The PHERCC Matrix. An Ethical Framework for Planning, Governing, and Evaluating Risk and Crisis Communication in the Context of Public Health Emergencies

Abstract Risk and crisis communication (RCC) is a current ethical issue subject to controversy, mainly due to the tension between individual liberty (a core component of fairness) and effectiveness. In this paper we propose a consistent definition of the RCC process in public health emergencies (PHERCC), which comprises six key elements: evidence, initiator, channel, publics, message, and feedback. Based on these elements and on a detailed analysis of their role in PHERCC, we present an ethical framework to help design, govern and evaluate PHERCC strategies. The framework aims to facilitate RCC, incorporating effectiveness, autonomy, and fairness. It comprises five operational ethical principles: openness, transparency, inclusivity, understandability, and privacy. The resulting matrix helps understanding the interplay between the PHERCC process and the principles of the framework. The paper includes suggestions and recommendations for the implementation of the PHERCC matrix.


INTRODUCTION
What Is Risk and Crisis Communication?
The COVID-19 pandemic has made clear the need to rethink risk and crisis communication (RCC), and to reflect upon the ethical framework in which this type of action should be framed.In this paper we propose a conceptualization of the RCC process in public health emergencies (which we refer to as "PHERCC"-Public Health Emergency Risk and Crisis Communication), we provide an analytical definition, and we propose an ethical framework to help design, govern and evaluate PHERCC strategies.
PHERCC is strictly connected to-and is a significant component of-preparedness and public health response activities.According to the World Health Organization (WHO), public health response aims at "developing policies and plans, and executing activities that reduce the public health impact of emergencies and disasters" by means of "prevention, mitigation, preparedness, early response and rehabilitation," and implementation of the International Health Regulations (WHO 2016b(WHO , 2016a)).The general aims of PHERCC are therefore derived from and aligned with those of public health response.PHERCC therefore aims at eliciting protective behaviors that allow risk reduction; at guaranteeing and fostering freedom of information (UN General Assembly 1946, 1948, pt. 19); at enabling informed decisions and maintaining public trust in institutions (Loss et al. 2021).
The importance of PHERCC has been widely acknowledged and prescribed in international law, including the International Health Regulations of the WHO, the Resolution 46/182 of the United Nations General Assembly, and the Sendai Framework for disaster risk reduction (2015)(2016)(2017)(2018)(2019)(2020)(2021)(2022)(2023)(2024)(2025)(2026)(2027)(2028)(2029)(2030).PHERCC includes the ability to detect, notify and report on public health threats, and disseminate information and recommendations for the population (WHO 2016a, 40-41;UN General Assembly 1991).The Sendai framework provides more details on the guiding principles and the aims of PHERCC, specifically in Priority 4, i.e., "disaster preparedness:" to increase communities' resilience to disasters, it is necessary to develop and strengthen people-centered multi-hazard communication mechanisms and social technologies.The aforementioned systems should be developed through a participatory process and tailored to the needs of users, including social and cultural requirements (UN Office for Disaster Risk Reduction 2015, 21).
The combination of recent epidemics/pandemics and new technologies that allow rapid and "horizontal" spread of information (e.g., social media) provided feedback on the strengths and limits of previous and current PHERCC strategies and approaches.In fact, "public reaction could be considered another outbreak to be controlled during an epidemic" (Hsu et al. 2017).Horizontal means of communication can be both an asset and a barrier-an asset, in that they allow rapid communication which reaches various types of public; a barrier, in that they can generate echo chambers, which in turn can foster the spread of rumors and fake news (Malecki, Keating, and Safdar 2021).Moreover, the "lasting emergency" of COVID-19 has been showing that there is a critical need for theoretical and practical tools to deal with uncertainty and changing evidence, advice, and content of PHERCC: information providers need to be able to dynamically adapt their messages over time, to different audiences with specific needs and characteristics (Malecki, Keating, and Safdar 2021).

Background
Although some work on the ethics of PHERCC exists (Sellnow and Seeger 2013, chap. 9), a framework that identifies and systematizes the issues, the stakeholders, and the approaches, providing both theoretical reflection and practical guidance for planning, governing, and evaluating PHERCC strategies is still missing.As the world attempts to transition, haltingly, to a postpandemic phase, this appears to be the right time to develop a detailed and comprehensive framework for PHERCC, and to incorporate it in the design and development of future response strategies to public health crises: on the one hand, because of the freshness of the available evidence and the depth of the theoretical debate on the subject; on the other, because of the moral and practical need to be prepared for the next public health emergency.
There is a significant body of literature on businessoriented risk and crisis communication (BORCC) (Quinn 2018), including some work on the ethics of risk and crisis communication.For example, Kim theorizes a model based on transparency (i.e.no information should be kept secret), two-way communication (i.e.listen to how the message is received), and right time (i.e.such message should be timely) (Kim 2015).Contreras-Pacheco proposes an approach based on care ethics, encompassing five principles geared toward the mitigation of negative outcomes-and reduction of reputational damage-when a business is responsible for a critical event: taking responsibility; apologizing for the pain; acknowledging the victims; honoring the victims' memory; deploying mechanisms to support the victims' families (Contreras-Pacheco 2018).
However, despite some similarities, the differences between BORCC and PHERCC are significant and make the transfer of ethical frameworks from one context to another appears to be problematic.The first difference is the goal: while PHERCC is a component of public health response, BORCC is mainly driven by the goals of businesses and thus geared toward reputation repair; in this context, communication is therefore mainly targeted at reducing the reputation damage caused by a crisis (Benoit 1997;Coombs 2007;Xu and Li 2013).Maintaining reputation and repairing trust are part of the communication strategies deployed by local institutions, health ministries, national governments, and by the World Health Organization in the context of the COVID-19 pandemic.However, the main goal of such institutions is to provide the public with accurate information, to elicit protective behaviors, and to keep people safe.This includes, for example, information on the virus' transmission mechanisms; on preventive measures, including the use of hand sanitizers or face masks; and on vaccines as an efficient way to reduce mortality, morbidity and transmission.
The second relevant difference between PHERCC and BORCC is due to the adopted definition of crisis: in BORCC a crisis is a low-probability event, which occurs unforeseen, and has the potential to generate a vastly negative impact on an organization and on its stakeholders (Burton, John, and Pearson 2016).The COVID-19 pandemic was not a low-probability event (Johnson et al. 2020), there have been similarities with other epidemics (Bowen and Heath 2007), its possibility was forecasted (Kreuder Johnson et al. 2015), and its impact did not hit specifically organizations or stakeholders, but the entire world.
The CERC (Crisis and Emergency Risk Communication) model represents the systematization and the current gold standard for preparing and organizing the content of risk and crisis communication in the context of public health emergencies (Reynolds and Seeger 2005).It defines five common stages of crises: (a) pre-crisis; (b) initial event; (c) maintenance; (d) resolution; and (e) evaluation; further, it details what the focus of each phase should be, suggesting specific strategies.These include educating the public and developing consensual plans between the information provider (initiator) and the public, as a preparedness strategy; establishing an empathetic communication-although structured and formalproviding information and reducing uncertainty; assessing public understanding and dispelling fake news; informing about post-disaster clean-up and remediation; evaluating, assessing, and planning for future actions.While the CERC model offers useful guidance for navigating an emergency, it solely focuses on communication strategies, without embedding and analyzing the efficacy of this type of communication in an ethical framework-which could ensure justice (following Rawls, intended as fairness and as respect for individual liberty) and increase the effectiveness of messages designed through communication strategies as defined by the CERC model.

