Article Text
Statistics from Altmetric.com
This article on the potential for patient online records access (ORA) to increase the likelihood of nocebo effects is timely,1 given the recent introduction of full prospective records access for primary care patients in England.2 Blease provides a convincing overview of the evidence for the nocebo effect and examines the complex interplay with health inequities. The article proposes two mechanisms for ORA augmenting nocebo effects through: (A) patients reading about possible negative outcomes of treatments and (B) a negative clinician tone or less supportive content in the notes diminishing patient trust and setting up negative expectations. Blease notes a small percentage of those who had used ORA at least once during the previous year in one study reported feeling more worried about their medications,3 and that those with mental health diagnoses were significantly more prone to this than those without such diagnoses.4
Clinicians are familiar with the ethical dilemma presented here—the need to balance ‘respect for honesty and transparency in patient care with the possibility of harm from nocebo effects’ (p1). The article does not, however, explore clinicians’ potential reluctance to downplay medication side effects for fear of damaging their credibility and negatively impacting on their relationships with patients. It also does not discuss the practice of ‘safety-netting’ in the clinical record, which is a third way in which ORA could augment nocebo effects. It is common practice, for example, for clinicians to document ‘red flags’, or include back-up plans should the initial treatment plan fail. For example, many clinicians will ‘safety net’ a diagnosis of musculoskeletal back pain or sciatica with a statement that a patient should seek urgent medical assistance if they experience leg weakness, bladder or bowel dysfunction, or perianal anaesthesia. The likelihood of cauda equina syndrome is very low, but the consequences of delaying treatment are severe. From both a patient safety and a medicolegal perspective it is prudent to document such potential but rare differential diagnoses.
With regard to the second proposed mechanism, Blease provides evidence that negative language in the notes can lead patients to question clinician competence or damage the clinician-patient relationship. The article notes that reminding clinicians about the potential for nocebo effects could encourage them to be more mindful of the language they use when making entries into the patient record. While we agree, we suspect that the desire to avoid ORA causing increased workload from more patient queries or complaints may be a stronger motivator.5 It is encouraging to hear that a web-based course for clinicians in the States is addressing the issue of documentation for an audience that includes patients, and we agree there is a need for healthcare education courses to tackle this issue. At Manchester, we are also engaging with clinicians and patients with a view to developing materials that could help clinicians write in the notes in a more layperson-friendly way while maintaining their clinical usefulness.
Various other solutions to the nocebo problem are presented, including the argument that we may wish to focus on how side effect risks are framed. Blease presents Leibowitz et al’s work,6 suggesting we may wish to consider informing patients that side effects are a sign the treatment is working, but does not examine the evidence for this contention. This reframing approach could be viewed as paternalistic or even manipulative, and could prevent a patient from asking for an alternative that may treat their condition just as well, but with fewer side effects. Full transparency may be a better option; clinicians can inform patients about potential side effects, and suggest that if problematic, alternative options are available. If all alternatives prove equally troublesome then clinicians may wish to discuss if these side effects could be the lesser of two evils. Properly informed, most patients have the capacity to decide if they prefer having side effects, or running the risk of their condition deteriorating and leading to more significant problems (such as hypertension leading to a stroke). Rather than reframing, we might be better focusing our attention on how to ensure patients have an accurate understanding of the risk of side effects, alternative options and the risks of no treatment.
The proposal that records systems could be adapted to automatically add information to aid patients’ understanding of their medications while minimising the potential for nocebo effects is appealing. Blease suggests ‘tooltips’ that offer to supply patients with information about side effects if they wish to read them. However, many clinicians may not wish to risk their patient suffering a serious adverse reaction and subsequently claiming they were not warned about the possibility that this might occur. Glyceryl trinitrate spray, for example, is known to cause a drop in blood pressure and so patients are routinely warned of this and advised to use it while sitting down—not revealing this potential side effect is a risk to patient safety. ‘Authorised concealment’ may sound good in theory, but a clinician may not fare well from a medico-legal perspective if they failed to warn a patient about a potential side effect who then went on to suffer a serious harm as a result (such as falling and sustaining a head injury).
The article compares the dilemma clinicians are faced with to that which confronted Odysseus navigating between Scylla and Charybdis. But full transparency is neither a six-headed monster nor a whirlpool, and patients have the right to make fully informed decisions about the potential side effects of treatments they are being prescribed. Many clinicians will prefer to err on the side of full transparency and consider the potential for nocebo effects a price worth paying, especially since (as Blease notes) ORA gives patients who may otherwise be disadvantaged more time to fully understand and engage with their health information. Blease concludes the article by noting that ‘further experimental research is now required to explore whether open notes influence nocebo effects’ (p7). We heartily agree, and look forward to further discussions regarding the methodology that such future research could employ.
Ethics statements
Patient consent for publication
Footnotes
Twitter @brwmcmillan
Funding Brian McMillan and Gail Davidge are funded by an NIHR Advanced Fellowship (NIHR300887).
Disclaimer The views expressed are those of the authors and not necessarily those of the NIHR or the Department of Health and Social Care.
Competing interests None declared.
Provenance and peer review Commissioned; internally peer reviewed.