To the Editor,

We enjoyed discussing the paper by Taneja et al. on discordance between documented and preferred resuscitation preferences1 at our virtual journal club, based in Manchester, UK. This study aimed to assess the degree of discordance between elicited and documented resuscitation preferences among medical inpatients at a tertiary hospital. We congratulate the authors for investigating a potentially controversial issue that is sadly particularly relevant in this time of the COVID-19 pandemic.

The means by which resuscitation preferences were elicited in Taneja’s study comprised research nurses conducting open conversations with patients during their hospital admission, after their resuscitation status had been documented by the medical team.1 This process identified discordance between documented status and patient preference in 90 out of 349 participants (25.8%). Nevertheless, this degree of discordance may be explained to some extent by the conversation as outlined being not only analytical but also intrinsically interventional. For example, the mere fact that a patient is approached for a second conversation about resuscitation preference might significantly influence their responses.2

The quotation referenced in the title of this letter is attributed to physicist Werner Heisenberg, author of the uncertainty principle, which describes how atoms can only be observed in one state, but actually exist across multiple states simultaneously.3 Building on Heisenberg’s work, Hanz-Dieter Zeh proposed the theory of quantum decoherence: if a quantum system remains perfectly isolated, it will maintain coherence indefinitely, but cannot be manipulated or investigated. The moment we disrupt isolation, coherence is shared with the environment and therefore lost.4 A similar phenomenon is potentially at play on a social level in Taneja’s study,1 and although “reflexivity” is more commonly associated with qualitative research, a richer description of the intervention and the positionality of those delivering it would have allowed the reader to better assess the extent to which this may have been the case.5

Taneja et al. state that where discordance was identified, it was “reconciled” in 77% of cases.1 While this certainly supports the utility of their intervention, we question whether such a definitive term is appropriate, as further discussion may identify (or provoke) further discordance, ad infinitum. Resuscitation preference should perhaps not be perceived as stable, particularly during acute hospital stays. Future research could usefully focus on characterising preference instability over time and finding ways of empowering patients to communicate with healthcare professionals about resuscitation preferences, while minimising the risk of introducing bias to shared decision-making.