PHERCC: Balancing a Triad
PHERCC is ripe with ethical implications and potential pitfalls.Active and passive freedom of information, vulnerability of individuals and communities, trust, and transparency are just some examples already identified and discussed in the literature (Kim 2015;Attademo 2022).As PHERCC's aims are derived from those of public health responses, the most relevant high level ethical implication is the need to balance three elements of a triad: utilitarian notions of effectiveness, intended as the ability to produce benefit in real life conditions (Cartwright 2009); autonomy, intended following Beauchamp and Childress' "nonideal conditions" as the combination of intentionality, understanding and non-control (Beauchamp and Childress 2013, 104-105); and fairness, understood in light of Rawls' liberty principle as a societally compatible implementation of autonomy which helps regulating the potential clash of individual autonomies, i.e.: a comprehensive set of basic rights and liberties that can coexist with similar rights for all (Rawls 1985).In other terms, PHERCC should be effective in generating a desirable outcome (people adopt protective behaviors) while at the same time protecting and even empowering (personal) autonomy, but without compromising (societal) fairness.

CONCEPTUALIZING THE PHERCC PROCESS
PHERCC is a multi-actor and multifaceted process whose core revolves around delivering information to different publics.Looking at recent examples emerging during the COVID pandemic, this included crafting and distributing messages concerning hygiene measures, supporting lockdowns, endorsing vaccination uptake, etc.Since the purpose of PHERCC is to deliver information and elicit behaviors, it needs to create a space for an "asynchronous conversation" between the initiator and the recipients of the action.This entails that (a) the initiator needs to understand the identity and characteristics of various publics, and these publics needs to understand the identity and characteristics of the initiator; (b) the initiator gathers all the evidence that can contribute to building an effective and precise message, tailored to each public based on its specific needs; (c) the initiator ensures the existence and maintenance of a solid infrastructure through which the conversation with each public can take place; (d) the initiator defines a set of messages, (e) finally, the initiator needs to consider and accept each public's voice and incorporate it as fundamental feedback for the next iterations of communication (Malecki, Keating, and Safdar 2021).In fact, as disasters tend to follow a "drop loop model," proceeding from baseline to recovery and development through deterioration caused by trigger events, acute crisis, and stabilization (Clarinval and Ahmad 2015), feedback and iterations are core components of the process.PHERCC processes can therefore be described as a looped ladder consisting of six elements: evidence, initiator, channel, publics, message, and feedback (Figure 1).

Evidence
PHERCC needs to be primarily and structurally grounded on scientific evidence (e.g.: how COVID-19 spreads).Emerging conspiracy theories or misinformation can provide useful information about which type of information is needed the most, and which information voids are currently existing.Such information also constitutes feedback-based evidence to inform refinements of the communication plan.Conspiracy theories, misinformation and disinformation should be debunked whenever evidence is available to do so, and this should play a relevant part in the PHERCC process, as disinformation is harmful and endangers proper public health responses (Love, Blumenberg, and Horowitz 2020).The initiator should ensure public health policy is aligned with evidence, provided in the form of academic research, or firsthand field experiences when research is not available, or even opinions from a group of experts when no research or field experiences are available for an THE AMERICAN JOURNAL OF BIOETHICS assessment.It is important to note that, especially when communicating during early stages of a crisis, when not much evidence on the underlying phenomenon causing a crisis is available, the initiator should ensure that the lack of information available does not prevent public health advice be given to the public.In the scarcity of evidence, safe and inexpensive, potentially effective, measures should be taken.For instance, early pandemic advises from WHO suggested international flights should not be halted as there was no evidence that SARS-CoV-2 could start circulating worldwide (WHO 2020).Later evidence suggesting the airborne nature of COVID-19 demonstrated this advice to be incompatible with a proper pandemic response (Lewis 2022;Zhang et al. 2020).Therefore, the growing and changing evidence needs to be reflected in PHERCC immediately and continuously (Ratzan, Sommariva, and Rauh 2020); at the same time, public health advice and policy, following the aforementioned principles, should ideally precede evidence-based knowledge generation when the response needs to be prompt.As mentioned above, feedback is an integral component to build evidence necessary to develop a message, and therefore evidence should also include how the message will possibly be received by each public, and ultimately how the message is perceived by each public.This includes understanding and analyzing communication strategies, the instruments adopted to convey a specific message (e.g., social media vs. traditional media, which social media channel, textual vs. graphical communication, etc.), the timeframe in which the information is provided, and the cultural, geographical, and socio-economic context of the specific public.

