhaps even more worryingly, however, such a link is likely to be found in all mental health conditions where the continuity of care is similarly poor. Over-reliance on specialist care has been highlighted as a potential contributing factor, with examples where the higher the number of community teams involved with a patient, the worse the clinical outcomes.^{Reference MacDonald, Adamis and Craig31} Other authors have also emphasised how patients may be significantly affected by the behaviour of clinicians, describing an ‘efficacy paradox’, whereby the therapeutic relationship had greater influence on clinical outcomes than the pharmacological treatment that was being tested^{Reference McQueen, Cohen and John-Smith32} or even the specific psychotherapeutic model adopted.^{Reference Norcross and Lambert33}

Shifting paradigms

The current limitations of how mental healthcare is organised have led some clinicians to look at different paradigms to deliver safer and more effective mental health interventions.³⁴ However, although there have been numerous attempts to change the way of delivering services or challenging the current way of thinking, few lasting results have been observed on healthcare institutions. For example, Barker^{Reference Barker35} proposed the Tidal model of care, with emphasis on metaphors and personal narratives to promote recovery. This has shown to be especially helpful in promoting sense of self and connectedness in people with SMI where continuous care is paramount, such as forensic patients^{Reference Cook, Phillips and Sadler36} and psychiatric populations with addictions.^{Reference Kusdemir, Oudshoorn and Ndayisenga37}

Sweeney et al^{Reference Sweeney, Filson and Kennedy38} instead highlighted the need for trauma-informed mental health services, which are derived from the notion that trauma is widespread among people living with mental health difficulties as well as in society in general. These promote a switch from a narrative based on asking ‘What is wrong with you?’ to a less blaming and pathologizing, ‘What happened to you?’,^{Reference Winfrey and Perry39} which is intended to result in a solution-focused approach. The re-traumatising processes of some of the current healthcare systems, in particular within in-patient settings, are what these approaches aim to avoid, with a focus on recovery and growth. A further attempt at changing perspectives may also be identified in the model of evolutionary psychiatry, which incorporates factors related to the history of human evolution (phylogenesis) and the development of individual human beings (ontogenesis) in the aetiology of mental health difficulties.

Although the models above have certainly contributed to a better understanding of healthcare challenges and offered alternative ways of conceiving service delivery, they have so far appeared to be unable to set clear principles and build a robust theoretical framework for the development of services that allow for a secure base, prompt responsiveness and relational continuity. For example, because of their strong focus on theoretical evolutionary meanings, the principles of evolutionary psychiatry have often been the object of speculation with regards to their effective translational value for clinicians.^{Reference Nesse40}

Furthermore, although often portrayed as ‘paradigm shifts’, such models do not represent a real shift according to the cycle of scientific revolution defined by Kuhn,^{Reference Kuhn41} which outlines distinct phases whereby science needs to go through a model drift, a model crisis and eventually, a model revolution. As many mental healthcare systems worldwide are certainly going through a model crisis phase, very few may ever reach the model revolution phase without a clear theoretical framework. The development of most mental health services over the past few decades has also appeared to rely on population-specific evidence bases and the allocation and management of finances. Although often essential, these driving processes hardly facilitate the adoption of robust and coherent theoretical frameworks that can help to understand human behaviour and emotional responses in everyday life, as well as in moments of crisis and suffering.

Attachment and mental healthcare systems

As we have seen, reductionist approaches appear to have failed at addressing some of the fundamental issues affecting mental health services, particularly as complexity needs to be accepted as an intrinsic part of mental healthcare systems. The lack of clear guiding principles in mental health service development might also be partially responsible for the fragmentation of services and the lack of continuity of care, which, in turn, prevent the development of meaningful and effective relationships with clinicians as well as services as a whole.^{Reference Adshead24,25,Reference MacDonald, Adamis and Craig31} And, as mentioned above, this is particularly important when dealing with SMIs, which typically require longer-term care.^{Reference Wiersma42}

