Care workers and managers' experiences of implementing infection control guidance in an epidemic context: a qualitative study in the South East of England, during the COVID­19 prevaccination era

,


Introduction
Te national response to the COVID-19 pandemic has heavily impacted the adult social care sector in England, with important consequences for both people providing and receiving care. Te most tragic of these has been the death of many people, especially among older adults (65+) living in care homes, with the UK witnessing one of the largest scales of loss of lives in Europe [1]. High mortality was also experienced in the workforce. As of May 2020, care workers in residential and domiciliary care accounted for the highest number of COVID-19 deaths among those working in social care and aged 20-64 [2].
Tis has raised questions on the timing and efectiveness of the policy response in adult social care settings and how it contrasted with the response in the National Health Service (NHS) [3,4]. Since the earliest stage of the pandemic, media reports, grey literature, and research have shown that in spite of its prompt response and the large sacrifces made by care workers and managers to provide care in the midst of a new pandemic [5], the sector faced many challenges. Tese included delay in the development of policies specifc to adult social care [3], unclear protocols around COVID-19 status of patients discharged from hospitals into care homes or domiciliary care [3,4], shortage of personal protective equipment (PPE) [3,[6][7][8], and sparse testing and contact tracing [6][7][8]. Along with these systemic challenges, social care staf in residential and domiciliary care had to rapidly implement new measures to protect themselves, their colleagues, and those receiving care from infection, with new responsibilities, added workload, and immense pressures [9] at a time of uncertainty.
Infection control in residential and domiciliary care settings is always complex for multiple reasons. Tese include the high prevalence of frailty and comorbidity among those receiving care [5,10], and physical proximity between carers and people receiving care. In domiciliary care, this is complicated by the domestic environment which makes it hard to segregate diferent types of caring activities. Tese factors might make it difcult to implement COVID-19 infection control measures, such as wearing masks, social distancing, shielding, and selfisolation [11][12][13]. While the epidemiological impact of the pandemic on the social care sector has been welldocumented, the experience on the ground of care staf is less well explored [5,7,14].
In the UK, adult social care refers to a wide range of activities and services that support older people (65+) to stay well, safe, and independent. Tese include "personal care" (i.e., assisting people with washing, using the toilet, and getting dressed), help with household tasks (e.g., cleaning and cooking), and support with leisure and social activities [15]. Residential and domiciliary care are part of adult social care. Residential care, also referred to as care homes, provides a living environment (with communal spaces shared with other residents) and person-centered care for older adults who need support with personal care and daily tasks. Nursing homes are a type of care homes that provide 24hour registered nurse care for people with higher-level care needs. Domiciliary care, also referred to as homecare, provides personal, domestic, and social care in a person's own home. In these settings, care workers are staf employed to provide diferent types of person-centered care described above, along with administrative tasks [16]. Tey are required to be trained in providing personal care but are not registered professionals (by contrast, for example, with nurses). Care work is emotionally and physically demanding, and it requires both interpersonal and practical skills [17]. Care workers perform a range of diverse tasks in complex environments (e.g., people's own homes) that pose diferent risks. Tis requires a constant adaptation of infection control practices to minimise the risk of infection transmission while also being attentive to a person's quality of life, wellbeing, and dignity [18].
In England, adult social care is both paid privately and commissioned by local authorities, provided mostly by private businesses of diferent sizes [17,18], and regulated by the Care Quality Commission (CQC), an independent body that sets codes and standards of care [6]. Registered providers have to comply with national infection prevention and control standards, which are assessed and monitored through inspections. In terms of infection control policy, this means that residential and domiciliary care providers follow national and local government advice, but this is variously adapted and implemented to refect each provider's structure and operational strategy. Infection prevention and control training is mandatory for care workers in registered domiciliary and residential care, although the content, assessment, and frequency of training, as well as access to PPE, vary across providers. Previous research [13,14] has shown that high staf turnover, inconsistent monitoring, lack of managerial support, sparse national policy, and work pressures are recurring issues in relation to infection prevention and control practices in adult social care.
As at 2020/2021, 1.54 million people worked in adult social care in England, a workforce bigger than the NHS (1.3milion) [18]. Te majority worked in domiciliary (44%) and residential care (40%), with care workers making up 53% of all jobs in the adult social care sector [18], which includes 17,700 CQC registered organisations. Te majority are micro and small businesses, with around 85% of organisations having fewer than 50 employees [18]. Despite a workforce bigger than the NHS, the adult social care sector is characterised by fragmented services and job market [6,19], high staf turnover and vacancy levels [18,20], low wages [21,22], and slower wage growth than in healthcare, increasing casualization, especially for domiciliary care workers [18] and low level of unionisation.
In this study, we aimed at understanding the experiences, concerns, and solutions of care workers and managers in implementing COVID-19 infection control guidance in practice. Te fndings presented in this paper refer primarily to the frst wave (March-July 2020) and prevaccination phase of the pandemic. Tis qualitative work was part of a mixed-methods implementation project aimed to understand what guidance and interventions would improve the working life of care staf in domiciliary and residential care whilst reducing the risk of SARS-CoV-2 transmission. Te project also included the development of a mathematical model of infectious disease transmission in domiciliary care, using COVID-19 as a test case [23], and an integrated review of national infection control guidance for social care and newspaper coverage of infection control issues in adult social care [24].