Initiator
PHERCC actions are usually initiated by local, regional, national, or international authorities, and often in combination, providing a mix of information of local and global relevance.PHERCC initiation requires a reputable and recognized authority, as free as possible from conflict of interest (e.g.: promoting specific protective measures and at the same time holding shares in the companies offering these products or services).The initiator needs to be acknowledged by the different publics as a leading institution, else the effectiveness of its communication, even when backed up by evidence, would be negatively impacted.This is one reason why the initiator should also ensure its publics are responsive and receptive and do not lack trust in the institution providing the information.The establishment of authority occurs when there is no ongoing public health crisis, whereas the reinforcement of institutional trust takes place in a time of crisis.

Channel
Channels are the operative system, or platforms, through which PHERCC actions are delivered-these include official websites, press releases, TV, and social media.The adequacy of the channel is crucial: PHERCC actors tend to be present on established channels as they need to have a broad set of followers (i.e.receivers of the message), in order to guarantee effective outreach and circulation of messages; however, some communication channels are structurally inadequate for specific PHERCC actions due to how contents are selected and displayed in the users' feeds (Hindman, Lubin, and Davis 2021), or due to a mismatch between the intentionality of the action and each public's expectation.Indeed, each channel has tailored communication mechanisms and a different public.Specific rules apply to each channel, thus content selection, censorship, and polarization (i.e., how information is displayed to users based on their own interests) are issues to be considered.Regulations on content selection and transparency of social media are currently being discussed, for instance, in the EU (Satariano 2022), but the road to effective enforcement will be long and complex.

Message
The message is the actual content of the PHERCC action, which is based on evidence wherever possible; it is prepared and transmitted by initiators through adequate channels to each public.In line with what discussed above, the same message should be adapted in different forms, based on the recipient and the communication channel, maintaining the same meaning and aim, but tailoring it to the needs of each public (SteelFisher et al. 2012;Spitale et al. 2021).The message is based on evidence produced by scientific literature and the public's feedback.In fact, this evidence should not be limited to the content of the message itself, but also to how the message is conveyed.The initiator should ensure the communication strategy is in line with scientific evidence in terms of effectiveness and design, but it should also consider experimenting with different forms to evaluate the efficacy of specific designs and strategies.Indeed, public health emergencies differ from each other, and the publics change their attitude, understanding and predisposition to listen over time, based on unfolding events, as well as the changing social, cultural, and economic context.

Publics
Each public comprises people or institutions for whom a specific PHERCC action is intended, i.e., the receiving end of the messages.Different people understand or misunderstand the same messages in different ways: there is growing evidence that "social and cultural factors, immediacy, uncertainty, familiarity, personal control, scientific uncertainty, and trust in institutions and media all shape perception and response to risk messaging" (Malecki, Keating, and Safdar 2021).Therefore, a better understanding of the public (intended as a plural, multifaceted, and diverse group) and its specific needs is paramount (Hu 2022).For this reason, throughout this manuscript we preferred using the plural "publics" instead of the singular "public."Publics can play two relevant roles in PHERCC actions: they provide feedback on the content of the messages, which can be used to fine-tune the communication strategy, and help to define and decide which emergency response strategies could be successfully implemented (e.g.: mask mandates, vaccine allocation prioritization schema).

Feedback
Feedback is a set of information on how different segments of the public receive and understand the message provided by the initiator.Feedback data are crucial for the following iterations of the PHERCC process, especially when facing long lasting emergencies, during which evidence might change as well as the public perception of the underlying issue.A recent and poignant example of this is the so called pandemic fatigue-"an expected and natural response to a prolonged public health crisis-not least because the severity and scale of the COVID-19 pandemic have called for the implementation of invasive measures with unprecedented impacts on the daily lives of everyone" (WHO Regional Office for Europe 2020).Thus, listening to publics and incorporating their feedback in future PHERCC actions plays a pivotal role.Its importance has been previously considered: the Sendai framework explicitly recommends developing people-centered multi-hazard, multisectoral emergency communication mechanisms through participatory processes, tailoring them "to the needs of users, including social and cultural requirements, in particular gender" (UN Office for Disaster Risk Reduction 2015).
Two approaches have emerged: active social listening and passive social listening.Passive approaches entail gathering and studying data, which are shared by people for their own purposes, on social media or messaging platforms.It is an observational approach, based on "collecting information from digital communities without engaging with them" (Spitale, Biller-Andorno, and Germani 2022).Active social listening, on the other hand, entails engaging with publics, asking them explicitly about their opinions and personal views, e.g., on the implementation of safety measures as well as their acceptability, considering their impact on people's lives, businesses, and personal versus societal risk evaluation.Passive social listening systems, although very effective in the short run, can be problematic: their systematic use can undermine the proper functioning of the public health system due to the erosion of public trust in public health institutions (Sekalala et al. 2020).Moreover, passive social listening systems can be used-and have been used-to legitimize discriminatory public health policies against minority groups (Sekalala et al. 2020).Active approaches, on the other hand, although slower and depending on effective bidirectional interfaces between publics and authorities, can help building trust rather than undermining it further (Spitale, Biller-Andorno, and Germani 2022).A limitation of active social listening approaches, however, is that the collected sample, and thus the opinions, of participants engaged in active discussion is limited and could potentially bias conclusions and cause misunderstanding, creating improper evidence, which could lead to a new iteration of ineffective, or even dangerous, PHERCC actions.Finally, active approaches are not immune from the risk of generating inequity or discrimination: although less problematic than passive approaches (as they rely only data explicitly shared by the publics), inequitable access to feedback systems could cause an underrepresentation of some opinions.

