Advance directives in mental health care are currently attracting interest, although there is some anxiety that they can restrict clinical freedom. The so-called ‘Ulysses contract’ is a form of opt-in to services that has been suggested in the USA. Psychiatrists might thus consider themselves to be the equivalent of Ulysses’ crew in being bound by the contract. This paper suggests, in some cases, that they might function more as Circe, who suggested the directions to Ulysses, and considers this in the light of contemporary relationships between psychiatrists and patients.

With these words, Ulysses (or Odysseus) instructed his crew as to their actions towards him as they cross the path of the Sirens and their wondrous songs, and created an early form of advance directive. Echoing his directions, the name ‘Ulysses contract’ was given to a form of advance directive in mental health in the USA (Dresser, Reference Dresser1982, Reference Dresser1984; Reference Winston, Winston and AppelbaumWinston et al, 1982), also termed an ‘opt-in’ advance directive, whereby patients agree in advance to treatment they may refuse later when ill. Although in Britain the emphasis probably has been on advance directives that opt out or refuse treatment, both ideas are current.

However else they are interpreted, an advance directive has at least two participants: the person who draws it up and the person(s) who must implement it. Advance directives, whether seen as enhancing patient autonomy or promoting partnerships and collaboration demand a mutual understanding of the directive and an appropriate response, whether this be Ulysses’ crew or a patient's psychiatrist. Psychiatrists have expressed concern about the implementation of advance directives, seeing them as a limitation to clinical judgement (Scottish Parliament, 2002a ). Ulysses’ advance directive was successful in that he and the crew navigated the hazard (i.e. the Sirens). Does an analysis of Ulysses’ story help us to understand contemporary relationships and point to successful outcomes?

Ulysses’ advance directive was noteworthy in a number of respects:

1. 1. The idea for the plan and its instructions did not originate with him, but with Circe, a known expert ^{Footnote †} on the Sirens.

2. 2. He wanted to experience something known to be dangerous, indeed almost certainly lethal, but to have a safety net.

3. 3. He required something to be done to him – namely restraint.

4. 4. He required something not to be done to him – his ears not to be plugged with wax.

5. 5. He made provision for a change of mind – he was to be ignored and restrained further.

6. 6. Conditions were put on the crew – their ears were to be plugged with beeswax.

7. 7. He and the crew were in agreement about the goal – to come safely past the Sirens.

8. 8. The plan did not require resources beyond those that the crew already had.

If asked to rank these, it is likely that point 7 would be important for staff: agreement about the desirability of outcome. Staff are likely to be happier to implement an advance directive if they believe that both the patient and they will have a good outcome. In the case of Ulysses, if points 3 and 5 were followed no harm would come to him. In many cases, staff fear that a refusal of treatment (point 4) will lead to harm for the patient and possibly negative consequences for themselves, whether this is in the difficult management of an ill patient or the extra call on limited resources by a longer period of illness (Reference Halpern and SzmucklerHalpern & Szmuckler, 1997). A recent study found advance directives to be of limited value based on staff-selected group outcomes, but individual patients may have found them empowering (Papaggergiou et al, 2002). It is, however, the potential for additional resources that concerns many in allowing patients to run the course of their illness episode; discussion of resource allocation is outside the scope of this article, but it will have an impact on attitudes.

The crew may have had faith in the plan because of its provenance – the goddess Circe. Advance directives based on experience and expertise may thus find more favour with staff, whether presented independently or drawn up with staff, than advance directives based on things that a patient believes but has not experienced. There is, arguably, a difference between someone who has experienced severe illness and electroconvulsive therapy and found it so distressing that they do not want it again, and someone who has experienced neither and is basing their decision on media reporting.

Although Ulysses’ directive is usually seen as opt-in, in that he required action to be taken, it can also be seen as an opt-out of what is arguably the more conservative option, namely plugging his ears with wax along with his crew and thus avoiding being bewitched by the songs. Ulysses thus wishes to experience something that, although dangerous, is potentially enriching and put in place a safety net to keep him from harm. Patients who choose to experience an episode of illness (whether they believe it to be an enriching experience or whether it is simply less bad than the alternative treatment) may need to be aware of what safety nets, including restraints, they are prepared to put in place to manage the illness episode. Staff may be more comfortable with advance directives that acknowledge the consequences of this and make provision for them. We are not told of the crew's response to having to restrain their captain, but the negative consequences to staff in using restraint have been noted (Scottish Parliament, 2002b ).

It seems clear from Ulysses’ account that he did not want to die. The limits of advance directives need to be stated. Advance directives made for end-of-life scenarios are clear about the outcome. Those in mental health may be less so, and no illness or episode is completely predictable. Directives should at least cover whether a person is prepared to die, or how long they are prepared to remain acutely ill before starting treatment.

Ulysses’ advance directive worked because it was, in effect, a limited forward plan that had the agreement of those who had to implement it. It dealt with a very specific set of circumstances, with known outcomes and agreed goals. It is likely that the more an advance directive conforms to these parameters, the more acceptable it will be to staff. It also had one overriding advantage that current advance directives do not. Ulysses’ was in charge, the crew was his and he had already brought them safely through many dangers. Advance directives necessarily challenge the authority from which staff usually operate. Psychiatrists may feel limited if they put themselves in the position of Ulysses’ crew and may prefer the role of Circe who proposes the plan. A proactive approach to future planning may promote patient choice and reassure psychiatrists that the plan is workable.

Advance directives require changes on the part of both staff and patients. Staff would have to accept not only patients’ choices, but the experience that leads to these choices. Patients would need to accept the responsibility that comes with having their choices honoured, even if these choices do not always have the expected outcomes. After all, Ulysses knew that passing the Sirens would only lead to the twin dangers of Scylla and Charybdis. And, in this respect, both patients and staff can find themselves between a rock and a hard place when balancing patient choice, uncertain outcomes and limited resources.

Declaration of interest

Dr Atkinson is in receipt of a research grant from the Nuffield Foundation to research advance directives in mental health.