2.1.
Design. Semistructured interviews were chosen as the main method of data collection as they allow to explore indepth participants' experiences, thoughts, views, and feelings on a particular topic [25]. Te interviews focused on participants' experiences, solutions, and concerns around the implementation of COVID-19 infection control guidance in residential and domiciliary care.
Te topic guide (Appendix 1) was developed by two research fellows (LBJ and LB) in collaboration with two public advisors (JF and EL) and was designed to explore topics general to all participants but also to be adapted to examine topics pertinent to roles, such as a manager or care workers.

2.2.
Participants. Ten participants (nine women and one men) met the following inclusion criteria: being 18+, speaking English, and having been working as managers or care workers in domiciliary or residential care in the Southeast of England within the period of interest of 1st February 2020 onwards.
Te recruitment proved complex due to the extremely difcult situation that the adult social care sector was facing at that point in the pandemic, and the need for researchers to adapt working practices and data governance to the novel situation of doing research remotely. Researchers were open to review recruitment strategies to respond to possible barriers and challenges, and multiple sampling methods were used to widen the identifcation of potential participants. Participants were recruited using a mixture of convenience and snowball sampling, both commonly used in qualitative research to capture community-based data [26]. Recruitment was narrowed down to places in Kent, Surrey, and Sussex, identifed as relevant through care and demographic statistics. Potential interviewees were identifed through existing and publicly available contacts (i.e., care providers listed on local authority websites) and were contacted by the researchers (LB and SS) either via e-mail or phone call.

Data Collection.
Between November 2020 and January 2021, we conducted semistructured interviews (n = 10) with care managers and workers in domiciliary and residential care in Kent, Surrey, and Sussex. Interviews were conducted by two research fellows (SS and LB) with expertise in qualitative research methods applied to health and care research. Following COVID-19 measures in place at the time of data collection, interviews were conducted remotely on Zoom. Interviews were audio recorded and lasted between 36 and 74 minutes. In line with the previous research works [26,27], the researchers addressed potential issues arising when conducting qualitative video and phone interviews [26,28,29]. Te equipment was checked regularly to minimise technical issues and participants were ofered support on how to download and use Zoom prior to the interview.
Te concept of "information power" [30,31] guided researchers' decisions around sample size. Information power in qualitative research refers to the idea that "the more information the sample holds, relevant for the actual study, the lower number of participants is needed" ( [32], p. 1759). Participants in this study held expert information and knowledge relative to the focus and aim of the research, being the ones implementing COVID-19 guidance in their everyday working life. Tis characteristic, along with researchers' experience of doing qualitative interviews, and the team's expertise on adult social care and infection control, made a small sample suitable for this study. Recruitment stopped when the research team agreed that the data collected ofered new insights relevant to the research question.

Data Analysis.
Interviews were audio recorded and transcribed by a professional transcription service. Transcriptions were anonymised and all identifying information were removed. Data were analysed using a framework method [33,34] consisting of familiarisation, identifcation of a thematic framework based on the interview topic guide, coding, charting, and interpreting. Coding was undertaken separately by LB and SS, and initial codes were entered in NVivo12. Temes were then discussed and agreed upon by the researchers. Te analysis was both inductive and deductive, as the themes were organised according to the topic guide and derived from the interviews. Temes were discussed and further refned through engagement with stakeholders in adult social care in Kent, Surrey, and Sussex [35].