Foundational Literature
In the recent past, some authors tried to identify ethical principles that should be incorporated in PHERCC actions.Beyond identifying principles, the aim of our work is to develop a comprehensive and viable solution for an ethics framework to be used in the context of PHERCC actions.We briefly present here these valuable contributions as the foundational literature on which our work is built upon.While studying the Chinese response to the 2002-2003 SARS outbreak, Bowden and Heath identified 5 key areas for RCC ethics: (a) a moral obligation to society-"the level of moral responsibility for an issue is determined by the amount of control the person or organization exercises over the decision;" (b) a need for enabling relationships-"providing candid and accurate information;" (c) do no harm and consider harm potential-ground decisions on the moral imperative of respect, not on self-interest; (d) maintain legitimacy through ethical actions-being dishonest and not forthcoming information undermines trust and legitimacy; (e) the obligation of veracity-"concealment might indicate a moral problem" (Bowen and Heath 2007).When analyzing the impact on human rights of public health surveillance measures in the context of the COVID-19 pandemic, Sekalala et al. concluded that "they should be evidence based, contribute to a comprehensive public health surveillance system, include sunset clauses, be nondiscriminatory, and ensure mechanisms for greater transparency and accountability" (Sekalala et al. 2020).Finally, although mostly geared toward developing practical recommendations for communication strategies, the work of Malecki et al. provides some insight that is relevant from an ethical point of view.This entails "data driven transparent decision making;" accepting the public as a partner; transparency, honesty, and acknowledging uncertainty; compassion and empathy; evaluation and reassessment of the strategy (Malecki, Keating, and Safdar 2021).

Fairness, Autonomy, or Effectiveness?
PHERCC processes and risk response strategies are generally perceived as requiring balancing between the three fundamental components of the triad: autonomy, fairness, and effectiveness.It is often challenging to attain the ideal level of each component, thus requiring the need for tradeoffs between them to prioritize one or more over the others.For example, when developing a new risk response strategy, initiators may face a situation where they have to choose between two options: one that is highly effective in reducing risk but could impact on the autonomy of the publics (e.g.: vaccine mandates), and another that protects autonomy, but may not be as effective in reducing risk and might be perceived as compromising fairness (e.g.: aggressive nudging strategies, such as COVID certificates).In this case, the initiator must weigh the tradeoff between autonomy, fairness and effectiveness, ultimately choosing the least worst option.This tradeoff could be true-to some degree-for enforceable risk response strategies like quarantine and isolation (Dong et al. 2022;Spitale 2020).However, the case of PHERCC is different: while measures like quarantine and isolation can be actively enforced (again, under the conditions defined by the Siracusa principles) communication and understanding cannot be enforced.To achieve its aims (i.e., eliciting specific protective behaviors across different publics and increasing risk awareness), PHERCC needs all the publics to be fully on board, i.e.: to ensure that the various elements of the PHERCC process incorporate this notion of justice as fairness.The same evidence should be adapted to different messages, containing the same core information, but crafted and disseminated to reach different publics.An excellent example in this sense is provided by the efforts to reach an adequate COVID-19 vaccination coverage in the Haredi communities in Israel: what proved to be crucial in this specific context was a communication campaign specifically tailored to this specific public (Schroeder, Numa, and Shapiro 2022).The vaccination campaign would not have been successful, had the communication not been tailored to that community.
In real-life scenarios, possibly characterized by urgency and limited resources, initiators could opt to try reaching as many people as possible, with messages that target primarily majority groups.This will predictably not meet the communications needs of groups who require communications that are targeted to their culture or that address long-standing legacies of distrust with public health authorities.Such strategies not only-unfairly-widen existing inequalities in healthcare (Abba-Aji et al. 2022;Bogart et al. 2022;Hussain et al. 2022), but can also cause additional burden to already stressed healthcare systems, affecting overall effectiveness (Bogart et al. 2021).Therefore, good PHERCC practices should not consider tradeoffs between fairness, autonomy and effectiveness: PHERCC actions will likely fail, when they are not perceived as fair by each public.

The PHERCC Matrix
Based on the mentioned considerations on justice intended as fairness-as a conditio sine qua non to effectiveness-we propose our framework in the shape of a matrix.We defined the PHERCC process, we identified relevant ethical principles, geared toward guaranteeing respect for autonomy and fairness across the whole process, and we propose the application of said principles in each step.There are two assumptions with meta-ethical relevance in this reasoning.First, that there are no tradeoffs between effectiveness, fairness, and autonomy; on the contrary, that aiming for fairness and respect for autonomy can increase the effectiveness of PHERCC actions.Second, in line with Rawls, that fairness and autonomy are desiderata in a modern constitutional democracy.In this sense our principles can play both an ethical role-they have intrinsic ethical value-and a pro-ethical role-they are instrumental to the pursuing of aims which have ethical value.Such principles, adopting a principlist view, have intrinsic ethical value; adopting a deontological perspective, they form a stack that contributes to the realization of fairness as a moral duty.Finally, it is important to stress the consistency of the model also from a utilitarian perspective: literature shows that fairness increases the effectiveness of the PHERCC action, producing desirable consequences (Malecki, Keating, and Safdar 2021), therefore the proposed framework has also ethical value from a utilitarian perspective.Thus, a systematic application of this framework in the planning, governing, and evaluation of PHERCC actions can be understood as an enabling factor for a fair and effective intervention, and-importantly-for a fair and effective public discourse.Our matrix is summarized in Figure 2.

