Client factors

Rather than identifying clients as complex, its more helpful to specify their complexity factors and describe the challenges that these could provoke within treatment. This needs to be balanced with reflection on facilitative factors and how these can be harnessed and supported. In spite of having long-term difficulties, David was very determined to overcome his problems and he was assertive in his approach to therapy. He was keen to form a trusting relationship with a therapist and appeared to have good alliance potential. Nevertheless, at the outset of therapy his problems appeared to be complicated and challenging to treat.

When considering whether a client has complicated problems, there is a risk of conflating this with related constructs such as severe, persistent, etc. The etymology of ‘complicated’ suggests a particular definition: multiple factors that are entwined or inter-woven. Problems that interact in a known, predictable way are unlikely to be complicated, whereas entwined problems are likely to be unpredictable, ambiguous or unstable. A number of David’s problems interacted with each other; for example, co-morbid depression and anxiety, but they did not interact in an entwined or unstable way – see Fig. 1. When he felt depressed he also felt more anxious and these states were mutually maintaining (Sharp and Harvey, Reference Sharp and Harvey2001). In contrast, David was perplexed by the difficulties he had experienced, over several decades, in his family relationships. Different members of the family held conflicting narratives about David’s father and the way he related to others. David’s beliefs about his father, and himself, were often confused, and several years of reflection had not produced a settled understanding of what had occurred in their relationship. Entwined family relationships had a deep and distressing effect on David over an extended period. David’s main complexity and facilitative factors are summarised in the case map depicted in Fig. 2.

Fig 2. Complexity and facilitative factors (complexity in red, facilitative in green).

Therapist factors

The therapist in David’s case was the first author. He had worked for 17 years in a specialist CBT service in the North East of England, with a particular interest in mood disorders. His continuing professional development included self-practice/self-reflection (Chaddock et al., Reference Chaddock, Thwaites, Bennett-Levy and Freeston2014) and training workshops on interpersonal processes. Prior to treatment, he appeared to be a good match to David’s needs in terms of CBT knowledge, skills, reflective capacity and interpersonal range. His weaknesses were a tendency to be overly theoretical and take too much responsibility for change. He and his supervisor were aware of these tendencies and they did not manifest with all clients. Nevertheless, they had the potential to influence David’s treatment in an unhelpful way – see Fig. 2.

Healthcare factors

Healthcare factors need to be considered from both client and therapist perspectives. From the therapist’s point of view, the healthcare system was in a state of flux during the COVID-19 pandemic, but arrangements were in place for David’s treatment to be provided remotely. With respect to meeting David’s needs, there were two key facilitative factors. Firstly, the therapist received specialist supervision from the second author who had 26 years of experience supervising complex cases. The therapist and supervisor had a strong supervisory alliance and shared an interest in mood disorders. Secondly, the tertiary service in which they were based was able to offer long-term treatment when clients needed it. Given that David’s previous CBT had been brief and its effects not lasting, the option of longer-term treatment was facilitative and supportive.

From the client’s point of view, David had strong alliances with some NHS staff which could have a facilitative effect in this treatment (Cooper et al., Reference Cooper, Gillmore and Hogg2016). However, he also had a number of adverse experiences in mental health services. He had requested trauma-focused therapy over a number of years and was repeatedly told it would be de-stabilising and he would not be strong enough to tolerate it. Five years previously, a locum member of staff in a Community Treatment Team (CTT) discharged him from the Trust, possibly to meet service targets, at a time when David’s depression was severe and his risk of suicide was moderately high. The locum staff member had no knowledge of the therapy service that David was subsequently referred to. Being discharged during a vulnerable period had a very negative impact on David which, unfortunately, re-activated trauma memories and worsened his mental health for several months. It took two years of repeated requests for him to be re-referred into services. Memories of these experiences were salient for David and had the potential to trigger a lot of distress – see Fig. 2.

Contextual factors

Contextual factors also need to be considered from both client and therapist perspectives. From the therapist’s point of view, he was settled and supported in his professional and personal life, with no significant issues adversely affecting his work. From the client’s point of view, David had limited sources of social support but the sources he had were highly engaged, consistent and committed to his treatment, particularly his wife and members of his church. However, David’s broader family situation was very fluid and stressful, with three of his four adult children having significant mental health problems (Gantt-Howrey et al., Reference Gantt-Howrey, Brookover and Robins2025). David and his wife were providing a high level of support to them and responding to unpredictable events and crises on a regular basis. This had the potential to disrupt the flow and focus of treatment – see Fig. 2.

Clinical cases have a combination of complexity and facilitative factors and how they interact dynamically over time influences the therapeutic process. One way of formulating this is with respect to the therapeutic relationship. If a sufficiently strong working alliance can form, there is potential for the client to benefit from therapy even if they have complicated problems. Conversely, if client, therapist, healthcare and contextual factors create barriers within the working alliance, the therapeutic process will be prone to friction, obstruction or rupture, and this will significantly limit therapeutic gains (Doukani et al., Reference Doukani, Free, Araya, Michelson, Cerga-Pashoja and Kakuma2022). This situation describes and defines clinical complexity: in the current model, treatment is complicated to the degree that there are barriers within the working alliance. Client, therapist, healthcare and contextual factors contribute to different degrees in different cases, but it is the presence of alliance barriers that constitutes clinical complexity.