Trustworthiness.
To ensure rigour and trustworthiness of data analysis, two researchers conducted the coding, charting and interpretation of the data. Interim fndings were discussed, further interpreted and fnalised with two public advisors and research team. Tese were also presented and discussed with stakeholders in adult social care at diferent stages. Tis process ensured that alternative interpretations were considered as the analysis progressed. Authors used the consolidated criteria for reporting qualitative studies (COREQ) checklist [36,37] to guide the report of this study.

Ethical Considerations.
Te researchers provided potential participants with a participant information sheet and consent form and invited them to ask any clarifcations before and at the point of data collection. A multiple yes/no questions consent form was read out by the researcher to the interviewee at the time of data collection, and consent was digitally recorded. On completion, each interviewee received a £15 voucher.

Findings
Five home-care managers, one care-home worker, one domiciliary care manager, and three workers in domiciliary care took part in the interviews. All participants had been working in the social care sector for many years (from three to twenty years) except one, who had just started working as domiciliary care worker at the beginning of the pandemic. Nine participants had worked in diferent health and care services before being employed in their current role, including ambulance services, nursing (both in NHS hospitals and nursing homes), and as therapists. Five participants had Table 1: Tematic analysis of interviews with staf and managers in domiciliary and residential care on their experiences of implementing COVID-19 infection control guidance during the frst wave of the pandemic.

Teme
Issues reported Examples (1) Increasing visibility and support for the sector (i) Lack of public recognition for adult social care workers compared to colleagues working in the NHS (ii) Huge sacrifces being made by individuals in order to keep colleagues and residents safe with little or no recognition and/or support (e.g., pay rise and incentives.) "I think it has been really, really difcult for a lot of our carers (. . .) because quite often they're short stafed and so they're having to work harder and they're exhausted and they're picking up extra shifts and I think it'd be nice for them to have a little bit more recognition than they get." (RM1) "I think a lot of the focus has been on the NHS, but I really do think they should be defnitely expanding that" (RM1) "We made ID cards for every single staf member that worked within the service to allow them access for priority shopping. One of our staf was refused entry to waitrose because he worked in a care home and not for the NHS" (RM2) "We don't get paid enough for what we do. And the government has just put a pay-freeze on care workers for the next three years. Tat's nice isn't it?" (RW1) (2) Impact of negative messaging about the sector (i) Concerns that there is not enough positive publicity about the sector and that the main image of care home is uncontrolled outbreaks "(What) would be good to come out of this, is some kind of positive messaging that we as care providers, did the best we could with the resources that we had" (RM2) (3) Feelings of isolation (i) Feelings of isolation and lack of moral support from other health and care colleagues/bodies (ii) Domiciliary care staf predominantly working alone, and in general, only engaging with colleagues online "I remember my clinical lead saying to me, just after one of our registered nurses died, "everybody says that we're all in this together but not one of them [in the CCG] has set foot in this building""(RM 2) "I think for the managers, erm, it would nice if they had that phone call from the CQC or social services or just, you know someone to let them know that actually ... they do a really good job. " (DM1) "Since COVID it almost seems like I've been isolated, in a way"(DW1) (4) Accessibility and usability of guidance Dissemination (i) Guidance often released on a Friday afternoon, with an expectation for implementation by the following Monday (ii) Guidance released by several diferent organisations, and uploaded to several diferent sites "Sending out the same document from ten diferent sources does more harm than good" (RM2) "It's a bit like now the lateral quick testing they give it all out, say "start using it on Monday" but don't give you the information until Friday and then suddenly we're all meant to have done the training and able to support this" (RM4) Format (i) Guidance was very lengthy and wordy and managers often had to create fow charts, diagrams, posters, and videos to communicate guidance efectively to staf and those receiving care "If there was a set time or a day that you knew it would come out and not on Friday at ten to fve (. . .) it would be fantastic" (DM1) "If you read the guidance around the use of masks it's like 20 pages long, but actually you can make it look cool but give the same message in a page, a page sized poster and that was, that's what I mean about accessibility. "(RM2) "Everything they send out, I read it from the beginning to the end, I print it of, and I put it in the COVID fle. And then more stuf will come through, so you read it. And you can't retain all that information, so I think sometimes it'd be really good if they picked up on the key points" (RM5) "I know it's got to be done [lateral fow test], but i don't think they've thought about it properly, you know. (...) the recording of the testing is just a nightmare" (RM5) "I think the care homes have had a lot of guidance, a lot more, sort of structure. Yeah we're just sort of fnding it out as we're going along (...) any policies coming out [for care homes] we could readapt [to homecare]" (DM1) (5) Social care staf as agents in producing and sharing good practice (i) Many examples of staf creating and using fow charts, posters, and videos to help colleagues, residents, clients, and family members understand and follow guidance "[Te ofce team put a video together for staf] so at least they could have a visual of what was expected in the reality of actually going to someone's house, getting out of your car. Wiping down keys, little things that would go alongside that PPE" (DM1) "Other ones have, like, done it with pictures, so they've, like, sent pictures of how to do it. And they've just done lots of "easy reads" really, for people to access things, so I think that's really helped." (DW1) "[We made] this fow charge process, erm, and it's across two organisations so it's exactly the same in every service and it is basically if you receive a positive result you do this, you do that, you call PHE, you call your regional manager, you write your letter, you send it to all of your relatives, you let all of your residents know and you tell all the other staf.""(RM2) (6) Managing expectations and the impact of conficting messages in media (i) Families can have expectations that are not consistent with guidance, often due to conficting information from the news and social media "Te reason that we didn't watch the news was because the guidance that was going out to members of the public and guidance that was specifc to us working like in care homes were two completely diferent things."(RM2) "It's trying to manage the staf, the members and the families" expectations that when they hear the news, you're constantly having to say to them "yes I know they're saying that but it's not happening" and I've felt that we've had to do this all the way through" (RM4) "During various stages people hearing diferent things from newspapers, news articles, online from this source, etcetera, erm, meaning that they question (. . .) the guidance we ofcially have" (RM3) (7) Improving communication with hospitals (i) Information on testing and isolation provided by hospital staf to residents and families was inconsistent with social care guidance (ii) Unclear responsibilities in liaising with other services (e.g. ambulance services) "My second outbreak in October was one of my members being discharged from hospital untested and came back here with COVID" (RM4) "Te manager at the ambulance service rang me and shouted at me because the ward had booked the transport, but the ward had failed to tell the ambulance crew that it was cancelled, so the ambulance crew then ring me and shout at me for booking an ambulance for someone who wasn't going to be returning home" (RM2) worked both in domiciliary and residential care during their career.
Te researchers identifed eight themes, as summarised in Table 1.

