Article Text

Download PDFPDF

Psychotherapy, placebos, and informed consent
  1. Garson Leder
  1. Alden March Bioethics Institute, Albany Medical College, Albany, NY 12208, USA
  1. Correspondence to Dr Garson Leder, Alden March Bioethics Institute, Albany Medical College, Albany, NY 12208, USA; leder.garson{at}


Several authors have recently argued that psychotherapy, as it is commonly practiced, is deceptive and undermines patients’ ability to give informed consent to treatment. This ‘deception’ claim is based on the findings that some, and possibly most, of the ameliorative effects in psychotherapeutic interventions are mediated by therapeutic common factors shared by successful treatments (eg, expectancy effects and therapist effects), rather than because of theory-specific techniques. These findings have led to claims that psychotherapy is, at least partly, likely a placebo, and that practitioners of psychotherapy have a duty to ‘go open’ to patients about the role of common factors in therapy (even if this risks negatively affecting the efficacy of treatment); to not ‘go open’ is supposed to unjustly restrict patients’ autonomy. This paper makes two related arguments against the ‘go open’ claim. (1) While therapies ought to provide patients with sufficient information to make informed treatment decisions, informed consent does not require that practitioners ‘go open’ about therapeutic common factors in psychotherapy, and (2) clarity about the mechanisms of change in psychotherapy shows us that the common-factors findings are consistent with, rather than undermining of, the truth of many theory-specific forms of psychotherapy; psychotherapy, as it is commonly practiced, is not deceptive and is not a placebo. The call to ‘go open’ should be resisted and may have serious detrimental effects on patients via the dissemination of a false view about how therapy works.

  • psychotherapy
  • informed consent
  • paternalism
  • ethics

Statistics from


The professional standards for informed consent to psychological treatment are often minimal and vague. For example, the American Psychological Association requires that practitioners performing ‘generally recognised techniques and procedures’ should use ‘reasonably understandable language’ to:

[I]nform clients/patients as early as is feasible in the therapeutic relationship about the nature and anticipated course of therapy, fees, involvement of third parties and limits of confidentiality and provide sufficient opportunity for the client/patient to ask questions and receive answers… (italics added).1

Similarly, the British Psychological Society recommends that patients:

[Are given] ample opportunity to understand the nature, purpose and anticipated consequences of any professional services or research participation, so that they may give informed consent to the extent that their capabilities allow.2

And we see comparable statements from the American Psychiatric Association and the UK’s Royal College of Psychiatrists.3 4 These organisations all maintain that patients should be given enough information to understand how therapy is supposed to work (ie, the ‘nature’ of psychotherapy). The traditional bioethics literature is not much more precise. A standard view is reflected in Beauchamp and Childress’s claim that respect for patient autonomy, minimally, requires that practitioners are obligated to disclose a ‘core set’ of information that includes ‘those facts or descriptions that patients…usually consider material in deciding whether to refuse or consent to the proposed intervention’ and ‘information the professional believes to be material’.5 The task, then, for practitioners is to identify what information about the nature of psychotherapy belongs in this ‘core set’ that should be disclosed to patients.

Informed consent to treatment

Recently, a number of authors have argued that the minimal requirements for informed consent are not being met in psychotherapy as normally practiced by psychologists, psychiatrists, counsellors and social workers. Notably, Charlotte Blease, Jens Gaab, Manuel Trachsel (and others) have offered a series of articles arguing that traditional psychotherapy is a likely placebo (or ‘placebo-related’) and that mental health practitioners have a duty to ‘go open’ to patients and disclose the role played by placebogenic factors in psychotherapy.6–8 According to Blease:

Patients are inadequately informed about how psychotherapy works…and, thus, their autonomy to make an informed treatment decision is being infringed on.8

Similarly, Gaab and colleagues state:

We recommend that the profession of psychotherapy goes open. No healthcare profession is entitled to gloss the rights of patients, nor the professional standards of adequate information and training provided to therapists.6

These claims to ‘go open’ are based on common-factors findings in psychotherapy. For many disorders, multiple forms of psychotherapy appear to be comparable in their efficacy (even if not always equally so).9 10 At minimum, this indicates that effective therapy need not be explicitly based on any particular theory or technique; therapies with differing, and often disparate, theories and techniques produce effective therapeutic outcomes. A prominent hypothesis explaining this multimodal efficacy is that therapeutic common factors (ie, general therapeutic ingredients common to multiple forms of successful therapy) mediate some, and possibly most, of the ameliorative effects in psychotherapeutic interventions.11 Numerous common factors have been postulated, and can roughly be grouped into five categories: client characteristics (eg, positive expectations and hope), therapist qualities (eg, the ability to cultivate positive client characteristics), change processes (eg, the acceptance of a theoretical rationale for the therapy on offer), treatment structure (eg, the delivery of concrete treatments and techniques) and therapeutic relationship (eg, the development of a working alliance between therapist and patient). While the relative weighting of the effects of common and specific therapeutic factors is controversial, the common-efficacy findings suggest that, at least sometimes, it does not matter what theory-specific rationale is being offered. Rather, the efficacy of therapy seems to depend, in part, on how the explanation is delivered, who is delivering it, and how it is interpreted by the patient.12