A potential answer to this gap in guiding principles might be found in the abovementioned theory of attachment.^{Reference Bowlby22,Reference Ainsworth23} With regards to service delivery, attachment theory suggests that caregivers should aim to provide a secure base to increase availability, flexibility, sensitivity and responsiveness to distress, while also intervening when needed to promote secure attachment.^{Reference Bowlby22} This appears even more relevant when, as highlighted by Adshead,^{Reference Adshead24} attachment styles are conceived ‘as a biobehavioural stress management system that is activated whenever a person is distressed and vulnerable and seeks care or protection’. Attachment styles have largely been found to play a major role not only in the relationship with healthcare professionals, but also in treatment adherence and clinical outcomes across a wide range of mental and physical health conditions,^{Reference Palmer Kelly, Tsilimigras and Hyer43,Reference Klest, Tamaian and Boughner44} and especially in individuals with SMI.^{Reference Pearse, Bucci and Raphael45} Secure attachment facilitates the development of healthy psychological habits, effective emotion regulation and appropriate interpersonal support in times of distress, as well as the acquisition of adequate emotional lexicon and skills that enable individuals to identify and describe their feelings.^{Reference Adshead24} It also promotes the development of autonomy, resilience and healthy interpersonal relationships, which, in turn, influence treatment adherence and long-term health outcomes.^{Reference Adshead24}

Nonetheless, as mentioned, the current organisation of mental health services does not necessarily favour secure attachment styles and continuity of care. The current emphasis on short-term treatments with insufficient time to establish a relationship that is safe and secure in attachment terms may promote insecure attachment, with obvious negative consequences for individuals.^{Reference Adshead24} Access to support is also often described as difficult, and rigid inclusion criteria and a large number of teams or specialisms provide fragmented experiences of care.^{Reference MacDonald, Adamis and Craig31} Similarly, Beale^{Reference Beale46} described how some mental healthcare systems appear to be designed either to coerce or to exclude, and how excluding patients from services has become common practice among clinicians, who think that it is clinically justified to do so. This, along with lack of resources and a risk-averse culture, may lead to systematic exclusion and iatrogenic harm in patients, as well as moral injury and erosion of compassion in mental health professionals.^{Reference Beale46}

The suggestion that attachment theory may inform the design of mental health services is not new. A little less than a decade ago, Bucci et al^{Reference Bucci, Roberts and Danquah47} reviewed the literature regarding attachment theory within mental health policy and service development, and concluded that it may indeed represent a viable theoretical framework. They identified eight key themes to guide the development of attachment-informed services, including service policies and evaluation, referrals, assessments and formulations, interventions, and support for staff and informal carers. The general lack of evidence on the adoption of attachment theory, as well as the lack of evaluation of services based on the ability to meet attachment needs, were also highlighted.^{Reference Bucci, Roberts and Danquah47}

A need to move from specialist to more generalist care to develop attachment-informed services has also been identified previously. Crossley and Lepping^{Reference Crossley and Lepping48} described the role confusion that current specialist-focused systems may create, and emphasised how more generalist systems may act with safety and best interest in mind rather than focusing on targeted changes, and may prevent patients from becoming lost in the system by attempting to reproduce secure attachment. A similar concept has been supported by Laugharne et al,^{Reference Laugharne, Thompson and Srivastava49} who defended generalist services as a place where continuity of care and therapeutic relationships are easier to establish.