Increasing Visibility and Support for the Sector.
Participants (n � 6) reported concerns about the lack of positive visibility and support for the sector from the public, policymakers, and politicians. Care workers and managers felt that public awareness and recognition focused on the NHS, and that the work and sacrifces made by those working in social care had received little or no public recognition at all. Tis was perceived as unfair, particularly in light of the large sacrifces made by individuals to keep colleagues and people receiving care safe, with an important impact on their lives both within and outside of work. For example, participants described living on-site (i.e., a care home) for long durations of time (e.g., several months), working with very few agency staf for infection control reasons or trying to not rely on agencies at all. Participants reported an immense amount of vigilance within and outside work, for instance, some discussed wearing diferent clothes every day and limiting or avoiding socialisation, including seeing friends or family. However, they felt that none of these was recognised in the public discourse, making their work and sacrifces invisible.
Te perception of lacking support and visibility is also linked to the policy response, with practical implications around infection control and support in place for care workers. Participants reported instances where they had to proactively mobilise to get access to benefts that other colleagues in the NHS had access to, such as priority shopping. For instance, a manager described having to personally fle a complaint on a major supermarket chain asking to recognize care home staf as eligible for priority shopping after one of their staf was turned away.

Te Impact of Negative Messaging about the Sector.
Some participants (n = 3) expressed concerns about negative messages about adult social care in the media and the public, where care homes were mostly associated with the image of outbreaks. Participants were aware that the adult social care sector was in the spotlight and shared the feeling that there was no enough positive publicity about the sector, for example, a participnat commented: "what would be good to come out of this [research], is some kind of positive messaging that we as care providers, did the best we could with the resources that we had" (residential-care manager 2). Tese concerns were widely shared by adult social care stakeholders in the region.