The advocates of the ‘go open’ argument claim that these findings support the view that psychotherapy is, or is likely to be, a placebo.6–8 While there is no universally accepted definition of ‘placebo’, this claim is based on the view that therapeutic placebos are treatments that work for reasons other than those specified by the theory grounding the treatment. Advocates of ‘go open’ argument often adopt Grünbaum’s distinction between ‘characteristic’ and ‘incidental’ ingredients to define therapeutic placebos.6–8 13 According to this view, a therapy (t) is a genuine (non-placebo) treatment for a target disorder (D) only if:

the therapeutic gain that ensued from t in the alleviation of D was due to those particular factors [ie, characteristic ingredients] in its dispensation that the advocates of t have theoretically designated as deserving the credit for the positive treatment outcome.13

Put simply, a therapy is a placebo if its efficacy is due to ‘incidental’ factors not designated by the theory grounding the therapy. So, according to Grünbaum’s definition, if the efficacy of therapy is due to ‘incidental’ common factors, rather than theory-specific ‘characteristic’ factors, then therapy is a placebo. And, according to ‘go open’ advocates, the common-factors findings support the claim that incidental factors deserve some, and possibly most, of the credit in the efficacy of therapy. The positive outcome of therapy, then, is likely to be placebogenic. It is this information that ‘go open’ advocates claim patients need to hear.

The ‘Go Open’ argument

The ‘open’ argument is straightforward: Informed consent in psychotherapy requires an accurate explanation of the causal mechanisms of change in therapeutic interventions. Therefore, it is argued, non-deceptive psychotherapy requires the disclosure of role the common factors play in therapy. This argument hinges on what counts as part of the ‘core set’ of information necessary for informed consent to treatment in psychotherapy. Minimally, this disclosure must include information that impacts patients’ ability to understand how any particular therapeutic intervention is supposed to work. And, according to ‘go open’ advocates, this causal explanation must include the role of the common factors in therapy. For example, Gaab and colleagues state:

If a patient asks, “How does [therapy] work?” it seems clear that this is a direct question about causal processes; “What is the real engine of treatment here?” When a therapist responds [by stating the theory-specific explanation] … then he or she has not directly answered the patient’s question…If we choose to sanction such oblique responses, we unwittingly endorse medical paternalism with the understanding that therapists are within their rights to tell their patients phony stories about how psychotherapy works.6

According to the ‘open’ argument, explanations for therapeutic interventions that focus exclusively on theory-specific treatment rationales are either withholding or ignoring an integral part of the nature of therapy (ie, they are ‘phony’ or ‘therapeutic fictions’). An explanation of the ‘real engine of treatment’, according to these views, must include the disclosure of the non-specific common factors. Therapy as it is commonly practiced, therefore, is argued to be deceptive. This deception can either be intentional (through a paternalistic overriding of the patient’s right to be informed), or unintentional (due to lack of awareness about the causal role of the common factors). Either way, ‘go open’ advocates argue that therapists have a moral obligation to end this supposed deception.

What, then, might an adequately informed therapy look like, according to the ‘go open’ paradigm? Blease offers us one example:

In addition to the construction of therapeutic fictions, evidence shows that if I speak to you in a positive, empathetic and encouraging tone of voice, if you have a high opinion of me as a health professional, and if I charge you a reassuringly expensive hourly rate, this will lead to therapeutic mind-body effects. Do you consent to these aspects of care?8

According to Blease, and other ‘go open’ advocates, something like this disclosure is ethically required for informed consent to treatment. While the specific wording may need to be refined (eg, ‘coherent rationale’ may be more preferable to some therapists than ‘therapeutic fictions’), the overall point is clear: respect for patient autonomy is supposed to require that the patient understands that the common factors explain part, and possibly most, of the effectiveness of therapy. To fail to do so is supposed to override the classic biomedical principle of respect for patient autonomy.7

If the ‘go open’ argument is correct, and if talk therapy normally involves deception, this is an ethical problem on a very large scale. And, if ethical deliverance of talk-therapy requires the dissemination of information that may undermine its effectiveness, this is also seriously problematic. This paper argues that the ‘go open’ project is flawed, and that there is no necessary moral problem with the delivery of theory-specific therapy as traditionally practiced. This paper makes two related arguments against the ‘go open’ claim: (1) practitioners (in most cases) need not ‘go open’ about the common factors in psychotherapy, because the ‘go open’ argument is based on a mistaken view of the mechanisms of change in psychotherapy, and (2) clarity about the mechanisms of change in psychotherapy shows us that the therapeutic common factors are consistent with, rather than undermining of, the truth of many forms of psychotherapy. The call to ‘go open’ should be resisted, and may have serious detrimental effects on patients via the dissemination of a false view about how therapy works.

