The psychological aspects of placebo

Classical definitions of placebo can introduce conceptual confusion rather than clarity when considering the mechanisms underpinning placebo effects. For example, defining placebos as inert substances leads one paradoxically to define placebo effects as the effects of inert substances. These difficulties may stem from a focus on placebo as a substance rather than placebo as a process.¹⁶ Indeed, current definitions now invoke notions of process. For example, placebo effects have been defined as changes in a person’s health status that result from the meaning and hope that the person attributes to a procedure, event or interaction in a health-care setting.^17,18 Colloca,¹⁹ a leading commentator on the placebo phenomenon, goes further in linking placebo effects to a specific mechanism, defining them as ‘powerful determinant[s] of health outcomes across many different diseases and encounters; the placebo effect is due to the expectancy of positive treatment outcomes’.¹⁹

Two main theories dominated early work on the psychological mechanisms underpinning placebo effects: (1) learning theory, specifically conditioning (i.e. placebo effects are underpinned by associative learning when placebos are paired with an active drug that triggers a physiological response); and (2) response expectancy theory (i.e. placebo effects are underpinned by the patient’s conscious or unconscious expectation that the placebo will have a particular effect). Experiments were designed to test competing hypotheses derived from these theories and the evidence amassed suggested that both conditioning and expectancy were involved in placebo effects in different circumstances.²⁰

Less divisive accounts of placebo mechanisms have now been proposed. Benedetti²¹ emphasises the importance of considering disease- or system-specific placebo mechanisms, particularly when considering mechanisms at the physiological level. Colloca and Miller²² integrate insights from learning theory and response expectancy theory, arguing that patient expectations are the central psychological mechanism that mediates placebo effects. According to this model, the brain decodes the psychosocial context, formulating (conscious or unconscious) expectations about outcome that then trigger placebo responses.

Colloca and Miller²² drew on previous work to suggest that expectations are shaped by learning mechanisms around three types of sign (or triggers) in the psychosocial context:

1. indices, which generate expectations through sensory- or memory-based associations for individuals (e.g. tablets can be indices when patients become conditioned to expect symptomatic benefit from taking them)²³
2. symbols, which generate expectations through culturally specific conventions, including language (e.g. the ritual and doctor–patient communication around surgery and the operating theatre foster particular expectations of benefit)²⁴
3. icons, which generate expectations through perceived similarities with the object (e.g. observing a similar person with similar symptoms responding to an intervention can foster positive expectations through social learning mechanisms).²⁵

Understanding the influence of these different elements on placebo effects within clinical trials can then inform attempts to design, manipulate and control placebo effects in surgery trials.

Five domains of the psychosocial context of health care that may influence patient outcomes have been suggested from reviews of the literature.²⁶ A recent review further explored how each of the five contextual domains are at play in clinical trials and identified specific design features and methods that might shape patient expectations and, therefore, placebo effects in trials²⁷ (Table 2).

TABLE 2

Different domains of the psychosocial context of health care on the surgical placebo response,

Qualitative research methods are helpful for exploring the psychosocial context of clinical trials and understanding the myriad influences and dynamic processes involved in shaping patients’ expectations. For example, interviews in a placebo-controlled trial²⁸ of acupuncture for osteoarthritis revealed that participants derived empathy not only from the acupuncturists themselves, but also from other trial personnel, from the friendly and polite reception staff and from the ease of making convenient appointments, all of which made the patients feel cared for.

How patients are informed about placebo surgery might be a key component of the psychosocial context of clinical trials that could shape patients’ expectations. Patient information leaflets in placebo-controlled trials typically seem to explain placebo effects in quite negative terms, if at all, with most devoting considerable space to describing the potential benefits and mechanisms of action of the trial treatment while describing placebos as ‘a ‘dummy treatment, which looks like the genuine medicine but contains no active ingredient’.²⁹ This is important because, arguably, it does not adequately inform patients about the potential for positive or negative outcomes from the placebo intervention.²⁹ Another acupuncture trial,³⁰ this time for irritable bowel syndrome, identified four main ways that patients conceptualised placebo effects: (1) placebos are necessary for research, (2) placebo effects are fake, (3) placebo acupuncture is not real acupuncture and (4) placebos have real effects mediated by psychological mechanisms. Negative views of placebos as fakes or illusory are potentially problematic for patients who receive a placebo intervention and experience real tangible benefit from it, as they may then struggle to make sense of this and integrate it into a coherent narrative that does not entail them feeling tricked or gullible.^30,31

Patient expectations are central to placebo effects and are driven by multiple components of the psychosocial context, including interpersonal interactions with clinicians and trial personnel, and information about the trial interventions. It is important to consider this when designing and conducting placebo-controlled trials in surgery.

The physiological aspects of placebo

The scientific overlap between the physiological and psychological aspects of placebo is significant and far from straightforward. Most work has been completed in the context of pain and pain relief, rather than methodological considerations for comparative groups in surgical trials. Therefore, direct application has limitations. However, in the context of pain modulation, it is well established that the mechanisms of placebo analgesia involve the antinociception brainstem pathway and spinal inhibition.³² Functional magnetic resonance imaging work has also linked observed placebo effects to the anticipation of reward.³³ If these effects are substantial for an intervention such as surgery, it is critical that they are taken account of in any trial design to enable fair comparison or deeper understanding of mechanism.

Furthermore, there is evidence from investigations of the mesolimbic reward system that certain personalities are more susceptible to such analgesic responses and placebo effects.³⁴ Many of the more qualitative features of placebo (e.g. expectancy, prior experience and belief systems) can influence various opioid and dopamine receptor systems to produce an effect,³⁵ giving physiological credence to centuries-old medical teaching (Galen) regarding the need for ‘confident’ physicians. The therapeutic setting can also produce very similar physiological changes using the same pathways.³⁵ What should not be forgotten is that the powerful antinociceptive or positive effects can also be mirrored with negative expectancy and ‘anxiety amplification’, the understanding of which continues to be driven by rapidly developing functional magnetic resonance imaging research.³⁶ Such nocebo or negative placebo effects also have a place in clinical trial design, depending on the research question being asked. It is likely that with large data set and machine learning input our understanding of these physiological areas will increase.