Clinical background

Ben was a man in his 40s. He had treatment-resistant schizophrenia, which for two decades had been unresponsive to multiple depot and oral antipsychotic medications, including dosages well over (up to 1000%) the British National Formulary (BNF) licensed maxima, and in combination with mood stabilisers. Ben refused clozapine and the necessary blood tests, so this had never been prescribed.

Detained under section 3 of the Mental Health Act 1983 and lacking capacity because of his floridly psychotic mental state, Ben was subject to involuntary treatment. Following psychotically driven assaults he moved up levels of secure care and after several years in seclusion was transferred to a high secure hospital. By the time of his arrival in high security Ben was very distressed and could not attend to his own most basic needs. He was dishevelled, went for long periods neither eating nor drinking and had lost a great deal of weight. He had reached a crisis point.

Clinical intervention

The team considered the various potential interventions and their associated risks and benefits (Table 1) and the treating consultant proposed the enforced nasogastric administration of clozapine, used in the USA (Reference FisherFisher 2003) but novel in the UK. However, the combination of the patient's extreme and at times childlike vulnerability, the anticipated further distress, uncertainty of outcome and ongoing risk of violence all evoked powerful emotions in the professionals involved and many had concerns about how best or whether to proceed.

TABLE 1 Risks and benefits of potential treatment strategies considered by the team

Consultation regarding the novel intervention and team concerns

In light of the difficulties in initiating a novel and unusual intervention, extensive discussions took place during the weekly multidisciplinary team meetings as well as with professionals and others inside and outside the hospital. These involved a medical peer review including the lead psychiatrist, key nursing, auxiliary and social care staff, the hospital's legal department and the team in charge of physical restraint. The treating consultant and social worker also consulted Ben's family. Both Ben's lawyer and an independent mental health advocate (IMHA) attended care team meetings and discussed the alternatives. For advice about the technical aspects of the procedure an intensive care physician from the local general hospital and a psychiatrist in North America who had described its use in a published paper (Reference FisherFisher 2003) were contacted.

As required by the Mental Health Act 1983, the team sought authorisation of the proposed treatment from a second opinion approved doctor (SOAD). The SOAD, who was an independent psychiatrist appointed by the Care Quality Commission (CQC), met with nursing staff and the team psychologist before agreeing to authorise the plan. The CQC was consulted as to whether the nasogastric route should be specified.Footnote a

Although anxious about the proposal, everyone was concerned about Ben's desperate condition. Many alternative treatments, such as atypical antipsychotics, high-dose regimes and covert medication, were suggested; all either lacked likely efficacy as they had already been unsuccessful over the previous two decades, or themselves presented serious problems in relation to harms and ethical difficulties.

Concerned team members were reassured by the wide-ranging consultation, inclusion of the patient's family and legal team as well as support from both within the hospital and externally. Nurses discussed the issues among themselves formally during handovers and reflective practice and informally. The team were also united in wanting to help a desperately ill individual and they had benefited from multiple shared experiences over many years of successfully initiating clozapine with dramatic and sustained improvements.

Treatment initiation and clinical progress

Ben was offered oral clozapine every day for a month. In the final week, the treating consultant visited daily, offering the oral preparation and explaining in simple terms the need for treatment. He eventually explained that if Ben continued to refuse medication, with no clinical improvement, he would be held and a ‘tube’ would be used to administer it.

Ben continued to refuse and so involuntary treatment was initiated for a 5-day trial. He was restrained by nursing staff while a junior doctor inserted a nasogastric tube; a stomach aspirate was taken to confirm position and then the dose of clozapine was given. Interventions lasted up to 15 minutes. The treating consultant was present for all but one restraint. This provided support for the team as a whole and the junior medical staff in particular. It also allowed the team to debrief and determine whether or not to persist. This decision tried to take into account the severity of Ben's physical and mental states, the distress and difficulty of the procedure for him and team members, the lack of less restrictive options likely to work and the risks of deterioration through dose re-titration if no clozapine was taken for 48 hours. This judgement could only be made effectively by first-hand knowledge of the process and it was apparent that some staff needed the psychological support of knowing that the consultant was present.

In light of rapid improvement in Ben's mental state, with reduced distress and improved food and fluid intake but ongoing resistance to oral medication, the initial 5-day plan was extended at the suggestion of the nursing staff, and medical and physical healthcare staff made themselves available at the weekends to ensure that there was no break in clozapine titration.