Openness
Openness in PHERCC is conceptualized similarly to the concept of openness in "open science."The Draft Recommendation on Open Science produced by the UNESCO General Conference defines it as "an inclusive construct [ … ] aiming to make multilingual scientific knowledge openly available, accessible and reusable for everyone, to increase scientific collaborations and sharing of information for the benefits of science and society, and to open the processes of scientific knowledge creation, evaluation and communication to societal actors beyond the traditional scientific community" (UNESCO General Conference 2021).From a theoretical point of view, "openness" can be understood in two ways: first, as the sheer availability of information, be it datasets resulting from primary research, or code; second, as the attitude and ability to create new knowledge, the will to share it and the ability to receive it.

Transparency
According to Turilli and Floridi, information transparency is not an ethical principle per se, but rather a "pro-ethical condition," in that it enables an ethical evaluation of the information, which per se could be considered ethically neutral (Turilli and Floridi 2009).In the context of PHERCC, information transparency cannot be considered ethically neutral, as it allows participation and informed decision making, fostering fairness and autonomy.Therefore, transparency can be considered as a full-fledged ethical principle.Transparency is intertwined with openness, as they are mutually enhancing (Ball 2009).Michener and Bersch developed a solid framework for transparency, identifying two hallmarks: "visibility of information, and its inferability-the ability to draw accurate conclusions from it" (Michener and Bersch 2013).Transparency and openness play, together, a relevant role in determining the scrutinizability (and thus the accuracy) of information, which in turn has an impact on autonomy and on fairness.

Inclusivity
In line with Rawls, reasonable pluralism is a basic feature of liberal democracies.It entails the societal co-existence of "a plurality of conflicting reasonable comprehensive doctrines, religious, philosophical, and moral."Liberal democracies must avoid "friend or foe" approaches to conflicting doctrines-as long as they are not incompatible with the very idea of a constitutional democratic regime (Rawls 2005, 441).Intended in this sense, inclusivity plays a central role in the PHERCC framework.For these reasons, information should be as effective as possible for as many people as possible, including minorities.In the same way, everyone should have the ability to talk back to the system, being actively engaged and having the possibility to partake in the public discourse.PHERCC, for example, should include strategies to deal with people lacking access to Internet, or people with no understanding of the initiator's preferred language of communication.Also, inclusive approaches should attempt to ensure information won't be polarized, and ideally use communication strategies that attempt to bypass biases produced by polarized channels of communication.For example, more transparency on knowledge gaps at the beginning of the COVID-19 pandemic was needed-for instance WHO did not recommended the use of masks until later during the pandemic (Sachs et al. 2022); clearly, earlier recommendations about masks were needed.At the same time, once the use of masks was added as a recommendation by WHO, the message has caused polarization of the debate about safety and efficacy of masks.Therefore, a communication strategy focused on the effectiveness of masks to prevent the spread of SARS-CoV-2, without blaming those who did not want to use masks or accusing them to endanger the population at large, could have mitigated political The PHERCC ethics matrix: the first row describes the PHERCC process, the first column describes the principles, the intersections describe the application of the principle in the different steps of the process.As PHERCC processes are context-specific, the questions are intended to be representative and not exhaustive of the interplay between the process and the principles.
categorization on social media of the masked versus unmasked debate during the COVID-19 pandemic.

Understandability
The principle of understandability further broadens the scope of inclusivity.PHERCC actions should consider that the recipients of information are coming from different socio-cultural backgrounds, hence they not only have different world views, but also different levels of education and different predisposition to understand specific concepts with a specific communication strategy and in a specific timeframe.Understandability is a matter of "fair opportunity"which guarantees everyone the possibility to be part of the PHERCC conversation, and the application of this principle helps delivering a comprehensible message for various publics.The role of each public in shaping the initiator's ability to produce an effective message has been discussed before-as such, understandability should be a guiding principle to shape actions.The initiator is generally a trusted institution with experts in public health, and thus should provide guidance and leadership in PHERCC.The trusted initiator should ideally be a nonprofit, internationally recognized organization, such as WHO (or local institutions) with no conflicts of interests; for example, pharmaceutical companies inform the public about the safety and effectiveness of their products.Despite this activity being scientifically regulated and sound, a skeptical public may perceive such communication purely as a marketing strategy, leading to a further decrease in general trust in scientific institutions at large.
The public, with its diversity, also provides a variety of voices that should be used to generate more messages, reaching out to more people.In fact, for a message to be effective, the initiator requires the involvement of the public, with its plurality and understanding of communication, not only as the receivers of the message, but also as the co-initiator of the message.This is possible: (a) through the feedback mechanisms described in the PHERCC matrix, and (b) by deploying influencers as initiators themselveswith their own ability to reach out to specific niches and with a specific, already existing, communication system that works well with their audiences.

Privacy
Privacy is a fundamental human right, based on the assumption that everybody should enjoy a free space for "development, interaction and liberty, a 'private sphere' with or without interaction with others, free from State intervention and from excessive unsolicited intervention by other uninvited individuals" (UN High Commissioner for Human Rights 2018).Our understanding of "privacy" must be broadened by the enlargement of digital horizons, and the growth of digital footprints: "the right to privacy is not only impacted by the examination or use of information about a person by a human or an algorithm.Even the mere generation and collection of data relating to a person's identity, family or life already affects the right to privacy, as through those steps an individual loses some control over information that could put his or her privacy at risk" (UN High Commissioner for Human Rights 2018).In the context of PHERCC, privacy plays a crucial role in several elements of the process; first in the delivery of the message to the publics, second in the collection and analysis of the feedback from each public, as digital delivery and feedback systems might allow digital fingerprinting or other forms of conventional tracking, such as cookies.Therefore, adequate measures should be put in place when choosing or building the communication channels.Interference with people's privacy is a limitation of a civil and political right recognized and protected by the International Covenant on Civil and Political Rights (UN High Commissioner for Human Rights 1966) and as such it should be managed following the Siracusa Principles on the Limitation and Derogation Provisions in the International Covenant on Civil and Political Rights (The American Association for the International Commission of Jurists 1985).It is justified only in the case of an emergency which has been legally proclaimed and notified, defined in scope, extent and duration, and should respect the principles of necessity and proportionality (UN High Commissioner for Human Rights 2018, art.10).

