Key points are not available for this paper at this time.
Design
Editorial
Answers are grounded in this paper and cite their sources.
This editorial highlights the practical, psychological, and legal barriers anaesthetists face in obtaining truly informed consent in clinical practice.
Obtaining consent for medical procedures is a process involving disclosure to, comprehension by, and permission from the patient. Ensuring valid consent requires both an understanding of the legal requirements and the ability to feasibly implement these within the constraints of clinical practice. The Association of Anaesthetists of Great Britain and Ireland (AAGBI) recently produced updated guidelines for consent in anaesthesia 1. Similar guidelines exist in Australia and elsewhere 2, 3. While commendable in principle, anaesthetists face many barriers to providing consent according to these guidelines. This editorial will explore the basis for these obstacles and the need to achieve better alignment between the principles of consent and the realities of clinical practice. Obtaining informed consent from a patient requires the anaesthetist to disclose the planned interventions and their risks and benefits, as well as any alternatives to the intended techniques. In theory, the risks of clinical procedures are simply the product of the incidence and severity of any adverse event. In practice, however, not only are incidences often difficult to quantify, but clinicians’ risk perceptions are frequently biased beyond this psychometric model 4. Furthermore, the significance of a particular outcome to a patient may be highly subjective and disclosure of all the variables influencing the risks to which the patient is exposed is unfeasible 5. In anaesthesia, the incidence of serious complications is extremely low. Furthermore, the separation between administration of anaesthesia and the manifestations of some complications may make them challenging for clinicians to detect or correlate with their techniques 6. Once patients leave the recovery ward and enter the charge of other clinicians, even serious adverse events such as awareness vary widely in their reported rates 7. It is well established that anaesthetists sometimes have difficulties applying statistics to their practice 8. Most would recognise that a technique performed a hundred times without incident still does not have zero risk, however many might not appreciate the incidence of a complication could still reasonably be as high as 1 in 33. This is based on Hanley's rule of thumb that zero instances in n cases only provides reassurance, within 95% confidence limits, that the incidence is less than 3/n with a 95%CI 9. As such, basing risk perception solely on their own practice may provide clinicians with limited insight into the true likelihood of adverse events. Pooled data are valuable to reveal the incidence of uncommon but potentially serious anaesthetic complications. The National Audit Projects in the UK have been a valuable tool to highlight complications and risk factors that might not be apparent from the practice of an individual clinician or even the morbidity/mortality data of a single anaesthetic department 10. Similarly, large scale, national reporting systems such as the WebAIRS project provide a simple method by which any anaesthetist in Australia and New Zealand can submit complications to an online database 11. These allow low incidence events to be analysed and risk factors identified, although they may be limited by biases of participation and reporting. Models to estimate survival have been shown to be highly predictive for some procedures such as aortic aneurysm surgery 12-14. Nevertheless, even when accurate numerical data are available, it is is not always easy to translate the population risk to the specific context and comorbidities of individual patients. Furthermore, cultural factors, personal experience and a particular fear of the adverse outcome are much stronger contributing factors to perceived risk than numerical data 15. For example, although two clinicians may know the statistical risk of epidural haematoma, one who has experienced this complication may subjectively perceive the risk to be greater and be more reluctant to perform neuraxial blocks. Conversely, longstanding freedom from a rare complication may lead a clinician to be more cavalier about use of a technique. Many studies have confirmed the inherently poor ability of clinicians to assess risks of tests and procedures 8. If not accounted for, these biases in clinicians’ understanding of risk may affect the their clinical judgement and the manner in which risk is disclosed to patients, potentially compromising the validity of consent and exposing them to unnecessary harm. In addition to understanding risk magnitude, anaesthetists must be aware of which risks to disclose 1. Negligence claims in relation to consent arise on the basis that the patient in question would have refused the procedure had the risks been known, a determination that will vary between individuals 16. In the absence of a specific enquiry by the patient, guidance on how to individualise consent is provided by the concept of ‘material risk’ which encourages the clinician to ask ‘what would this particular patient regard as relevant?’ 