Ben received nasogastric clozapine under restraint four times over 19 days. The other days he took the oral clozapine reluctantly. By day 20, Ben ate and drank regularly, took clozapine without demur and was markedly less fearful and hostile. Seclusion ended the next day. His level of functioning markedly improved and other patients who recognised his vulnerability befriended him. Seven months after the initiation of clozapine, and following some negotiation, he transferred to a low secure hospital. Shortly before he moved from high security he had escorted community leave for the first time in years, to a nearby beach and for fish and chips. When he left, he thanked the care team looking after him, giving both them and his treating consultant a thank you card. He remained in telephone communication with the ward nursing staff for over a year, because he liked them.

Ben continues to be markedly improved, taking oral clozapine, and he has been transferred out of forensic services to a rehabilitation hospital. He is able to take unescorted leave and travels alone across his home city to visit a relative. His current team are looking to discharge him to a supported community setting. When asked about his experiences he was neither distressed nor appeared to have meaningful recall of what had happened.

Restraint and restriction to achieve a beneficent outcome

The primary ethical dilemma was whether the intended benefit of the proposed intervention justified or obliged the degree of physical intrusion, restriction and restraint.

Professional and team anxieties

Despite legal authorisation and ethical justification, anxieties among the care team remained that led to conscious (and possibly unconscious) obstructions.

How are ethical decisions made?

One might suppose that ethical decision-making is done by carefully collating all the available evidence and weighing up the likelihoods and consequences of the anticipated outcomes in an aware and reflective way, supported by a societal consensus regarding rights and values, and then reaching a decision, without emotional influences, that can be applied within a legal framework.

How likely is that and what comes first anyway, the decision or the reasoning? A major theme of moral psychology is the investigation of what drives moral decision-making. Psychiatrists in particular should not be surprised by Haidt's social intuition theory: this posits that, rather than the series of careful cognitive steps suggested above, moral decisions are intuitive, with moral reasoning taking the form of post hoc rationalisations made to influence others, which in turn may be influenced or changed through discussion (Reference HaidtHaidt 2001). Reference Greene and HaidtGreene & Haidt (2002) then developed the dual process theory of moral reasoning, acknowledging that both emotional and reasoned judgements can take place.

Functional magnetic resonance imaging (fMRI) supports the hypothesised relationships between intuitive judgements and emotion, and between consequentialist approaches and reasoned thought, by demonstrating that principle-based decisions are made quickly and with activation of the amygdala, whereas consequentialist decision-making takes longer and there is frontal lobe activation (Reference Greene, Sommerville and NystromGreene 2001, Reference Greene and Gazzaniga2009).

The emotion of disgust is particularly cited as one that may result in moral judgements and reasoning in ways that are irrational in terms of consequences. The ‘yuk factor’ (Reference Haidt, Rozin and McCauleyHaidt 1997) – in the case of Ben, the restraint of a weak and vulnerable man to place a nasogastric tube for treatment he resists – is likely to evoke various painful emotions, such as anger, fear and resentment, but from different perspectives and resulting in opposed conclusions. For example, from team members: ‘He is weak and vulnerable. Restraint and a nasogastric tube is cruel/disgusting. We must do everything we can to protect him’. From the treating consultant: ‘He will die. I am afraid I/we will be blamed. We must do everything we can to save him’. Alternatively, but perhaps less likely, more thought-out conclusions from principle-based and consequentialist positions might have been in opposition.

When decisions of any sort are being made, decision-making that seems to be based on purely cognitive appraisal may in fact be affected by the short-term experience of positive or negative emotions (Reference Kahneman and TverskyKahneman 1979). If decision makers are facing difficult, unpleasant, risky and uncertain choices, unconscious and automatic emotional reactions have a significant influence on their attitudes, appraisals and actions (Reference DamasioDamasio 2005; Reference Slovic and PetersSlovic 2006). This has a much wider application in psychiatric professional judgements, particularly about risk.

Resolving ethical conflicts

Ethical disagreement in healthcare teams is clearly not new, strange or undesirable. For some emotionally demanding types of work it may be inevitable (Reference Kovacs, Kovacs and HegedusKovacs 2010). However, the key issue to consider in terms of team functioning is the degree to which team members may lack awareness of these differences and their emotional responses to them. They may be unaware of the different value perspectives that they are taking and of the similarities and differences between these different perspectives. This can cause anxiety in people trained to make difficulties disappear, and this may be acted out in ways that affect the quality of team functioning.