Suggestions for Implementation
PHERCC is a multi-actor process.It involves research institutions generating evidence; local, regional, national or international initiators; software engineers and media experts developing or improving communication channels; copy strategist, copywriters, graphic designers and translators transforming evidence-based recommendations into segmented messages; publics themselves, as co-actors providing feedback in the form of evidence, and supporting the role of the initiator in shaping PHERCC; and again, research institutions processing the feedback and developing further contributions to the evidence-based process.
The implementation of the framework requires everyone involved in the process to be aware of the general aim of PHERCC, of the specificities of the step they oversee, and of the ethical values that should help shaping it.This stands true for the planning of an action, for its governing, and for its evaluation.In this sense, the initiator can assume a pro-active role, as a "primus inter pares," ensuring that the various coactors are aware of and apply the principles to the step they are involved in.This entails education and training actions-i.e.empowerment, to be conceived as a structural part of disaster preparedness-as recommended by the Sendai framework (UN Office for Disaster Risk Reduction 2015).

Strengths and Opportunities
Openness and transparency through the process foster trust and allow independent third-party verifications or audits.Adopting inclusive approaches, which consider understandability, allows the design of better tools, strategies, and messages.This in turn allows people to effectively perceive themselves as co-actors, rather than as the recipients-once more, fostering trust and adherence to evidence-based policies shaped to address a public health emergency.Privacy, clear data processes, and data management by reputable, open, transparent, and trustworthy institutions put people in the condition to freely express their opinions, even when they are very critical-and thus important to consider when aiming for autonomy, fairness and effectiveness.
Adopting this framework offers several opportunities.Populations can be provided with information, which is not only relevant, precise, and timely, but also based on their individual needs, understanding of the ongoing crisis, and their reactions to current or future measures to address the emergency.Such information should be simple to understand, and geared toward eliciting specific protective behaviors, in a way which is perceived as nonjudgmental-especially on morally loaded topics.This approach would improve fairness, and if implemented through single channels that allow the generation of messages tailored to different publics, this would not require the use of multiple channels, resulting in slow or resource intensive processes; one example is the PubliCo platform, which based on information provided by users through a survey provides personalized information based on a set of rules applied to individual users' scores (Spitale et al. 2021).We propose a paradigm shift-from policy-makers "teaching the public" with a (limited and failing) paternalistic approach, toward fully acknowledging different publics as co-actors in the process.During the COVID-19 pandemic we witnessed the explosion of an arms race in info wars: misinformation rampaged, both on mainstream and on new emerging channels (Cuan-Baltazar et al. 2020); consequently, efforts to track, debunk and contain such misinformation rampaged as well.Adopting the PHERCC framework and incorporating its principles can help avoiding or at least curbing the issue of rampaging misinformation and conspiracy theories.In fact, understanding concerns from different publics, with often polarized views, can help structure a communication strategy that suits best the different needs of each public, responding to their doubts, concerns, and the presence of information voids.

Censorship
In a crisis, such as the ongoing COVID-19 pandemic, initial assessment of scientific evidence shaping messages and public health policies is typically performed in an environment that was unprepared and in a context with lack of information.In this information void, disinformation and conspiracy theories can emerge at a quick pace, especially if guidance is not promptly provided to each public and to policy-makers.In a short timeframe and in an unprepared environment, curbing misinformation is difficult and likely relies on stronger measures; in the case of the COVID-19 pandemic we have seen limitations of personal freedoms in the form of lockdowns, and for misinformation we have seen attempts to censor harmful information, such as messages containing misinformation about vaccines (Germani and Biller-Andorno 2022).Censorship is a double-edged sword which can save lives in the short term, but can impact trust in institutions (i.e., the initiator) providing information (Niemiec 2020).Thus, over time and when a crisis is not resolved in a short period of time, censorship can allow a spiral formation and reinforcement of conspiratorial behaviors, which can damage the societal fabric and have a negative impact on the expected effects of policies aimed at addressing the emergency (Chang et al. 2022).Censorship of information can be compared to lockdown measures to reduce the spread of a virus-they are effective but short lived.Our framework considers an active role of different publics in providing feedback, which should be considered as evidence, to build an effective communication strategy.Thus, publics in this framework act as co-actors in a process of information production.The involvement of different publics in the PHERCC process can have positive consequences in the medium and long-term, thus creating an environment which discourages the spread of misinformation.This approach can be compared to mass vaccinations that equip the public with resilience against disinformation and conspiracy theories.Besides, the effectiveness of our PHERCC framework is enhanced when the public is resilient to misinformation ahead of a crisis.Preparedness requires educational measures and research to understand and identify the best tools and teaching strategies to build resilience.This would allow, with the emergence of a new public health crisis, a reduced necessity to impose strong measures and a higher involvement of different publics as coactors from an early stage of the crisis.
Public health crises can further exacerbate political and societal tensions within democratic societies.Therefore, PHERCC and the potential use of censorship to deal with misinformation can have larger effects on society than those such messages and censorship are designed for.For example, the use and abuse of censorship can lead to public surveillance, attempts to identify citizens with "deviant" opinions, as well as the generation of predictive modeling as a first step toward controlling the population.In fact, most conspiracy theories emerged during the COVID-19 pandemic have something to do with the idea that a political elite is controlling or attempting to control people's behaviors (Germani and Biller-Andorno 2021).Examples include: that vaccines are implanted chips controlled by 5G technology, or that vaccines are tools to control the increasing population size, etc. (Spitale, Biller-Andorno, and Germani 2022).These "Brave new world" scenarios emerge as potential derangements of actual risks stemming from the use of censorship, especially when used for a prolonged period.
Convincing versus Building: Top-Down versus Feedback-Loop Approaches to Information Providing information can follow two distinct paths, one coming from an initiator and directed to the public, and another, as proposed in our framework, directed from the initiator with an active feedback role provided by the publics, and directed to both the publics and the initiator.The second approach, which is a cyclic, feedback-loop approach, is preferable according to our PHERCC framework.If the process of evidence creation and knowledge formation is shared between initiator and the public, and if the latter perceives it is playing a role in such process, then knowledge will be perceived as shared and not imposed.This is relevant, as it may help reduce misinformation and debunk conspiracy theories.Therefore, the information provided should build knowledge and evidence, rather than simply convince different publics to adopt a certain behavior.Let us analyze an example from the COVID-19 pandemic: masks can effectively reduce the spread of SARS-CoV-2 (Howard et al. 2021); there are two approaches to inform the public: (a) convincing them that evidence exists to claim masks are effective-and thus encourage their use-this is a convincing effort shaped as a top-down approach to information; (b) explaining how viral transmission occurs, explaining how it can be prevented (e.g.masks) and listening to whether or not the public would want to use masks, and why, despite such evidence exists, they may prefer not to use masks, and under which conditions.Additionally, based on the feedback received from the public regarding the declining use of masks, the reasons for such decline would be made apparent.This information can be used to shape a new communication strategy aimed at informing the public about the importance of masks in preventing the spread of SARS-CoV-2.By considering the reasons for declining the use of masks and incorporating them into the communication strategy, a more effective approach to promoting the use of masks can be developed.This approach can help increase public understanding and adherence to recommended mask use, thereby reducing the spread of the virus.This is a feedback-loop approach to information.

