Unheard and unseen: The hidden impact of nocebo communication in the Intensive Care Unit

The nocebo effect is the psychobiological phenomenon of adverse treatment effects that cannot be ascribed to a specific treatment mechanism, shaped by patient and clinician expectation. ¹ This is in sharp contrast to placebo effect, whereby an non-active intervention has a positive outcome, thus leading to the description of nocebo as the placebo effect’s ‘evil twin’. ² There is a significant and increasing body of evidence demonstrating that the way in which clinicians communicate can be a substantial source of nocebo effects, resulting in adverse symptoms that are far more pervasive than previously thought.^{3 –5}

The evidence

The nocebo effect comes with significant costs to the health system. It has been implicated in the increased reporting of undesirable drug effects, significant dropout rates in clinical trials, high incidence of reported adverse events in the placebo arm of covid19 vaccine studies, as well as perceived lack of drug effectiveness, sometimes on a mass scale.^{6 –10} In studies examining peri-procedural nocebo, the nocebo effect is associated with higher pain scores, increased vocalisations of pain and patients more likely to pull away due to pain.^4,11,12 Patient impact is also exhibited in the form of decreased quality of life, increased psychological and physical distress, non-adherence to treatment due to the perceived adverse effect and/or loss of drug efficacy, as well as financial costs to health services and societal costs.^6,13 It is therefore incumbent on clinicians to recognise and minimise nocebo from their clinical practice.

The neurobiological mechanisms of the nocebo effect have been explored using PET scanning and fMRI. During events whereby pain is enhanced by negative expectation, there is higher intensity activation of the anterior cingulate gyrus, parietal operculum, insula, thalamus and prefrontal cortex compared to activation of those areas without applied nocebo or decreased expectations of pain. ¹⁴ The obliteration of the nocebo effect with the injection of the cholecystokinin antagonist proglumide suggests that the cholecystokinin pathway is an important mechanism in the nocebo response. ¹⁵

Despite the evidence and the demonstrated neurobiological basis, there is little awareness regarding the nocebo effect in current clinical practice in general, and on the Intensive Care Unit (ICU) in particular, making it a phenomenon not often recognised and therefore unseen and unheard. With the exception of the Australian Medical Colleges of Anaesthesia, Australian College of General Practice and the Australian Medical Association,^{16 –18} the authors could not find any reference to the nocebo effect on the website of any medical college or regulatory body including those in the US or UK.

Clinical impact in the ICU

The ICU represents a fertile ground in which the nocebo effect is likely to be exhibited inadvertently through health care staff-patient interactions. We ourselves have conducted a study during the insertion of peripherally inserted central catheters (PICC) in the intensive care setting which suggests a high incidence of nocebo. We have identified three commonplace situations in intensive care practice where the nocebo effect is likely to be active: the consent process, during invasive procedures, and family meetings.

The nocebo effect during consent

The nocebo effect is borne from the complex interplay between three components: the expectations and beliefs harboured by the patient, the physician and the relationship that is engendered between the two, all of which come into play during the consent process. However, the very process of describing potential adverse events may lead to intensivists inducing nocebo responses, leading to increased fear, anxiety and subsequently harm, which are in direct conflict with the principles of beneficence and non-maleficence. ⁶

An alternative is to reframe the risks of the procedure in a way that reassures the patient and their family, rather than causing anxiety. For example, rather than say that ‘oesophageal perforation is a risk of a tracheostomy insertion’, one could say ‘as damage to nearby tissues such as the oesophagus can occur, we always perform this procedure with the use a special camera in the breathing pipe (a bronchoscope) to make the procedure as safe as possible’. A recent study explored the concept of a ‘layered consent’ process whereby patients chose the amount of information they consumed, with a focus on the different information needs of patients and a preference for understanding benefits over risk. ¹⁹ Such an approach may not only improve patient agency but potentially reduce the risk of nocebo effects.

