Can deception ever be justified on therapeutic grounds?

Case history

The behaviour of X in childhood had been described as ‘difficult’, ‘aggressive’ and ‘bizarre’, including bizarre gestures, mannerisms, ‘making scenes in public’, ripping her clothes and trying to set fires.

In early adulthood she had had a number of placements in establishments for people with learning disabilities. Major difficulties had occurred in managing her behaviour in all these settings and she had had repeated hospital admissions, followed by attempts at placement in the community. She exhibited preoccupations, repetitive questioning, periods of agitation (during which she exhibited paranoid ideas and auditory hallucinations), high-pitched screaming, violent threats towards staff and residents (if her demands were not met) and aggression towards herself, property and other people. These behaviours were unchanged despite treatment with behaviour modification and psychotropic medication.

In her 30s, following a failed hospital placement, she was admitted under Section 3 of the Mental Health Act 1983 to a local psychiatric unit, and later transferred to an open forensic unit for people with mild/borderline learning disability (Johnson et al, 1993; Smith, 1988). Reassessment with the Wechsler Adult Intelligence Scale (revised) gave her a full-scale IQ of between 64 and 77. A diagnosis of autism was confirmed.

Her quality of life on the admission ward was poor. Despite this, she would often say repetitively “I am never going to leave this hospital, am I doctor?”, with increasing agitation. Her anticipatory anxiety prior to visits to potential placements was so extreme that it resulted in serious aggressive outbursts. Her resistance to change, a feature of her autism, was so severe that even the suggestions of looking at alternative accommodation caused distress and violence. Her resistance to change left her trapped in an unsuitable long-term environment — she remained an in-patient on the admission ward for several years.

She was assessed for placement in a new small unit designed specifically for people with autism. It was the clear, unequivocal and multi-disciplinary consensus that the environment would be an appropriate long-term placement that would both enhance her quality of life and prove therapeutic, but it was known that she would be highly resistant to the transfer. An intense period of discussion occurred between all staff concerned. A multidisciplinary meeting was held to discuss the ethical issues surrounding not informing her of the move. The decision was made not to tell her.

Several weeks prior to the move staff from the new unit came to work alongside her ‘on placement’. They became familiar to the patient. On the day of the move her room was recreated in the new unit using her personal belongings while she was elsewhere in the hospital grounds. She was told that she was going on a trip. When she arrived at the new unit there were several staff with whom she was familiar. She entered the unit in a calm and dignified manner. When she saw her room she was reported as saying “I like it here, this is where I am going to stay”.

One year after the move there have been no incidents of aggression. Her level of self-care has improved and her behaviour has been manageable.

Ethical discussion

This centres on the issue raised from moving a detained patient between facilities without his or her consent or even knowledge. It raises a number of issues, most importantly: is consent needed to move a detained patient, when is it right to withhold information from a patient and should residence be considered part of treatment?

Consent is important when considering issues of treatment. It is not clear if transfer to another facility (ward, hospital or hostel) is an issue of treatment or rather one of security, geography, quality of life or bed availability.

With voluntary patients consent is generally assumed by their acquiescence to a move. The legal purpose of consent is to provide those concerned with a defence against assault, battery or trespass to the person (Lord Donaldson MR in ReJ (a minor) (medical treatment), 1992). For detained patients, it is their legal detention that provides the defence against these charges and false imprisonment. Once detained under the Mental Health Act, patients have a right to receive appropriate care but no right to determine where that care is undertaken (e.g. a patient may be detained outside his or her home area and be transferred to the local unit even if against his or her wishes). Good practice implies that we should inform patients, but do we have a duty to inform?

This is at the heart of this case. It is about a patient's basic rights. It is unusual to withhold information from a patient, even if he or she is detained under the Mental Health Act, although the act itself allows provision for non-disclosure (Department of Health, 1996). The danger is that in withholding information, even though not directly lying, we may be effectively deceiving our patients.