Barrier 1 – The client’s past experiences of services

Because David had been declined trauma-focused therapy in the past, and been discharged from services inappropriately, he was anxious to find out how the therapist would respond to his request to explore the causes of his depression, not just manage symptoms. David did not meet all the criteria for PTSD and he had not been placed on a trauma pathway. His main presenting problem was recurrent depression and the request to explore causes raised the therapist’s concern that this could lead to depressive rumination. However, in the initial sessions, it would have been a significant barrier to engagement if the therapist had taken attention away from the causes of David’s depression. Rather than commence with a standard course of cognitive therapy (Beck et al., Reference Beck, Rush, Shaw, Emery, DeRubeis and Hollon2024) or behavioural activation (Martell et al., Reference Martell, Dimidjian and Herman-Dunn2021), which David had received in the past, the therapist encouraged him to share information about his relationship with his father. He also encouraged him to talk about his experiences in mental health services, both positive and negative. This initial response to David’s request helped to build a lot of personal alliance which formed rapidly over the first three sessions. David felt supported, taken seriously and hopeful. A potential barrier to engagement was overcome, because empathy and compassion were prioritised in response to David’s concerns. However, personal alliance needed to be balanced with task alliance and there were three inter-related barriers to achieving this.

Barrier 2 – The client’s multiple interacting problems

As indicated in Fig. 1, David was not only living with distressing memories of his father. On a regular basis, he was managing depressed mood, elevated anxiety, a chronic pain condition and a turbulent family situation. The therapist was concerned that treatment could become diffuse or ‘pinball’ if too many of David’s problems were tackled simultaneously. This was discussed explicitly and David agreed that therapy needed a consistent focus. He had a clear preference to limit attention to his relationship with his father, and this became the agreed target problem. Limiting attention to an agreed target helped David prepare for trauma-focused work. His family problems affected him intermittently in the following months, but they did not interfere with the therapy or obstruct it. Having an agreed target strengthened the task alliance; however, after approximately six sessions there was no agreed plan how to treat the relational trauma.

Barrier 3 – The therapist’s inflated responsibility

The therapist was conscious that previous courses of therapy had supported David and given him tools to manage distress, but they had not helped him to heal from trauma or reduce his risk of depression. The therapist realised that a strong personal alliance was forming, but he felt uncertain about the optimal treatment plan and started to feel overly responsible for initiating a change process. This was partly because David’s case did not fit neatly into a depression or PTSD protocol. The therapist was conscious of avoiding further failed treatment, and he started to put pressure on himself to find a rationale for change. This strained the collaboration at key points; for example, the therapist assumed that David felt anger towards his father, rather than asking him. He proposed a piece of compassion-focused work to develop David’s self-soothing skills, but this was confusing and frustrating for David because he had already let the therapist know that a previous course of compassion focused therapy (Leaviss and Uttley, Reference Leaviss and Uttley2015) – while helpful at the time – had not addressed the relational trauma. Fortunately, David was open about this at the next session and it was discussed explicitly. In subsequent supervision, the therapist realised that he had fallen into the trap of taking too much responsibility and he had acted unilaterally in guiding the therapy. The personal alliance was strong enough to contain this and David tolerated the associated frustration. Rather than it rupturing the therapeutic relationship, it had a clarifying effect on next steps.

Barrier 4 – Uncertainty about a bespoke treatment plan

As described in Barton et al. (Reference Barton, Armstrong, Meares, Bromley and Whitton2024), a specific therapy plan was agreed after 11 sessions. Although David did not meet all the criteria for PTSD, the formulation of intrusive memories and disturbances in self-organisation had the potential to benefit from memory processing from trauma-focused CBT (Ehlers, Reference Ehlers2013; Ehlers and Clark, Reference Ehlers and Clark2000). Although David’s main presenting problem was recurrent depression, a treatment component from a PTSD protocol was a better fit to the emerging formulation, and explicit conversations in supervision were essential to clarify this and target the therapy appropriately. The 11 sessions spent building the working alliance gave David the opportunity he needed to prepare for trauma-focused therapy, and when he commenced memory processing he was ready and well-supported to do so. Across the remainder of treatment, the working alliance continued to strengthen across a total of 47 sessions. At each step, the interpersonal bond was strong enough to support and contain the tasks of therapy. David’s therapeutic gains helped to deepen mutual trust and shared engagement in the change process, ultimately leading to a very positive clinical outcome.

In each of the alliance barriers in David’s case, they were overcome through an iterative process of: (a) noticing the barriers, with attention brought to them; (b) reflecting on the barriers, seeking to formulate and understand them; and (c) discussing the barriers explicitly, in supervision and therapy. Problems in the therapeutic relationship can sometimes be addressed implicitly, for example, through non-verbal communication, but more significant barriers need explicit conversations in supervision and therapy (Safran et al., Reference Safran, Muran and Eubanks-Carter2011).