The common factors and theories of psychological healing

The ‘go open’ argument errs in assuming that the common factors provide an explanation for the efficacy of therapy. The argument confuses the possible mediators of change in psychotherapy with mechanisms of change. Mechanisms of change in psychology are ‘the processes or events that are responsible for the change’ and are ‘the reasons why change occurred or how change came about’; the mediators are the variables that are statistically correlated with this change.14 The common factors may be mediators of change, but they are not necessarily mechanisms. The ‘go open’ argument is correct in stating that informed consent requires that patients be given an explanation to the ‘How does it work?’ question, however it is a mistake to think that providing patients with a list of therapeutic common factors achieves this. Rather, the answer to this question requires an explanation for why the presence of any therapeutic ingredient (specific or common) may be salubrious. Therapy, for example, does not work because a therapist charges ‘a reassuringly expensive hour rate’ or speaks in an ‘empathetic and encouraging tone of voice’. Knowing what therapeutic ingredients may be salubrious does not explain why they are salubrious. An explanation of the ‘real engine of treatment’ will need to provide the patient with an account of the mechanisms of change that are responsible for the efficacy of treatment. And, crucially, a number of theory-specific therapies attempt to do just this.

Consider, for example, cognitive therapy, one of the most widely applied and studied therapies in the Anglophone world.15 Cognitive theories of psychopathology posit that mental disorders are primarily caused by maladaptive cognitions (ie, thoughts, beliefs, and assumptions), and that the alteration of cognitions is the primary mechanism of change in psychotherapy.16 Cognitive therapy for specific disorders focusses on theory-specific therapeutic techniques, such as identifying and challenging maladaptive thoughts and beliefs and training patients to challenge maladaptive patterns of thought (eg, all-or-nothing thinking, catastrophising, and overgeneralisation). However, a cognitive theory of healing (ie, the theoretical explanation for why therapy is effective), need not posit that any particular theory-specific techniques are necessary for successful therapy.17 For instance, Aaron Beck, one of the founders of cognitive therapy, is explicit that cognitive theory is supposed to be compatible with the common factors findings and the use of multiple therapeutic techniques. Beck states:

I believe the ‘common factors’ among the various psychotherapies rely primarily on cognitive change…. Cognitive therapy is best defined in terms of the theoretical structure and presumed mechanism of action rather than the techniques derived from it… (italics added).18

According to Grünbaum’s formulation, a therapy would be a placebo if its efficacy was due to incidental factors that were not ‘theoretically designated as deserving the credit for the positive treatment outcome’ by advocates of the therapy.13 This is not the case here. Beck’s cognitive theory, rather than being necessarily undermined by the common-factors findings, is supposed to explain the efficacy of the therapeutic common factors. Cognitive therapy is hypothesised to work because it alters maladaptive cognitions; this is hypothesised to be done either directly (via cognitive therapy) or indirectly (via other therapeutic techniques). Note that this paper is not advocating for the plausibility of cognitive theory (or of any other theory of psychopathology); the point here is that the causal role of the common factors in the delivery of psychotherapy does not necessarily undermine the truth of a theory-specific explanation for the efficacy of psychotherapy. It can be true that both (1) theory non-specific common factors mediate some, and possibly, most of the efficacy of psychotherapy, and (2) theory-specific causal explanations for therapeutic healing may be accurate. In contrast to the ‘go open’ argument, theory-specific explanations for the ‘How does it work?’ question need not be deceptive and need not be placebogenic.

The ‘go open’ authors argue that the common-factors findings show that ‘incidental’ (ie, placebogenic) treatment effects are responsible for some or most of the efficacy of therapy,6–8 but this is a mistake. As the cognitive therapy examples shows, the common-factors findings are consistent with theory-specific explanations of healing; the common factors can be part of a theory’s characteristic ingredients. For example, in cognitive therapy it is possible for it to be true for a therapist to make the theory-specific claim that ‘Therapy works by modifying maladaptive core beliefs’, while it also being true that common factors (eg, alliance effects, expectancy effects, and rational-based effects) are the mediators that enable this success. The primary mechanism of change, according to cognitive theory, is the modification of cognitions and cognitive patterns; the common factors are merely possible hypothesised mediators for this change.