In the case of Ben, the clinical team were made to feel like ‘bad’ people for doing something they thought was right, even after a lengthy process of ethical and legal debate supporting their plan. In his description of values-based practice in mental healthcare, Reference FulfordFulford (2008) argues that different ethical positions may reflect important differences in value perception and weighting, and that it is important for teams to explore this ethical ‘dis-sensus’, because there may be more than one ethically sustainable argument. It can be hard for healthcare professionals to accept that in some situations there is no single ‘good’ intervention that will save the patient and allow them to retain their sense of themselves as ‘good’ people.

Reference HaidtHaidt (2001) proposes that the main purposes of moral positions are to influence others, maintain social relationships and preserve self-definitional attitudes. Moral positions can be changed through private reflection and also by the reasoned persuasion, emotions and intuitions of others. It is in the sharing of one another's judgements as well as reasons that the emotions, intuitions and ideas of others can move. In this case, the treating consultant shared his fears about the consequences of inaction – death through neglect – as well as the more formal approaches of weighing up and analysing. This may well have helped shift the intuitions, and hence conclusions, of opposing team members, which may have been dominated by fears, for Ben's welfare or of criticism from others. Sharing thinking might also have helped team members take a consequentialist approach, looking at the outcomes of each of the unpleasant alternatives.

A particular difficulty of this case was that, although many other treatment avenues were suggested, each had been tried repeatedly and failed (apart from electroconvulsive therapy (ECT), which would have entailed greater physical health risks). But as well as the discussions at care team meetings, many others were involved, were seen to be involved, visited the ward and talked with staff about their judgements and reasoning. This allowed many opportunities for social persuasion, the influence of one person's judgement on the other's intuition, as well as reasoned persuasion, the influence of one person's reasoning on the other's intuition. This was quite clear when nurses who spoke of their fears to the SOAD were told: ‘You should not be worried about doing this; you should be worried about not doing this’. This was a clear statement of a consequentialist position and it had a very powerful effect.

What seems key here is the degree to which professionals can become aware of their emotional biases and the extent to which they can become aware of the anxiety, fear, competitive arousal and anger that may influence decision-making in difficult circumstances. Reference Woodbridge and FulfordWoodbridge & Fulford (2004) suggest that teams in ethical conflict have group discussions to explore their different perspectives, similar to the reflective practice processes that are now regularly recommended for teams who work in emotionally demanding clinical settings such as mental healthcare (Reference GrahamGraham 2000; Reference Yakeley, Hale and JohnstonYakeley 2014). It seems especially important in mental health settings, where professionals have to work closely with people whose mental disorders make them irrational, that those professionals pay close attention to their own capacity for ‘irrationality’.

Irresolvable ethical disagreement

Although not affecting the action in this case, some ethical disagreements are irresolvable. These are associated with moral conviction, a strong absolute belief that something is right or wrong, moral or amoral (Reference Skitka, Bauman and SargisSkitka 2005). Typical examples are abortion and capital punishment. These beliefs tend to be perceived by the holder as universal matters of fact and are difficult, if not impossible, to change through discussion. Interestingly, at least one vocal critic maintains the view that, although retrospectively the decision was correct in Ben's case, a similar course of action should never be taken again.

What was learnt?

In this case, the team's assessment was accurate, and therapeutic success was achieved, more quickly and to a higher degree than expected. The anticipated good outcomes (based on clinical experience and evidence) did happen; the only ‘bad’ consequences were that Ben suffered some short-term discomfort and loss of dignity, and some staff were unhappy and anxious, also in the short term. However, nursing staff in particular radically shifted position from initial fear and skepticism to actively suggesting a similar course of action with subsequent patients. The team worked together cohesively once Ben's condition began to improve. The ability to take a flexible approach was probably very important, although this caused some difficulties. A series of questions were raised in the discussions before the restraints began: ‘How long will we carry on for?’, ‘When will we stop?’, ‘What if Ben makes no improvement?’, ‘How will we know if it is worth it?’ Good questions, with no easy answers, other than: ‘We will make a decision together, taking everything into account’. Had Ben shown little or no improvement, treatment would probably not have been pursued beyond a fortnight, but then the alternatives would have been very limited.

After an interval of about 2 years, further patients at Ashworth Hospital have received similar interventions, including from consultants who had initially been very fearful. At least one psychologist has remarked: ‘Why didn't we do this earlier?’ About half the patients for whom the intervention was planned accepted oral medication without recourse to restraint or nasogastric administration at all (some had been informed of the plan for nasogastric administration after oral clozapine had been refused, others were persuaded before the planned use of nasogastric clozapine was revealed). Subsequent team decision-making has not been associated with conflict at anything like the degree first experienced. Instead, there has been great uncertainty from external teams asked to assess patients who have markedly improved following some use of nasogastric clozapine and for whom transfer out of high secure hospital has been requested.