1, 2. The need to individualise material risk can arise because of patient-specific factors that make them more vulnerable to, or increase the significance of, a specific complication, particularly where alternative techniques exist. Exceptions to the need to disclose material risk are extremely limited, essentially amounting to the patient persistently declining risk disclosure or situations where doing so risks serious harm to the patient (beyond distress or simply the refusal of beneficial treatment) 1. Despite this, it can be argued that the risks actually disclosed to patients are arbitrary and that many complications of anaesthesia which might reasonably represent a material risk are not typically consented for. The various elements that together constitute ‘an anaesthetic’ can be divided into procedures (e.g. intravascular lines, regional blocks) and process elements (e.g. choice of monitoring, medications, devices or techniques) that inevitably arise from the routine sequence of decisions made during provision of anaesthesia. While consent for discretionary procedural elements of anaesthesia is more reliably sought, even this is inconsistent in the authors’ experience. Both neuraxial blocks and arterial lines are rarely associated with catastrophic complications but detailed disclosure of risks appears usual for the former and uncommon for the latter. Conversely, obtaining consent for the process elements of anaesthesia is rare 5. Choices of airway devices, medications, routes of administration, use of monitoring and antibiotic/thromboprophylaxis, combined with the potential for the treating clinicians to be gaining proficiency with new devices and techniques during the process of patient care, may dramatically influence not only the magnitude, but also the nature, of risks to which the patient is exposed. These decisions – which balance potential adverse consequences including nausea, anaphylaxis, haemorrhage, permanent hearing impairment and limb ischaemia against vastly different alternative risks such as awareness, pneumonia, thrombosis, sepsis and haemodynamic compromise – are typically made without any discussion to ascertain their significance to the patient, presumably because they are perceived by anaesthetists as matters of ‘clinical judgement’, made in the patient's best interests to decrease the overall level of risk. Yet any of these vastly different outcomes could represent a material risk that might reasonably result in a patient exercising their right to refuse a particular intervention, or even the entire primary therapeutic procedure for which anaesthesia was required in the rare situation where a particular element was viewed as integral to the anaesthetic. The AAGBI guidelines state that it would be ‘unreasonable’ to get patients to approve every drug and technique comprising an anaesthetic, citing the Montgomery ruling 1, 17 which cautions against overwhelming the patient with overly technical information they cannot comprehend. They advocate, however, that discussion should include ‘components that are of importance to that patient’. If thiopentone rather than propofol were selected as an induction agent it would be hard to assert that the increased risk of fatal anaphylaxis, or potential loss of a limb (if injected into an unrecognised arterially placed cannula), represent any less of a material risk than the complications of neuraxial anaesthesia. Similarly, disclosing these risks and the rationale for assuming them requires no greater ‘technical understanding’ than explaining the relative risks of epidural vs. intravenous techniques for postoperative analgesia. If the principles of informed consent outlined in the guidelines are to be upheld, it is logical that disclosing both procedural and process risks, along with their alternatives, is equally necessary. Moreover, patients should be given the opportunity to selectively decline individual elements giving rise to these risks. Such disclosures are likely to be unrealistic, however, due to the myriad of alternative components making up every anaesthetic, each with its respective risks. This makes the routine obtaining of informed consent for anaesthesia extremely challenging, if not impossible. Issues of language or capacity can impede the ability of a patient to participate in the consent process. Clinicians’ use of either technical or mitigating language may substantially influence patients’ perception of the risks disclosed, even in English speakers 18. In patients with limited English, the legal requirement to use impartial interpreters (rather than family members) to translate may not be appreciated and may influence the validity of the consent obtained 19, 20. Furthermore, clinicians may be unclear on the legal nuances dictating when a patient lacks capacity to consent, who has legal standing to provide consent in these situations, and the limits of their authority 21. For example, in Australia, the requirement to act in a patient's best interests, as well as specific legislative requirements in most jurisdictions, prevent Jehovah's witnesses from refusing permission for a blood transfusion for their children where this is judged to be medically necessary 22. In the UK, the situation is less clear. As either inappropriately withholding life-saving therapy, or its administration against the wishes of the consenting party, may have serious consequences for both patient and clinician, it is advisable to seek specific legal advice on a case-by-case basis in these circumstances. Recognition of the need for such legal guidance, however, presupposes awareness of the potentially uncertain validity of the refusal of consent – which is unlikely to be intuitively appreciated by most anaesthetists. The practicalities of modern anaesthetic care exacerbate these issues. Staggered, day of surgery admissions mean that anaesthetists are frequently faced with patients for the first time in the operating complex, immediately before surgery, limiting the availability of interpreters and patient representatives. Such time constraints also restrict exploration of what represents a material risk to a particular patient and make the detailed documentation of discussions surrounding consent outlined by guidelines unrealistic, particularly during high turnover lists. Yet, despite this, and the fact that numerous other interactions/procedures must also be undertaken during the short interval between a patient's arrival in the theatre complex and the initiation of anaesthesia, courts have tended not to view expediency as a defence in consent cases. Limits on the time and physical space available for discussion also have implications for patient comprehension of risk and their ability to give permission, free of coercion. Adequately explaining techniques and alternatives may take time in itself, in addition to the time required for patients to process, discuss and consider the options presented to them. Even where patients are seen by a different anaesthetist, some days or weeks before the procedure, and written information is provided, the AAGBI