Based on the six themes identified above, we present four observations with important implications for governments and health policy makers with regard to planning for influenza pandemics and public health emergencies more broadly.
First, our analysis identified a notable level of cynicism among participants about the health system's current capacity to respond effectively to an influenza pandemic. The health system was described as already in crisis and barely able to meet existing demand, let alone the overwhelming demand of an influenza pandemic. Participants also articulated working assumptions about the behavioural responses of healthcare workers in a crisis. On the one hand, there were concerns that healthcare workers would 'protect their own' or avoid coming to work if other colleagues contracted the influenza virus. On the other hand, participants expressed considerable empathy toward the burden of moral responsibility healthcare workers are faced with making 'tough choices' about patients' lives. Finally, while participants recognized that standard of care might not be achievable in a pandemic influenza crisis, they emphasized that care was owed to all patients even if it were simply comfort measures. Understanding the public's 'starting points' may have important implications for policy makers in terms of governments being able to effectively engage, inform, and communicate with the public in a timely manner, especially during public health emergencies like an influenza pandemic. For example, communicating clearly the reasons why healthcare workers might be prioritized over non-healthcare workers for scarce resources during a pandemic (e.g. because they put their lives in danger by the nature of their work) would need to take into account that the public may be somewhat distrustful of healthcare workers (e.g. the finding that some participants believed that the healthcare system is already biased in favor of healthcare workers). Building trust may signal to the public that scarce resources are being valued for what they are, scare resources, and that these resources are not being 'wasted' frivolously.
Second, participants underscored the challenges of empirical uncertainty in ethically charged decision-making moments. While participants often sought more facts about the priority setting scenario as presented, when more facts were not available, participants explored alternate formulations of the priority setting scenario based on different empirical assumptions to test their positions, values, or conclusions. For example, the scenario did not specify how Ms. A (the nurse) had contracted the influenza virus. In all three town halls, participants considered explicitly whether it made a difference in their assessment of the relative claim of Ms A and Mr M for an ICU bed if Ms A had contracted the virus at work rather than in the community. Notably, this would often include deliberation on the social value of 'essential workers' in an influenza pandemic and the appropriateness of considering social value in allocating limited health resources. Participants recognized that in an influenza pandemic, policymakers and clinicians would be faced with making decisions with incomplete information. Given this uncertainty, participants placed great emphasis on the need for planning and agreement about decision-making principles and values to bridge these empirical gaps and ease the moral and psychological burden of decision-makers in the midst of a crisis. Our findings suggest a considerable appreciation among town hall participants of the inherent uncertainty of pandemic response. Clear communication by decision makers about the nature and extent of this uncertainty may make a greater contribution to building public trust than issuing unfounded assurances of certainty intended to allay public fears.
Third, when given time to deliberate about ethical quandaries, like priority setting in the context of influenza pandemics, participants demonstrated nuanced ethical reasoning. Participants struggled with regard to which values should guide priority setting decisions. Overall, the participants rejected the pragmatic constraints imposed by the assumption of extreme scarcity. While participants acknowledged that scarcity was a reality, the deliberation would often start with the assumption that 'something could be done' and focus on alternate practical solutions to bridge the scarcity gap (e.g. whether healthcare staff outside the ICU could be taught the basic necessities of the job; whether resources could be reallocated from different departments of the hospital to ease the ICU's burden; whether the lay community might be able to fill the gap). When 'nothing could be done' to bridge the scarcity gap, participants articulated the struggle in moral terms about what values and criteria should guide priority setting decisions. Participants acknowledged and understood the inherent ethical difficulties in having to make allocation decisions under duress, uncertainty, and time-constraints. Participants reinforced the importance of planning not only as a means for easing the moral and psychological burden of decision-makers, but also as a means of building a broad consensus and ensuring transparency about how priority setting decisions would be made before the constraints of a crisis created an impediment to deliberation and engagement.
Finally, the participants proposed a number of ethical values and criteria that they perceived to be relevant to allocating ICU beds during an influenza pandemic; this is consistent with other empirical studies examining public perspectives on priority setting in a pandemic. For example, Ritvo and colleagues conducted a national telephone survey of Canadian residents, in which participants were asked to identify who should have priority for access to scarce hospital resources . Participants attributed high priority to children, healthcare workers infected while serving patients, the sickest patients, and adults with dependents, which suggests a number of prioritization principles were at work (e.g., fair innings, reciprocity, compassion). Much like Ritvo et al's study and our Canadian town halls, Vawter and colleagues, who conducted town hall meetings with Minnesota residents, found that participants also elected to ration ventilators on the basis of varying values (e.g. solidarity and mutual responsibility) . Moreover, the Minnesota town hall participants contended that life expectancy and socio-economic status should not be considerations in allocating ventilators during an influenza pandemic. One should note that our empirical findings (along with those of Ritvo et al. and Vawter et al.) seem to be at odds with the majority of theoretical papers that have espoused utility, i.e., maximizing good outcomes with available resources, as the overarching ethical consideration in allocating ICU beds during influenza pandemics,[7–12] including those which argue against giving priority to health care workers . The limits of utility-based theories of emergency ICU bed allocation have also been critiqued in the theoretical literature, especially on the basis of equity with regard to criteria that may disadvantage existing vulnerable populations [6, 13].
A possible limitation of our findings is their generalizability to other health systems or other public health outbreaks and challenges. As a qualitative study, our goal was not generalizability. However, we expect that readers in other health systems or faced with other public health challenges may see themselves in our findings. For example, the balancing of different ethical criteria and the need for public input have been found in other studies on public engagement in priority setting in healthcare beyond the context of pandemic planning [30–32]. Moreover, access to ICU resources is a perennial challenge during seasonal influenza outbreaks, for which the importance of pre-planning, public accountability, and transparency are relevant factors in establishing and maintaining public trust. Another question that one may arise is the applicability and usefulness of qualitative data in the context of healthcare policy discussions. The CanPREP town halls were part of a larger research project looking at pandemic ethics, which included the previously mentioned telephone survey by Ritvo and colleagues . While the national telephone survey provided a breadth of information about Canadians' perspectives on resource allocation and other ethical issues in a pandemic, the town hall deliberations provided insight into the underlying justification or rationale for the survey findings. For example, although the survey indicated a preference for healthcare workers having priority access to for scarce medical resources, it is through the town hall deliberations that we were able to identify intrinsic (e.g. 'society owes them') and instrumental (e.g. 'society needs happy healthcare workers') reasons for this preference whilst also elucidating the ethical and practical complexities entailed in acting on this preference. Finally, our findings may not be broadly representative of Canadians' perspectives on priority setting of intensive care resources in an influenza pandemic. This is due in due in part to the limitations of our sampling strategy, which was not designed to achieve representativeness; it may also be due to the pragmatic demands of participating in a full-day town hall, which may not have been feasible for some interested individuals due to employment, child care, or other constraints. Notwithstanding these limitations, the town hall participants were remarkably diverse in their experience and provided a rich narrative to enhance our understanding of the survey findings and shed light on how some Canadians might address the ethical issues raised by an influenza pandemic.