Deception is generally seen as wrong. It undermines the trust element of the doctor—patient relationship and questions the independence and presumed competence of the deceived individual. Despite this, deceptive practice has a long history in medicine (e.g. placebos). It is generally justified on therapeutic grounds, usually on a consequentialist argument (e.g. good outcome justifies the means). Therefore, if one is going to deceive a patient it needs to be justified in much the same way as any other potentially harmful procedure in medicine. The General Medical Council position is clear:

“Patients have a right to information about their condition and the treatment options available to them… You should not withhold information unless you feel the information would cause serious harm — serious harm does not mean the patient would become upset.” (General Medical Council, 1999).

Information was withheld in this case on the grounds that it would have caused her serious harm, not just upset. The serious harm was clinically judged to have been a severe deterioration in her mental state, which would have led to the loss of a suitable well-designed placement and her continued residence in an unsuitable, anti-therapeutic and possibly deleterious ward environment.

There is, of course, the inherent difficulty in applying a consequentialist argument to an incompetent patient. Who makes the assessment of the possible harm and potential benefits of a course of action? Normally the patient assesses the risk—benefit equation, but this is not possible when a patient is incompetent or the harmful act is non-disclosure. In this case we thought the patient to be clearly lacking the capacity to consent to the move and acted in what we believed were the patient's best interests. What we did might be defined as a paternalistic deception (Bok, 1978). In our case we saw this as a clinical decision that was taken and agreed by both multidisciplinary teams. We might have been criticised for not asking for an independent ethical opinion before proceeding along this path. However, yet another opinion would have been by an individual who did not have the knowledge and understanding of the patient that our team had acquired through 5 years of care and would have no more right to decide for the patient than the multi-disciplinary team. There is currently no court of protection with regards to overseeing medical decisions of incompetent patients.

In many ways this case demonstrates the limits of the classical bioethic model. This has evolved around short-term treatment packages rather than what is often the reality of long-term incapacity and questions of quality of life and on-going relationships, of disability rather than disease (see Agich, 1999).

Nevertheless, current ethical and legal opinion generally views living environment as part of the treatment and therefore subject to the same safeguards as other treatments. Indeed, social milieu is often the primary treatment in autism (Larkin, 1997). If residence is therapeutic then transfer may also be considered therapeutic as it necessitates the patient arriving in a suitably prepared state to undergo the treatment. Therefore, not only residence in the unit but also the process of transfer to the unit can be seen as part of this patient's therapy. It is therefore subject to the guidance on the treatment of psychiatric patients contained in the Declaration of Hawaii. It states:

“No procedure shall be performed nor treatment given against or independent of a patient's own will, unless because of mental illness, the patient cannot form a judgement as to what is in his or her own best interest and without which treatment serious impairment is likely to occur to the patient or others.” (World Psychiatric Association, 1981)

We assessed that her autism meant that she would not understand information presented to her in a clear way or weigh it in the balance (as shown in Re C Adult Refusal of Medical Treatment, 1994). She therefore lacked capacity to consent and was unable to see what was in her best interest. It could be argued that in her case consent was meaningfully given by proxy through her compliance with our treatment plan. This raises the issue of presumed consent in incompetent individuals, as recently discussed following the Bournewood case (L v. Bournewood Community and Mental Health NHS Trust, 1998). Applying this to our case, her acquiescence to the move implies that the patient seemingly consented to a care plan involving her own deception.

Conclusion

We judged that the patient's transfer to a specialist autistic unit constituted a therapeutic procedure, which would ordinarily have required informed discussion. In view of her mental disorder she lacked the capacity to consent and we made a substituted judgement that the move was in her best interest. Furthermore, we thought that withholding information about the move was justified in that she would have suffered serious harm to her mental state had we informed her and this could have damaged the therapeutic value of the move. We believe that the approach was justified and ethical and would have remained so had the placement not gone successfully. Its success is a reflection of the accuracy of our 5-year long clinical assessment.