With regards to attachment-informed systems, a virtuous example is perhaps represented by Sweden's ‘Esther’ model. Developed in the late 1990s from the negative experience of a fictional elderly patient (‘Esther’),^{Reference Gray, Winblad and Sarnak50} its overarching principle was a shift from narrow, organisation-centred views – focused on the best course of action for a specific unit or department – to a more patient-centred approach, summarised by the question ‘What is best for Esther?’.^{Reference Vackerberg, Andersson and Peterson51} The Esther model has so far been successfully implemented for individuals with complex health needs in some places across different countries, including the UK^{Reference Lee, Vibhuti, Therkildsen, Manley, Wilson and Øye52} and Singapore.^{Reference Lim, Khee and Thor53} There, the model created a system in which healthcare organisations provide care in a collaborative, organised and supportive fashion⁵⁴ by focusing on ‘what the individual wants, not what professionals think they should have’.^{Reference Lee, Vibhuti, Therkildsen, Manley, Wilson and Øye52} This level of cohesion is believed to simulate the stability of a nurturing and healthy parental relationship, and thus, a stable and safe base for individuals navigating the system and receiving interventions. Ultimately, the adoption of the principles required to promote a secure attachment in the Esther model may represent a catalyst for recovery when caring for people with SMI and other conditions requiring longer-term care (e.g. elderly patients with complex needs^{Reference Gray, Winblad and Sarnak50}), in whom attachment is known to play a crucial role with regards to functioning.^{Reference Pearse, Bucci and Raphael45}

The same determining factors of secure attachment – such as security, respect, responsiveness, negotiation, promotion of human rights and autonomy – also underpin the good outcomes of the abovementioned model of mental healthcare in Trieste.^{Reference Mezzina and Vidoni29} However, the principles of the Trieste model are not bound by a theoretical framework, which may be one of the reasons why it has been historically difficult to replicate on a large scale, both within Italy as well as elsewhere in the world.^{Reference Portacolone, Segal and Mezzina27,Reference Mezzina30}

A few ideas for the future

Despite receiving previous attention,^{Reference Bucci, Roberts and Danquah47} and besides few sporadic examples such as the Esther and Trieste models,^{Reference Mezzina30,54} attachment theory still struggles to find its way within the development of mental healthcare services. It is perhaps easier to think of attachment at the individual level, with a focus on how the attachment style of patients affects presentations and how the attachment style of healthcare providers affects responses. However, it could be argued that mental healthcare systems as a whole can be conceived of ‘organisational caregivers’ within attachment theory, thereby focusing on whether their organisational climate promotes secure levels of attachment for both patients and staff.

The quality and functions of an effective individual caregiver – such as availability, proactiveness and responsiveness – need to be translated into service delivery as accessible, proactive and continuous care at an organisational level. Mental health services may also need to challenge the current reductionist trends and move toward more generalist, inclusive, responsive and ‘defragmented’ models of care characterised by less stringent inclusion criteria. For example, barriers identified in current referral systems would need to be challenged, minimised and possibly eliminated in a timely fashion, including issues such as unhelpful financial management processes.

The meaning of multiagency and multidisciplinary work should also be invested with renewed significance, since the flexibility and responsiveness of both individual interventions and services as a whole should be at the centre of any type of development. The continuous collaboration between psychiatrists, clinical psychologists and other mental health professionals should be encouraged further – not only to ensure the delivery of effective integrated models of care, but also to promote the adoption of biopsychosocial and systemic approaches in mental health areas where strong pathologising narratives still persist.^{Reference Zarotti, Dale and Eccles8–Reference Dale, Wood and Zarotti10}

Finally, but importantly, a potential misconception should be clarified, i.e. that attachment-based services will create dependence. A clear distinction should be made between offering attachment-informed services and promoting dependence, as the main goal of mental healthcare should remain the promotion of independence and self-care.^{Reference Bucci, Roberts and Danquah47} Where mental health services truly promote a secure attachment by creating a safe place, independence should follow naturally.

About the authors

Alberto Salmoiraghi is a consultant psychiatrist and Medical Director for the Mental Health and Learning Disabilities Division in North Wales, Betsi Cadwaladr University Health Board, Bangor, UK; an honorary senior lecturer at Bangor University, Bangor, UK; and visiting professor at Wrexham University, Wrexham, UK. Nicolò Zarotti is an academic and clinical psychologist with the Department of Clinical Neuropsychology at the Manchester Centre for Clinical Neurosciences (MCCN), Salford Royal NHS Foundation Trust, Salford, UK; and the Division of Health Research at Lancaster University, Lancaster, UK.