Patient and public co‐creation of healthcare safety and healthcare system resilience: The case of COVID‐19

Abstract Introduction Healthcare system resilience is a conceptual approach that seeks to explore how health services adapt and respond to variability in demand and resources. As has been witnessed since the beginning of the COVID‐19 pandemic, healthcare services have undergone many reconfigurations. One understudied aspect of how the ‘system’ is able to adapt and respond is the contribution of key stakeholders—patients and families, and in the context of the pandemic, the general public as a whole. This study aimed to understand what people were doing during the first wave of the pandemic to protect the safety of their health, and the health of others from COVID‐19, and the resilience of the healthcare system. Methods Social media (Twitter) was used as a method of recruitment due to its ability for social reach. Twenty‐one participants took part in 57 semistructured interviews over three time points from June to September 2020. The included an initial interview and invitation to two follow‐up interviews after 3 and 6 weeks. Interviews were conducted virtually using Zoom—an encrypted secure video conferencing software. A reflexive thematic analysis approach to analysis was used. Results Three themes, each with its own subthemes were identified in the analysis: (1) A ‘new safety normal’; (2) Existing vulnerabilities and heightened safety and (3) Are we all in this together? Conclusion This study found that the public had a role in supporting the resilience of healthcare services and systems during the first wave of the pandemic by adapting their behaviour to protect themselves and others, and to avoid overwhelming the National Health Service. People who had existing vulnerabilities were more likely to experience safety gaps in their care, and be required to step in to support their safety, despite it being more difficult for them to do so. It may be that the most vulnerable were previously required to do this extra work to support the safety of their care and that the pandemic has just illuminated this issue. Future research should explore existing vulnerabilities and inequalities, and the heightened safety consequences created by the pandemic. Patient and Public Contribution The National Institute for Health Research (NIHR) Yorkshire and Humber Patient Safety Translational Research Centre (NIHR Yorkshire and Humber PSTRC), Patient and Public Involvement and Engagement Research Fellow and NIHR Yorkshire and Humber PSTRC Patient Involvement in Patient Safety theme lay leader are involved in the preparation of a lay version of the findings within this manuscript.


| INTRODUCTION
The resilience of a system is defined as its ability to maintain operations when there is a disruption. 1 To do this, resilient systems are posited to be those able to respond, monitor, learn and anticipate 2 in the face of organizational threats. To those responsible for health systems, the COVID-19 pandemic can be categorized as an irregular threat that is unlikely to happen and is difficult to prepare for. 3 The COVID-19 pandemic has been a global health crisis which has demanded large-scale population behaviour change. Globally, the daily lives of the general public were abruptly impacted in early 2020, with many countries instituting nationwide lockdowns in the interest of public safety. In April 2020, people across the United Kingdom were ordered to stay at home and leave only under certain circumstances. National Health Services (NHS) were suddenly reorganized, and resources were redirected to prepare for the COVID-19 outbreak and treat increasing numbers of  patients. This had enormous implications for the way, both regular and irregular users of the NHS, experienced and accessed healthcare services, and for the quality and safety of care. 4,5 Given the large-scale adaptations that have been required nationally and globally, the COVID-19 pandemic in many ways represents a very visible, tangible expression of the concept of resilience, 6 and one upon which it is useful to apply a resilient healthcare conceptual 'lens'. Foundational to the concept of resilient healthcare, is that people acting within a system represent one of the main sources of adaptive capacity, and are thus key to the ability to maintain stable performance across variable conditions. 1 Whilst staff within healthcare organizations have traditionally been seen as the main protagonists in this source of adaptive capacity, 7 it is increasingly becoming recognized that patients and their families also contribute to the capacity of a system to adapt to fluctuating conditions. [8][9][10] The ways in which patients and families might do this are many and varied. For example, patients and families have reported chasing appointments, following up on missed care and checking medications. [9][10][11] This type of activity has come to be thought of as variously 'scaffolding' the healthcare system to support its ongoing performance, 7 or 'propping it up' when care is suboptimal or there are significant structural 'safety gaps'. 12,13 Whilst the role of patients and families in supporting their safety is increasingly being recognized and documented, 9-11,13-15 hitherto this supporting role has been limited to those directly 'using' services.
However, given the enormous change to the fabric of society that was experienced in the early phases of the pandemic-often with the expressed purpose of 'flattening' the curve, or 'saving lives' 16 -it is clear that members of the general public, including those who were not active users of health services during the COVID-19 outbreak, were suddenly asked to undertake activity to support the NHS. This paper presents the findings of a study within which we examined the experience of members of the public during the early phases of the COVID-19 pandemic, within the United Kingdom. We aimed to understand what people were doing during the first wave of the pandemic to support the safety of their health and the health of others, and in doing so, if and how this may have supported the resilience of the healthcare system. To do this, we asked the following research questions: (1) What are the public doing to individually and collectively support the safety of themselves, their immediate networks, society and healthcare systems?
(2) How did this change over time?

