Abstract
See Review by Berk et al., 2012, 46(2): 92–99
Berk et al, 2012 propose a palliative care model for individuals in the later stages of mental disorder. The purpose of this commentary is to critically appraise the case they make, and to suggest reasons why a palliative care model may not be in the best interests of such patients.
The focus of this model is those patients with severe mental illness who either do not respond or who respond only partially to front-line treatments. It is an unfortunate fact of life that there are a significant proportion of patients with severe illness who remain with considerable problems despite considerable therapeutic effort. It is suggested that a palliative care model will protect such patients from aggressive and high-risk treatments that have a low likelihood of success, which could in turn have a demoralising effect on the patient and family. Thus the ostensible aim of the model is to spare the patient pointless harm and disappointment.
This line of reasoning relies on a particular understanding of the longitudinal course of the illness, for which the concept of staging is thought to be relevant. It is here that the first cracks appear in the edifice. Staging has much to offer in certain physical illnesses, where criteria are independent of outcome; in these cases the later stages can be empirically shown to be associated with very low chances of cure or recovery, and preparations for the expected end justify a palliative approach. Several of the authors have previously aired the idea of staging for mental illness (references 7 to 10 in Berk et al., 2012). The first problem is that as yet there is no agreed set of stages. At best, a robust and agreed set of stages needs to be established before we can justify applying palliative thinking to the later stages. Secondly, to be heuristic or predictive, stages must be definable by features other than treatment response – indeed one of the main values of a stage is that it can guide treatment. To base staging on treatment response defeats this purpose, and represents a clear circularity of argument.
A further problem with the staging idea is that it assumes an ordered set of states, and it is by no means clear that this exists for all or most individuals who have been ill for a long time. Among those whom we call chronically and severely mentally ill there are generally different courses discernible for those with schizophrenia (flat or downward) and those with affective disorders (episodic). Is the same set of stages meant to apply to both groups? In contemplating longitudinal course it is important not to confuse deterioration with chronicity. The former implies a progressive increase in severity or loss of function, while the latter implies a steady state of either severity or impairment, or both. The chronic patient may not be deteriorating, and may indeed be fairly steady, particularly when adjustments for age and iatrogenic factors are made. What may be following a downward trajectory in this case are the enthusiasm, energy and confidence of the clinician(s) to find an effective treatment.
So, while staging may not be the appropriate model, it still makes sense to adjust and adapt treatment according to the patient’s requirements. If a palliative model should spare a chronic patient burdensome and complex treatments, then one would expect it to be simple and basic. However, the potential interventions listed in Table 1 of Berk et al. (2012) are cumulative: each stage adds a number of interventions to those of the earlier stage that the patient has presumably already traversed, so in the final stage the poor patient is being subjected to everything that has ever been tried. This is of course quite the opposite of what a palliative model is supposed to deliver. If the authors are serious about sparing the patient burdensome treatments then they need to specify what they would no longer do. Looking at what should be provided to the chronic patient, the authors list such things as the setting of attainable goals, reduction of side effects, targeting of psychological and social problems, dealing with comorbidities, and maximising quality of life. But surely these objectives are relevant to any patient at any stage of their illness. As noted in Wikipedia ‘palliative medicine is appropriate for patients in all disease stages’ (Wikipedia, 2011). The lack of specificity of palliative care for the chronically mentally ill seriously reduces the utility of the concept.
It is fair to say that the authors are sensitive to the accusation that thinking palliatively about the chronically mentally ill will be perceived as defeatist, and so they are careful to hedge their argument with several caveats. One of these is that the model should ‘only be applicable to those who have truly exhausted appropriate algorithm- or guideline-based care’. Local guidelines for schizophrenia say that ‘Comprehensive psychosocial interventions should be routinely available to all patients and their families, and provided by appropriately trained mental health professionals with time to devote to the task’ (Royal Australian and New Zealand College of Psychiatrists Clinical Practice Guidelines Team for the Treatment of Schizophrenia and Related Disorders, 2005: page 2) and the NICE guidelines (National Institute for Health and Clinical Excellence, 2009: page 9) advise that cognitive-behavioural therapy should be offered to all patients with schizophrenia. We may reasonably ask how many actually receive such interventions; the evidence is not encouraging. A recent survey of Australians living with a psychotic illness (Morgan et al., 2011: page 74) found that only 22.3% had used CBT in the past year; this was one of several findings that led the report to conclude that ‘the proportion of participants accessing psychosocial therapies was relatively small’. Thus we must infer that even the authors, using their own criteria, would have to exclude many chronic patients in our current mental health system from a palliative form of care.