Trust
As discussed in the introduction, risk and crisis communication has been typically studied and understood with a "business-oriented approach," i.e., the company or organization uses a set of strategies to deal with the negative impact caused by an unforeseen event.As we discussed, the role of PHERCC goes beyond that, but nonetheless we can consider some relevant aspects based on a "business-oriented approach" and our PHERCC framework.The initiator of a PHERCC process (e.g., WHO) has the role, according to our framework, to integrate public feedback in the evidence-building process necessary to shape messages directed to policy-makers and to each individual public.As discussed, failing to integrate different publics would mean adopting top-down approaches to information, which can create a distance between initiator and public, and break the loop of information creation and sharing in the PHERCC process.Such distance reflects on public trust in the institution providing the information, thus creating a negative environment in which the initiator is compared to an elite controlling information and shadowing opposing worldviews, understandings (and misunderstandings) of the crisis.The initiator, to maintain high public trust, should therefore (a) involve various publics as co-actors, as previously explained; (b) consider that its role as information provider should involve branding and advertising itself, as a business-oriented institution would attempt to do, depicting itself as a service providing value to the public, and not just to themselves: the public needs to "buy the product" (i.e.: the message) and provide feedback on how to make such product more effective, understandable and inclusive.For example, during the pandemic WHO has worked with various private enterprises, such as Google or Facebook, to create strategies and products to help reducing COVID-19 related misinformation online (Germani, Pattison, and Reinfelde 2022).Despite this, WHO hasn't advertised such collaboration for a long time.Showcasing such work could help building trust through increased transparency and increased understanding of the initiator's work.

Regulating Information and Free Speech
The debate over free speech on social media is relevant in this discussion and is particularly timely.The rise of misinformation on social media has posed great risks to people's lives during the COVID-19 pandemic.Thus, measures including censorship were taken by social media including Facebook and Twitter, in line with guidelines provided by WHO (Mosseri 2017;Twitter 2021).Recent debate on Twitter's policies and free speech initiated by Elon Musk is embedded in the debate about how PHERCC should be handled-and about how information should be handled in general.Two opposing world views address the issue from different perspectives: (a) a regulatory (censoring) approach, sees information (i.e., misinformation) as a potential weapon, whereas purists' free speech views, including Musk's, see regulating misinformation as an assault on free speech, with the consequences of fostering mistrust and further conspiratorial behaviors (Musk 2022a(Musk , 2022b)).Our framework addresses how resilience-building methods should be preferred, as resilience to misinformation bypasses the problem of regulating information versus free speech.That said, our framework relies on educational and training measures as tools to build resilience and thus increase preparedness in the event of a new public health crisis requiring PHERCC, but these effective educational strategies haven't been identified yet.

Education
Education/teaching measures to build the public's resilience to misinformation is fundamental in the emerging phases of a public health crisis, especially when evidence-based information is not readily available, since the underlying cause of the crisis is not yet fully understood.In this context, information voids are usually filled by ill-founded conspiracy theories and misinformation, unless the public is already "vaccinated," i.e., resilient, to these dynamics (Roozenbeek et al. 2022).Our framework theorizes that such voids could be filled by designing simple and effective education strategies, which are ideally inclusive and not costly, understandable, and not time consuming.So far, we identify two major limitations to this approach: (1) there is no substantial evidence on what skills should be taught to build information literacy (Albitz 2007), and (2) there is no understanding of which education plan, strategy and design would be effective in building such skills (Willingham 2020).Such limitations will need to be addressed by research institutions as quickly as possible, as the understanding of these aspects may have a relevant impact on the PHERCC process and the functioning of the proposed framework once the next public health emergency arises.Importantly, education strategies, to be effective, require trust in the initiator proposing, designing, and implementing the education strategy.In this context, collaborations between public institutions (e.g.WHO) and private agencies (e.g.social media platforms) could facilitate the process and train users with effective educational interventions while they go about their daily lives (Germani, Pattison, and Reinfelde 2022).