The nocebo during invasive procedures

Many invasive procedures in intensive care require an injection of local anaesthetic. Accompanying the injection of local anaesthetic with the warning that the patient is about to receive a painful injection may be thought to be a kindness by being open and transparent. However, there is clear evidence that phrases with negative suggestions that include words such as ‘pain’ and ‘sting’ can result in hyperalgesia, distress and anxiety, as well as increased vocalisations and patients pulling away in pain.^3,4,12 The nocebo effect does not require a contemporaneous, potentially painful intervention to be active: the act of asking postoperative patients to rate their pain using a pain score can increase pain and requests for analgesia postoperatively four-fold. ²⁰

Alternatives to negatively loaded words such as ‘burn’ or ‘sting’ during procedures can include a simple statement such as ‘I am about to inject local anaesthetic that numbs the skin’, or ‘insert a dilator that makes it easier to pass the central line’. These statements represent neutral information with a clearly outlined therapeutic meaning, that are less likely to bias patient expectations towards increased pain and anxiety. It also reframes a procedure, such as the insertion of a central venous catheter as a positive part of the patient’s care: ‘This will allow us to give you strong medication that will help keep you safe and get you better’.

The nocebo effect in family meetings

The ICU represents a complex area of medicine that has a significant impact on the mental health and wellbeing of patients and their families. Patient relatives are likely to be exposed to nocebo communications during end-of-life discussions with the use of phrases such as ‘withdrawal of care’ or ‘limitation of treatment’. These nocebo communications may harm both patients and families by generating unnecessary anxiety, fear and a sense of abandonment by medical teams. ²¹ An alternative phrase to ‘withdrawal of care’ may be ‘changing the focus of our care to comfort and dignity’. Instead of ‘limitation of treatment’, the phrase ‘appropriate treatment’ or ‘goals of care’ may be used. A similar approach to the nuance of language has been recognised in maternal care and anaesthesia,^22,23 thereby replacing nocebo effects with placebo effects.

The counterargument

Clinicians are obligated to explain the potential adverse events when prescribing medications or recommending procedures. It is well documented that a meticulous description of all potential adverse events during the process of informed consent leads to an increased incidence of adverse effects being reported, ⁶ which creates a dilemma for the physician. By giving the patient reasonable information as part of the informed consent process, allowing them to make personal choices which impact them, there is an unavoidable development of nocebo leading to maleficence, however, elimination of risk information is not an option. This creates an ethical tension for intensivists who seek full disclosure of potential adverse events versus the risk of harm via unintentional nocebo effects. However, as the examples above illustrate, it is possible to combine negative signals of risk with positive ones, such as the steps taken to reduce risk and reframing the context of treatment. The ‘layered consent’ process may also lend itself well to the clinical environment, thus giving patients greater agency. Interestingly, the ‘layered consent’ process is being used in a current multicentre trial, suggesting an alternative to the standard consent process for a clinical trial ¹⁹ . These strategies have the multiple benefits of reducing the nocebo effect while also reassuring and empowering the patient as well as keeping within guidelines of ethical practice.

It could be argued that employing strategies to minimise the nocebo cannot protect patients and their loved ones from the trauma, physical or psychological, associated with critical illness and an ICU stay. However, an increased awareness of the clinical effects of nocebo language may allow communication during difficult conversations to be reframed to articulate compassion, engender trust, and improve the experiences of patients and their relatives, thus demonstrating the need for nocebo minimisation strategies. Though there are many factors outside the intensivist’s control when a patient is admitted, working by the principle of primum non nocere ‘first do no harm’, is still an essential principle to abide by.

Furthermore, unlike many ICU interventions, nocebo elimination or minimisation is cost-free, non-invasive. and has benefit for both the patient and their loved ones. It can be implemented by all clinicians, regardless of their discipline or seniority, provided appropriate training has taken place. A nocebo minimisation strategy works synergistically with the tools that many intensivists pride themselves on: that is, their communication skills.

Conclusion

The ICU likely represents a setting with many unrecognised nocebo situations directly impacting patient care. It is likely that many intensivists have been communicating in ways that are counter to the evidence with an inadvertent risk of increased harm to patients and families. Avoiding nocebo represents a simple, cost-effective intervention to deliver patient-centred care in any health care setting. Though further studies in intensive care need to be performed, the nocebo effect can be ignored no longer. ²¹