| Design, recruitment of participants and setting
This study used a qualitative interview method. The ontological paradigm of relativism was assumed, which suggests that researchers can never gain a fully objective access to the world since the world in an objective sense cannot be known. Using a postpositivist epistemological stance, researchers were, therefore, able to use the qualitative interview method to access, explore and gain insights into subjective human experiences, via a collaborative process mediated by human understanding. 17 Smith and Osborn 18 ,p.53 refer to this process as 'the participants are trying to make sense of their world, the researcher is trying to make sense of the participants trying to make sense of their world'.
Participants were recruited via the social media platform Twitter.
An invitation to participate was posted on Twitter, with recruitment proceeding using convenience sampling. Initial eligibility criteria were purposefully broad, including anyone over the age of 18 who could speak English. Those who expressed interest were emailed the information sheet, also available in an easy-read format, and scheduled at a mutually convenient time for interview if they still wished to participate. However, once 10 participants were recruited, demographic information was reviewed and purposive sampling was conducted for the remaining sample to ensure diverse and representative views were captured.

| Analysis
Fifty-five interview transcripts were included in the analysis. Our approach to analysis was principally inductive but organized around the sensitizing concept 20 of patients, family or public supporting ('scaffolding') their health and healthcare service level outcomes. This approach allowed us to explore our findings with reference to the broad research objectives, whilst providing a point of reference to guide the analysis of the data.
A reflexive thematic analysis approach was used. 21  The summary documents formed the unit of analysis for regular meetings between all authors, revisiting the raw data where necessary. The coding framework was discussed and refined until a consensus was reached. This included drawing upon the data set to gain a holistic view and highlighting significant extracts to define each theme to ensure that the analysis captured information relevant to the research questions, but also included novel perspectives. A detailed log of theme development was kept.
The study followed the consolidated criteria for reporting qualitative (COREQ) research. 22

| RESULTS
The results are presented in three themes, each with its own subthemes. These are: (1) A 'new safety normal' (Feeling cast adrift, Adapting to support and Balancing risks and unanticipated outcomes); (2) Existing vulnerabilities and heightened safety (Personal circumstances and safety 'capital', and Inequity of healthcare and information access) and (3) Are we all in this together? (Interpreting others' safety behaviours and motivations, and Government mistrust and its influence on public approaches to safety).

| Personal circumstances and safety 'capital'
The ability of people to 'reach in' into the system to support themselves was dependent on their capital and resources, meaning that some could support their own safety more easily than others. The unequal provision of information and support was also highlighted, whereby government communication to people from ethnic minorities who had health conditions had been poor and was felt to be directly linked to the deaths of friends.
Not one of them was receiving any information from the government, or support, and they were all isolating bar one…. Only one gentleman, he was Caucasian, he died in his flat, but he was receiving food parcels, he was receiving support for his mental health, but all the others wasn't receiving anything, and they were all ethnic minorities. So that does concern me. P14 (Interview 1)

| Are we all in this together?
There was a general sense of empathy for how difficult others' circumstances were during the pandemic, contributing to societal efforts to 'flatten the curve'. Differences in peoples' approaches to risk and safety were highlighted over time. Perceptions of the government's response to the virus appeared to directly impact peoples' safety behaviours, resulting in many using their own initiative rather than following government guidance.