And, importantly, other theories of healing are (at least in principle) consistent with the common-factors findings. It is open to practitioners of other therapies to argue that if any technique is effective in therapy, it is only successful insofar as it engenders the mechanisms of change posited by that specific theory. The answer to the ‘How does therapy work?’ question requires an explanation of the hypothesised mechanisms of change in psychotherapy. The question is not answered by providing the patient with a list of possible mediators of this change (be they incidental constituents or theory-specific techniques); the common factors are part of the explicandum that a theory of healing is supposed to explain. Many theory-specific psychotherapies (including cognitive therapy) attempt to provide just this type of explanation. The point here is not that all forms of psychotherapy are equally as plausible (they are not). Rather, the take-away point is that the ‘go open’ argument is correct in noting that informed consent requires that patients understand how any proposed treatment is supposed to cause the therapeutic effect,7 however, in contrast to the ‘go open’ argument, theory-specific explanations for treatment can, and often do, meet this requirement.

What is the harm of ‘going open’?

This paper has argued that practitioners do not have a duty to disclose the role common factors play in the therapeutic encounter. However, this paper is not arguing that therapists ought to withhold this information from their patients. Practitioners should discuss the mediators of change in therapy with patients to whatever level of specificity that they choose (including no discussion), as long as they do so accurately. The ‘go open’ argument misinforms patients by wrongly claiming that the common factors are causes of the therapeutic effect, rather than being merely mediators. This misinformation may seriously harm patients by providing them with a mistaken view of the state of knowledge about psychological healing that may influence both treatment decisions and outcomes. ‘Going open’, ironically, may put the patient in a worse epistemic position than therapy as traditionally practiced and negatively impacts patients’ ability to give informed consent to treatment.

An advocate of the ‘go open’ argument may object that practitioners have a duty to disclose information about both potential mediators and mechanism of change in therapy. This objection would grant that the common factors do not answer the ‘How does therapy work?’ question, while still arguing that the standards of informed consent require that patients understand all of the potential mediators of change in therapy. This, however, is setting a far higher standard for psychotherapy than the rest of healthcare. As Colloca and Benedetti note, ‘all medical procedures are associated with a complex psychosocial context that might affect the therapeutic outcome’.19 Psychosocial factors, such as the perceived warmth and competence of the practitioner, the practitioner’s outfit, and the patient’s hopefulness are statistically significant mediators of change in medical interventions ranging from allergy treatments to recovery from cardiac surgery.20–22 Informed consent requires that the patient be given a ‘core set’ of information about the nature of the treatment. This does not necessitate that medical practitioners explain every mediating factor that is merely statistically correlated with producing the treatment effect; the ‘core set’ of information required for informed consent to cardiac surgery certainly does not include the potential impact of the prescribing practitioner’s sartorial choices.

Given the pervasiveness of psychosocial factors on healing, it is far too onerous a requirement that medical treatments include an exhaustive discussion of all potential mediators of change. Just as consent to medical interventions typically does not require that patients have substantial understanding of the biochemical details of treatment (which often would require significant background knowledge and training),23 consent to psychotherapeutic interventions does not require that patients have a substantial understanding of every factor that may be statistically correlated with healing (which also would require significant time, training, and background knowledge). What information should be provided to the patient? Here, this paper agrees with the ‘go open’ argument: patients need to understand the ‘real engine of treatment’. This requires that patients understand the hypothesised mechanisms of change for any given therapy.

The ‘go open’ argument is correct that practitioners need to ‘go open’, it is just wrong about what they need to go open about. An informative explanation for the ‘How does therapy work?’ question requires that patients understand that there is currently no consensus about the mechanisms of change in psychotherapy,14 and that the therapy on offer (including cognitive therapy) is based on disputed theoretical foundations.24 Patients also need to understand that multiple therapies (sometimes based on different hypothesised mechanisms of change) appear to be comparable in their efficacy for treating many disorders. A theory-specific therapy would be deceptive if it claimed that its methods and techniques were the only way to achieve the therapeutic effect. However, as argued above, theory-specific therapies (such as cognitive therapy) need not make such a claim.


The ‘go open’ argument is based on a mistaken view about the mechanisms of change in psychotherapy and threatens to harm patients by undermining their ability to make informed treatment decisions. This paper has argued that the prima facie ethical problem raised by the ‘go open’ argument is diffused if we clear up a conceptual confusion about what, exactly, we should be going open about. Therapists should be open with patients about the differing theories of the mechanisms of change in psychotherapy; this can, but need not involve discussing information about the therapeutic common factors.

Ethics statements



  • Contributors GL is the sole author.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests None declared.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

Other content recommended for you