The authors are also sensitive to the negative connotations of the word palliative, and they take some pains to counteract the impression that palliative equates to extinguishing hope. But the terms hope and hopelessness are used inconsistently. In some places, maintenance of hope refers to the expectation of good care, but in others it refers to the expectation of relief from psychotic symptoms. The message appears to be that, for these individuals, we should promote hope for good care but downplay hope of a cure. Such nuances are probably beside the point because for most clinicians the word palliative will evoke associations with terminal illness and end-of-life care. Therefore the authors counsel that the term needs to be ‘used with caution, if at all, lest a message of hopelessness is interpreted’ (Berk et al., 2012, page 97). How are we to understand the phrase ‘if at all’? There is a recognition that in the age of the internet patients, and their family members, representatives and advocates, will have access to the professional discourse. By hinting that maybe the term should not be used at all, or maybe substituted for some other (as yet unspecified) term without clear negative connotations, the authors practically concede that, even if the concept of palliative care for this population had some merit, the word itself carries too much baggage to be helpful.
At this point it is worth reflecting on the likely thought processes of a patient who has just learned that their treating clinician has recently allocated them to a palliative category or stage. This is particularly pertinent because perhaps the main basis of patients’ expectations is their perception of their doctor’s expectation. There is evidence that staff generally underestimate patients’ expectations of the non-tangible elements of service quality; one study (O’Connor et al., 1994) found patients held significantly higher expectations of service responsiveness (willingness to help and provide prompt service) than physicians perceived them to hold. The result, which was subsequently replicated and extended to medical and nursing students (O’Connor et al., 2000), led to the conclusion that healthcare staff need education in ‘how much value patients place on particular dimensions of service quality, such as responsiveness and empathy’ (O’Connor et al., 1994: page 33). The potential damage done to the expectations of patients (and their families) by them detecting, rightly or wrongly, that they have been placed into an ‘incurable’ category, would be most unfortunate and difficult to repair. It is interesting that a palliative model is being proposed at a time when recovery appears to be the emergent paradigm, since they appear to be at odds with each other. While recovery is mentioned in the title, it only appears twice in the text, and tangentially at that. Hope is a central component of recovery (Slade, 2009), definitions of which usually revolve around leading a satisfying life despite ongoing symptoms. It is interesting therefore that at least some people with chronic schizophrenia (presumed candidates for the palliative stage) believe that full recovery could not be said to have been achieved until they stopped medication and had a steady job (Ng et al., 2008).
Perhaps the most serious objections to the palliative model as presented are not so much scientific and theoretical, but values-based and ethical. The paper implies that active multimodal treatment of the chronically ill, by being ineffective, is also wasteful. But waste can occur at both ends of the early–late spectrum. Just as treatments with a low chance of success may be given unnecessarily to patients in the late stage, so also can expensive treatments be given unnecessarily to early cases who are likely to recover anyway, or even to people who only have elevated risk of developing illness. Ominously, the authors imply that high costs of treatment of the chronically ill may constitute a ‘misallocation of finite resources’ that may ‘reduce access to care for other patients in need’ (Berk et al., 2012, page 95). This sounds like the preparation of a case to transfer resources away from the chronically ill (who will get predominantly palliative care) and toward the early/first episode sector, where, on average, current pharmacological treatments have better results. The choice of how a health service, and ultimately a society, chooses to distribute its resources is not simply a financial calculation – it is a question of values. Contrary to the palliative model, it could be argued that this group of patients deserve extra, not less, effort to find effective treatments.