guidelines emphasise the need for the treating anaesthetist to confirm the adequacy of consent (rather than simply its documentation) and provide the opportunity for ongoing discussion. While the opportunity to refuse to consent to an intervention must always remain a realistic option for the patient, perceived pressure to proceed is likely to be increased along with the physical and temporal proximity of consent to the intervention. Similarly, it is foreseeable that anaesthetists might see it as callous to raise the possibility of serious complications or death for the first time immediately before induction of anaesthesia, resulting in a tendency to omit specific mention of these risks so as to assuage patient anxiety. Arguably for anaesthetists, the above issues have implications beyond just ensuring valid consent for anaesthetic interventions. As patient advocate, the anaesthetist must first believe that informed consent exists for the primary therapy for which anaesthesia is required and has an ethical and legal obligation to speak out when this is not the case. A signed consent form represents acknowledgement that some consent process took place but does not confirm its validity. The authors are aware of many situations in which issues of language or capacity that invalidated consent for the primary surgical procedure – as well as circumstances in which the patient has signed a consent form despite no attempt to obtain consent for surgery ever having been made – have been identified by the anaesthetist. The anaesthetist is in a position both to observe, and be impacted by, deficits in the conduct of the surgical consent yet surprisingly the responsibility to attend to this is not addressed in the guidelines. An untenable disparity exists between the principles of informed consent and real-world anaesthetic practice. The issues outlined above create a situation where deficiencies in providing what might be legally regarded as valid consent are unrecognised or accepted as inevitable. Since the overwhelming majority of anaesthetics administered do not result in adverse outcomes, let alone negligence suits for consent, these daily inadequacies go unacknowledged. Yet clinicians must appreciate that if the validity of the consent process is ever scrutinsed, it will probably only be in the context of these rare consequences having already arisen. It is likely that most anaesthetists reassure themselves that their practice in relation to consent is consistent with that of their colleagues. If an issue arises though, the standard by which the adequacy of consent will be judged is not whether the ‘reasonable anaesthetist’ would have disclosed a particular risk but what a particular patient would reasonably want to know 1, 16. The current situation therefore does little to either respect patient autonomy or protect anaesthetists from litigation. If the legal requirements for obtaining anaesthetic consent were strictly adhered to, it is likely patients would simply become overwhelmed by the array of alternative material risks presented to them. These requirements are, however, determined by the courts and legislation. Any perceived need to ‘recalibrate’ legal expectations so they better reflect reasonably achievable standards for provision of informed anaesthetic consent, would need to be addressed in these forums. Guidelines for anaesthetic consent simply serve to advise anaesthetists how to apply the existing legal requirements in clinical practice. Where knowledge deficits exist, guidelines provide an opportunity for anaesthetists to re-evaluate their practice. It is likely many anaesthetists underestimate the complexity of the issues involved and require more training on the legal requirements and technique for obtaining consent. The guidelines also highlight (yet do not acknowledge) the concerning unfeasibility of adhering to many of the principles outlined. Overcoming these obstacles would require significant changes to current processes of anaesthetic practice, which would be likely to affect throughput. It is unfortunate the guidelines are not phrased in a way (analogous to the minimum standards for monitoring) that would allow clinicians to more readily hold healthcare institutions accountable for facilitating the requisite changes to workflow, so as to ensure valid consent processes could be practicably implemented. While information leaflets, provided well in advance of anaesthesia, are helpful, they may be too detailed to be processed by patients and cannot address the subjective issues that give rise to material risk in a specific individual. Innundating patients with exhaustive lists of potential complications is unlikely to satisfy the requirements for valid consent 1, 3. Information must be modified according to the needs of individuals but anaesthetists have limited opportunities to determine these needs. One strategy might be to rethink the role of the documentation distributed to patients in the lead up to procedures requiring anaesthesia. Rather than a comprehensive list simply detailing the risks disclosed, questions could be included to help ascertain what issues are of particular significance to each patient, as has been suggested for surgical consent forms 22. The responses to these questions would help make more efficient use of the anaesthetist's interaction with the patient as well as providing a record of the issues raised. This is more in line with the concept stressed in the guidelines: consent as a process, composed of an ongoing dialogue tailored to the needs of the patient. Providing informed consent to the standards outlined in the guidelines represents a significant practical challenge. While it is not met, anaesthetists will continue to assume responsibility for undermining the legal rights of patients to make decisions about their own care, in the interests of keeping the system running. The authors would like to acknowledge Dr Jane Anderson and Dr Roderick McRae for their assistance in the preparation of this editorial. No external funding or competing interests declared.
No takes yet. Share an insight, caveat, or question.
Chrimes et al. (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: