Current guidelines on placebos

At the outset we can note that, in practice, physicians are not resistant to prescribing placebos. A recent survey of US internists and rheumatologists, for example, indicated that around 50% of respondents regularly prescribed placebo treatments and 62% believed that the practice was ‘ethically permissible’.5 In addition, 68% of those physicians who used placebos described them to patients as ‘potentially beneficial medicines which were not typically employed in treatment’; in only 5% of cases did physicians explicitly describe the form of treatment as a placebo. This suggests a sort of intuitive moral sensibility and/or pragmatic stance on the issue among physicians, regardless of stipulated medical codes.

The GMC proffers no specific guidelines on placebo use, rendering this a predicament which physicians must resolve on their own. The AMA, however, attempts to clarify the issue, but its statement provokes more questions than it answers. It stipulates that ‘the use of a placebo without the patient's knowledge may undermine trust’ and ‘compromise the patient–physician relationship’.6 What can it mean to request consent before employing a form of therapy that is inherently illusory? The AMA guidelines continue (and this is the rub): ‘A placebo may still be effective if the patient knows it will be used but cannot identify it and does not know the precise timing of its use … The physician need neither identify the placebo nor seek specific consent before its administration’. This raises two issues: (1) the question of how a placebo can be a ‘placebo’ if it doesn't deceive; and (2) whether suppressing details of placebo use can be consistent with those resolute stipulations which speak against withholding information.

Turning to the latter issue, it seems clear that there is an irresolvable tension in the AMA's guidelines: deceiving patients about placebo use seems, on any level, to be a flagrant dismissal of the imperative that physicians be wholly honest. Prima facie, the first issue presents a way to resolve this: if placebos do not involve deception, then there is no such ethical dilemma. However, this is curious: if placebos don't involve deception, why can the physician not identify the placebo and the timing of its use? More to the point, what can it mean for a patient to expect treatment from a form of ‘medication’ that is not in itself curative but for this not to involve deception?

The notion of ‘non-deceptive placebos’ draws on limited and controversial research that indicates that divulging to a patient that he is receiving a placebo need not lessen its effectiveness.7 This research appears to show that patients are responding to placebos as a result of conditioning: taking pills or being injected with medication are acts which we are strongly conditioned to associate with recovery. Indeed, other well-supported studies reveal that the type of placebo medication even has an impact on the size of the response: smaller sugar pills are less effective than larger ones8; pills taken four times a day are more effective than those taken twice a day9; and pills make for less effective placebos than saline injections.10 However, the ‘non-deceptive placebo’ theory is problematic for at least two reasons. First, no control was employed in the original study in 1965 (and this work has never been replicated); second, it is not clear whether subjects today would not be more sophisticated (ie, sensitive) in response to the revelation that they were being administered a ‘placebo’. It seems that the AMA is drawing on insubstantial research in order to ‘licence’ clinical and ethical employment of placebos.

In summary, more robust research is needed to ascertain whether it is somehow possible to prescribe placebos without deception. Given that placebos are deemed to be effective (to varying degrees) in the treatment of various conditions, how should ethics committees and physicians proceed? There are two choices available, given current (reliable) research. In the interests of consistency, ethical bodies can either counsel that placebos do involve deception and therefore should never be employed,11–14 or admit that deception is involved but assert that placebos are too important to be eliminated from clinical practice.15–17 In the following sections I further investigate the nature of deception in clinical practice by examining the treatment of depression.

Depression and placebos: emerging concerns

In this section I argue that the limited efficacy of antidepressants is a serious issue but one that has started to gain the attention of the medical community. I argue that nascent but highly significant research adds further importance to responding carefully to the issue of deception.

Depression is deemed to be the most prevalent mental disorder in the world: in the USA alone, diagnostic rates are estimated at around 10% of the adult population per annum.18 Antidepressants (such as Prozac, Effexor, Zoloft and Citalopram) are the treatment of choice among physicians in the USA; in any month 10% of women and 4% of men are prescribed antidepressant medication19 and more than 31 million prescriptions were written for antidepressants by physicians in the UK in 2006.20 Clearly, any research that questions the effectiveness of these drugs demands serious attention by medical professionals.

In fact, recent work reveals disquieting statistics on the efficacy of antidepressants. In noteworthy meta-analyses of antidepressant usage, it has been argued that any difference between placebos and antidepressants in clinical settings is due to the significant side effects of antidepressants—that is, patients and physicians are able to intuit that sugar pills have not been administered and this, it is argued, undermines the requisite research standard of ‘double blind’ trials.21 22 Indeed, the greater the number of side effects experienced by patients in such trials, it has been observed, the more patients improved. Furthermore, when trials included ‘active’ placebos (placebos which induce similar antidepressant side effects), no clinically significant disparities in the effectiveness of placebos and antidepressants were seen to occur.23 24 This research is singularly important because, with respect to National Institute for Health and Clinical Excellence (NICE) criteria, it indicates that antidepressants are not more effective in a clinical setting than placebos.25

While it should be reiterated that further research is needed to ascertain whether the mechanism of action of antidepressants is the same as for placebos, these findings raise potential issues of concern for medical ethicists and physicians. Should it be discovered that antidepressants function as placebos, then it is possible that they have already (unwittingly) been employed as a form of ‘deception medication’. On the other hand, if antidepressants are not in fact, placebos, given the well-documented severity of side effects associated with them, should physicians be prescribing sugar pills instead? Although this latter scenario may involve deception on the part of the patient and physician, is this preferable to treatments with extensive common side effects (drowsiness, low sex drive, weight gain, and so on), more especially when there is a possible link between suicidal ideation and such medication (notably in the case of paediatric treatment)?26–28 Research into the effectiveness and mechanisms of action of antidepressants highlights the importance of resolving the tension between patient deception, on the one hand, and provision of the utmost medical care, on the other; indeed, there is a pressing need for more studies in this area. However, before I address the issue of alternative treatments and how they might work, we need to consider the issue of precisely what it is we want our treatments to achieve.

Positive illusions and ‘depressive realism’

Is deception an irrevocable part of life? Well-replicated evidence shows that positive illusions—that is, positive unrealistic self-evaluations, an inflated perception of one's control over events and an exaggerated optimism about the future (what psychologists dub the ‘Pollyanna principle’)—are characteristic of good mental health.29–31 The health-promoting effects of these positive illusions include an increased capacity to care for others and an enhanced aptitude for creative and productive work. Indeed, individuals who display the positive illusion that they are better off than others are less likely to develop depression.32 This research is fundamental to how we conceive of mental health. If individuals deviate from these illusions—that is, these forms of positive deception—it seems that it is the job of medical treatment to restore them.

Indeed, the notion that depressives exhibit realistic cognition when it comes to self-evaluation and their mastery of events is one that has been mooted in the psychological literature.33 34 Assuming the correctness of such a hypothesis, and in the interests of upholding current medical ethical guidelines, it would appear that physicians ought not even to treat depression. Clearly, this is an absurd consequence, but it further underscores the importance of a more penetrating revision of prevailing views on excluding deception from medical treatment.

In fact, growing evidence supports the view that depressives display illusions on the opposite end of the spectrum from the mentally well (they appear to exhibit negative illusions about themselves and their future prospects35 36); it would appear that the onset of depression involves a swing towards pessimistic biases from normal positive cognitive biases.37 However, regardless of the standing of these hypotheses on depression, neither hypothesis affects the prevailing consensual view held among social psychologists that mental functioning involves deception. Counterintuitive as it may seem, research indicates that it is the job of physicians to redress this imbalance by restoring positive illusions in patients.

Conclusion: a spoonful of deception helps the medicine go down?

Returning to the issue of successful treatment, what are the alternatives to antidepressants? Meta-analyses show that cognitive behavioural therapy (a form of psychotherapy) is the most successful long-term treatment for depression.38 39 This is a form of therapy which encourages patients to identity negative thought patterns and subsequent behaviour and to consider whether such thought patterns and responses are useful or helpful to them. Over a long period of time (many months), patients are persuaded to adopt thoughts which are more ‘realistic’ (but which are, in fact, moderately positive) and which induce different behavioural responses to their usual course of action.40 41 Individuals who successfully complete such forms of treatment end up endorsing positive illusions about themselves and the world around them. They often consider that their illness has ‘happened for a reason’ and that it has given them an opportunity to ‘grow’ as individuals; indeed, the notion that they are fortunate to have suffered appears to extend to a downward comparison with those who have not had the chance to experience such things.30 More than this, these psychological therapies involve a more prolonged form of deception than placebos; any deception about the efficacy of prescribed sugar pills pales when contrasted with the promotion of highly personal deep-seated illusions about oneself that are induced in the successful treatment of a patient with depression. In short, by finding meaning in misfortune and integrating it with positive views on oneself and the benevolence of the world, individuals heal.42 One psychologist has dubbed this form of therapy the ‘ultimate placebo’.24

In conclusion, given the intrinsic importance of positive illusions to psychological well-being, medical ethics committees need to be more candid about the importance of placebos in clinical practice. We await robust research to the contrary of the current prevailing view that placebos involve deception. Moreover, treatment for depression cannot even be undertaken without the active endorsement of positive illusions on the part of the medical community. Medical bodies need to accept that a spoonful of deception may be fundamentally (and unavoidably) therapeutic.