Data Access and Contextualization
Openness is a core principle in the proposed PHERCC matrix.As explained, this implies that data should be available to initiator and publics alike.One issue is that, without public resilience to misinformation and with limited instruments to understand and read publicly available data, such undigested pieces of information can be freely misinterpreted or can be manipulated to fill information voids, thus having a negative effect on the PHERCC process.Educational approaches will also need to provide tools to make publicly available data understandable without restricting access to them.Other issues which limit openness and the possibility to freely share data between initiator and public are due to the current functioning of the scientific publishing business, and in particular the limited, albeit growing, presence of open access journals and accessible publications, as well as the existence of structural and stylistic limitations in the communication of academic research.

Feedback Bias and Propagation
As proposed, publics should participate in the PHERCC process not only as recipient of a message, but also as co-actor in building evidence to generate new messages.The role of the initiator, as specified, is to provide guidance and leadership in the process, but integrating public feedback into an effective action has the risk of propagating the initiator's bias toward evidence and the evidence received through public feedback.For example, a specific public may report to the initiator the need to discuss the role and differences played by traditional vaccines versus mRNA vaccines, as the public seem to be concerned that mRNA vaccines could alter DNA.The initiator could then attempt to explain how mRNA vaccines work and why they do not alter DNA-however this message can be read and understood by some as an attempt to defend the roll-out of vaccines despite lacking evidence to protect the interest of pharmaceutical companies.To prevent these misunderstandings, we propose that the initiator investigates the potential effects of a given action on a public sample of interest, who reports their feedback actively-thus, again, allowing the public to use active social listening systems to participate in the PHERCC process as coactor.Such feedback would provide immense value and understanding of how a certain message, its wording, design, and adopted communication channel could lead to a desired or undesired effect.

The "Boaty Mc Boatface" Problem and the Is-Ought Issue
In March 2016 the United Kingdom's Natural Environment Research Council (NERC) launched an online poll to find a name for a new, big, and expensive polar research ship.Everyone from the public could participate, suggesting a name or voting for one of the suggestions.The results were clear: NERC's new jewel, 15,000 tons and 129 meters of might, would have been called … "Boaty Mc Boatface."NERC was clearly hoping for something different, more highbrow and less prankish-in fact, they overruled the public's decision and chose a more sober solution, opting for "RSS Sir David Attenborough" (NERC 2016).The winning crowdsourced name, "Boaty Mc Boatface," lives on in two ways: as the name of one of the ship's autonomous submersibles, and as a memento: when you let the public decide, there may be surprising and disappointing results.This same phenomenon, which makes Boaty Mc Boatface's story such a good joke, might have tragic outcomes in the PHERCC context.What if through the feedback loop of a PHERCC action the initiator realizes that the public is strongly in favor of actions that are morally unjustifiable, such as killing all infected people, or denying treatment to minorities on the base of vaccine status or racial arguments, the elderly, or people with preexisting conditions?And what if, despite all evidence provided, there is a strong preference for mitigation measures that do not have any scientific backing, such as stocking hydroxychloroquine rather than vaccines for COVID-19?This is just a special case of Hume's is-ought problem-i.e.: we cannot derive prescriptive conclusions from descriptive premises.MacIntyre proposed a solution based on the notion of telos: "human beings [ … ] have a specific nature; and that nature is such that they have certain aims and goals, such that they move by nature toward a specific telos."Good is therefore whatever allows to pursue that aim, hence it is possible to ground moral judgment on facts (MacIntyre 2007, 148-50).It is not our intention to delve into metaethics with this work, which in that respect aims to be as agnostic as possible.But this issue, one of the fundamental struggles of empirical ethics, cannot be ignored.In this sense, it is important to keep in mind that we propose to incorporate the publics in a multiactor discourse, which is not the same as "letting people decide what to do."Long-term empowerment strategies, as detailed before, will mitigate the risk of incurring in Boaty McBoatface scenarios.But reality and contingency will call for flexibility in the implementation of the framework.

CONCLUSION
Although we started to probe feasibility and explore the implementation of this approach in different settings, ranging from Switzerland to Hong Kong and Singapore, much more research is needed: PHERCC is a complex, multifaceted, and multi-actor process, and a crucial component of risk preparedness strategies.As the COVID-19 pandemic demonstrated, PHERCC is dense of ethical implications and potential pitfalls.A common misconception is to perceive that PHERCC actions require a tradeoff between justice-intended as fairness, autonomy, and effectiveness: while this can be true for other risk control measures, those which can be enforced, this is not true for communication and understanding-which cannot be enforced by any means.We therefore identify fairness as the main driving moral value of PHERCC actions, and propose a framework encompassing five principles: openness, transparency, inclusivity, understandability, and privacy.The matrix we propose can help planning, governing, and evaluating PHERCC actions, incorporating these principles across the board.A consistent implementation of the PHERCC matrix, keeping into account its strengths and limitations, fosters societal resilience by strengthening democracies.Finally, it can help to develop and bolster people-centered multi-hazard communication mechanisms and social technologies that are both fair and effective, in accordance with the recommendations of the United Nations Sendai Framework for Disaster Risk Reduction.

Figure 1 .
Figure1.The looped ladder of PHERCC: evidence is used by an initiator to deliver a message to the public through a specific channel that allows feedback to be incorporated as new evidence to craft the next iteration of communication between the initiator and the public.

Figure
Figure2.The PHERCC ethics matrix: the first row describes the PHERCC process, the first column describes the principles, the intersections describe the application of the principle in the different steps of the process.As PHERCC processes are context-specific, the questions are intended to be representative and not exhaustive of the interplay between the process and the principles.