| Interpreting others' safety behaviours and motivations
Participants became aware that others' behaviour could be a danger to their own safety and paid attention to others' safety behaviours and motivations because there was suddenly a collective risk to health. Most participants acknowledged that their approach and behavioural adaptations were not shared by everyone. Some felt that this was a result of a lack of understanding or disregard for the rules.
A few participants felt it was young people who were not using common sense and causing the virus to spread, a perspective that links to the development of divisions in society during the pandemic.
Society seems to be splitting into those, say, under thirty who seem to believe that they can go out, they can go to raves, they can go to parties, and those other who are saying, oh, well, we'll go out and we'll be sensible and we'll think about things. P5 (Interview 2) Participants tended to have the view that they wanted to 'do their bit' and make a positive contribution to society by following government guidance, but also taking extra precautions they felt necessary. This was with a view that others ought to do the same to be beneficial at a societal level. However, some considered that this view was not shared by others, who acted against guidance, albeit understandably, rather than pulling together for the greater good.
You see all this queuing at Primark and I'm just thinking … why? But I think there is something more ALBUTT ET AL.
| 1473 than just going shopping, it's actually a day out for some people. 'Cause lockdown is not easy, we are social creatures. P3 (Interview 1)

| Government mistrust and its influence on public approaches to safety
While personal perceptions of the government's response to the virus varied, many felt that there had been a shift in emphasis from what was saving lives and protecting the NHS to protecting the economy. A small number of participants who were more mistrustful of the government felt they were acting disingenuously. On the other hand, many individuals spoke about 'muddling through' with the sense that nobody really knew what the best approach was, even those setting the guidance.
I think that's the difficult thing about everything, isn't it, we don't know, and even the government, they don't know, they've never dealt with it before. We're all just really trying to go along in the dark, aren't we blindly. P20 (Interview 2) For those finding the government guidelines to be particularly ambiguous, there was an increasing responsibility to behave according to their own initiative. Examples of the government not following the guidelines it set out fuelled government mistrust and resulted in participants taking greater personal responsibility for their decisions. In the perceived absence of trustworthy information, many participants referred to using 'common sense', whether that be to impose stricter behavioural adaptations, or to interpret the rules in a way they deemed logical. This included those who were shielding and who were anxious about the easing of measures for economic reasons. Guidance was in some cases adapted to suit their circumstances, for example, bubbling two households of more than one person.
Other participants interpreted guidance with a focus on maintaining as much normalcy as possible for themselves and their families. This was particularly pertinent for one participant who cared for a child with autism. There was also some evidence that as time progressed, people saw others violating the rules, including prominent public figures, and so the credibility of the public messaging experienced attrition.
Researchers have characterized the impact of the pandemic, and evidence that others had not followed the rules around isolation, as being morally injurious. 29 They argue that efforts to limit the spread required people to prioritize the safety of others over their own moral commitments to friends and family, for example, seeing and visiting relatives and attending funerals. 29 Whilst our participants did not describe their own experiences as being morally injurious, the impact of isolating and monitoring information to keep safe impacted their mental well-being.

| Limitations
It should be noted that social media (Twitter) was used to recruit participants for the study. Although Twitter does have the ability for social reach and was felt to be an appropriate method of recruitment given the limitations created by the pandemic, this may have biased the sample to include only those who use social media. Using a different method of recruitment that did not rely on social media may have resulted in a sample of people with different perspectives than those reported. Finally, we were not able to follow up with all participants across all time points, which might have impacted the inferences from our study relating to changes in experiences over time.

| CONCLUSION
Our study exploring how the public co-created health and healthcare safety during the COVID-19 pandemic, found that they had an important role in supporting the resilience of healthcare services and systems during the first wave of the pandemic. This was evident through significant adaptions to their behaviour, explicitly undertaken to protect themselves and others and to avoid overwhelming the NHS. Our findings illuminate existing inequalities and inequities for those who are most vulnerable, and who were already required to safely navigate inadequacies or 'gaps' in their healthcare. These