Group dynamics

Within teams, a range of negative behaviours can be observed that profoundly influence group dynamics: competition, pairing and scapegoating (Reference Yank, Lindsay and BarberYank 1992). Unrecognised, these behaviours can create hierarchical autocratic atmospheres, splitting and rigidity, and isolation of solitary opinions. Conflicts within groups and teams have been reported to the UK's National Clinical Assessment Service as a common reason for calling into question the professional capacity of doctors (Reference Donaldson, Panesar and McAvoyDonaldson 2014).

At present within the National Health Service (NHS), group consensus is increasingly valued in healthcare teams. However, increasingly, clinical teams can struggle to manage the anxiety that goes with the management of complex cases, as staff are subject to steadily higher levels of scrutiny, especially following critical incidents or inquiries, in a working climate where blame and ‘finger-pointing’ at individuals are favoured (Mid Staffordshire NHS Foundation Trust Inquiry 2010). In clinical contexts where the risk is high and there is strong emotion, anxiety about who will be held accountable if things go wrong can lead to hostile behaviours and interactions within and between teams. Staff may be reluctant to take risks or be innovative, even if this means that patient welfare suffers (as seems to have been the case in Mid Staffordshire). These pressures may be even more painful in mental healthcare, where the evidence base is less well established and there are many empirical uncertainties about how best to treat really complex conditions.

Unconscious heuristics in decision-making

Unconscious emotions can have a significant influence on high-risk and high-cost decisions. In the case of Ben, the patient was so disturbed and distressed that it was inevitable his care would evoke very strong feelings in all those involved, no matter how peripherally. The treatment proposed was novel, came with no guarantee of success, was inherently intrusive and would result in at least temporary distress. It is likely that the anxiety about doing harm made it difficult for staff to ‘see’ the complexity of the different therapeutic options.

Particularly in mental healthcare, without an acute crisis, there is potentially a heuristic bias against doing something rather than nothing, especially in relation to high-risk cases, i.e. professionals may not see inaction as a form of action in itself that may cause harm (Reference Spranca, Minsk and BaronSpranca 1991). In the UK, nursing staff, and society in general, have very low approval of most restrictive practices such as mechanical restraint, although interestingly not of seclusion (Reference Whittington, Bowers and NolanWhittington 2009). Therefore actions involving involuntary treatment and physical restraint, which do not fit with the traditional professional ideal of the ‘good’ carer, may be met with ambivalence. But UK preferences are not ubiquitous in psychiatry as a whole. Across Europe, the frequency and type of coercive interventions, and legal restrictions on them, vary widely: in the UK (Reference Raboch, Kalisova and NawkaRaboch 2010) seclusion is particularly common, but mechanical restraint is rarely if ever used (Reference Bowers, van der Werf and VokkolainenBowers 2007).

It is possible that there is still some lingering remnant of the ethos that allowed Mary Barnes to live for 5 years in the experimental therapeutic community at Kingsley Hall, London, smeared with her own excrement (Reference Bjarnason and EdgarBjarnason 2001); what commentators have called ‘rotting with their rights on’ (Reference GutheilGutheil 1980). Unconscious disgust and fear can lead to staff avoiding action in the name of respect for liberty and freedom, a freedom that people who lack capacity cannot enjoy or exercise.

Clozapine – unconscious heuristics?

There may also be unconscious anxiety and suspicion about the use of clozapine. Despite clozapine's positive evidence base, it is not used involuntarily like other medication in mental healthcare and arguably remains underutilised (Reference Mistry and OsbornMistry 2011; Reference Stroup, Gerhard and CrystalStroup 2014). A UK survey reported that, where clozapine had been enforced, restraint to take blood was uncommon, and a significant minority of professionals viewed the use of clozapine under restraint as ‘always wrong’ (Reference Pereira, Beer and PatonPereira 1999a,Reference Pereira, Beer and Patonb). One psychiatrist has argued that studies of involuntary clozapine should not have been published at all, suggesting (possibly tongue in cheek) that they were ‘instructions in martial arts for psychiatrists’ (Reference FahyFahy 2000). Other commentators have offered technical objections, or recommended involuntary ECT, which appears to be at least as restrictive, invasive and hazardous (Reference BarnesBarnes 1999), especially in patients like Ben, whose physical health was compromised by dehydration and starvation. In some professional minds, the chief objection to forced nasogastric administration of clozapine appears to be the association with the protests of the oppressed, evoking historical images of the treatment of political prisoners, suffragettes or detainees in Guantanamo Bay.