Feelings of Isolation.
Feelings of isolation were reported by four participants, although the sources and dynamics of isolation varied. Managers and staf in care homes reported a lack of feeling of closeness and support from other health and care colleagues and bodies, for example, a participant told us: "I remember my clinical lead saying to me, just after one of our registered nurses died, "everybody says that we're all in this together but not one of them [Clinical Commissioning Group (CCG) professionals] has set foot in this building" (residential care manager 2).
For people working in domiciliary care, isolation also emerged as a condition related to the type of work and care provided. In fact, domiciliary care staf are predominantly working alone, and in general, only engaging with colleagues online; therefore, they are less able to instill team working practices in the way that their counterparts in care homes are able to.

Accessibility and Usability of Guidance.
All managers (n � 6) interviewed in our study discussed challenges around the accessibility and usability of infection control guidance. We grouped these challenges into three subthemes. strongly, strongly worded conversations with some discharge coordinators" (RM2) "(At the beginning of the pandemic) I couldn't get the visor, it just took forever to come, um, I think I waited something like eight weeks for it, or maybe even longer" (DW2) "(At the beginning) it was really hard to ... you know, all our suppliers you'd go to and there would be backlogs, gloves weren't coming through. We didn't have the PPE either (DM1)" "(Up to May, I would call PHE on a daily basis about testing) I wasn't signposted for any support" (RM4)

Dissemination.
Managers indicated that guidance was often released at inconvenient times and at short notice, for example, on Friday or late afternoons (around 5pm), and with an expectation that this is implemented on Monday. Such expectations were considered impractical and frustrating. Where the guidance released was also problematic as the same guidance was released by several diferent organisations and uploaded to several diferent sites. Participants found this unhelpful, because they had to search for the most up-to-date guidance at any one time.

Format.
Participants described the guidance as being lengthy and wordy, and this created a number of challenges. Many stated how challenging it was to have to read, digest, and remember 75 pages of guidance, particularly at a time where information was being updated constantly. Managers and staf also found that guidance could be more clearly and concisely written, and they stated that either themselves, their colleagues, or their organisations had created fow charts, diagrams, posters, and videos to communicate the latest guidance simply and efectively to staf and to the people receiving care.

Suitability.
Participants working in domiciliary care indicated that there was often no specifc infection control guidance for care provided in the person's home and they needed to adapt guidance from the care home sector. Tey also pointed out that guidance for care homes was more and better structured compared to that available for domiciliary care, which was quite disjointed, as explained by a domiciliary care manager: "I think the care homes have had a lot of guidance, a lot more, sort of structure. We're just sort of fnding it out as we're going along" (domiciliary care manager 1).

Social Care Staf as Agents in Producing and Sharing Good
Practice. All participants reported examples of the active role of the sector in producing and sharing good practices, particularly in relation to guidance, communication with families, and staf wellbeing. Managers and staf drew on and adapted previous training and experiences of infection control (e.g., from nursing hospitals, other care homes) to navigate the new risks and lack of specifc guidance. For example, a care home manager commented as follows: I quite actively went through foot and mouth disease which was back in 1999/2000 and that was an airborne transmitted disease (. . .) having that memory of what we went through (. . .) and then converting that to working in this sort of setting, it was really helpful (Residential Care Manager 2).
As guidance was difcult to implement in its current form, managers and staf created and used fow charts, posters, and videos to help colleagues, residents, clients, and family members. Tese included the following questions: How to clean a room after someone dies, how to put on PPE and clean one's car (for domiciliary care workers), how to wear a mask, and what to do (e.g., steps to take and whom to contact) if there is a positive test result.
Some managers and staf (n � 4) also described taking actions to manage families' expectations and keep communication going under new restrictions. Tese included discussing advance care and funeral plans with families, keeping regular communication with families via e-mail and online meetings, active social media presence, and updates, and dedicating time for online meetings with families and residents or clients.
Finally, we heard examples of initiatives taken to support the wellbeing and morale of colleagues and staf. Tese included practical support, such as organising laundry and transport services for staf and staf living on-site, cooking, and shopping for each other; wellbeing and mental health support, such as signposting to formal local mental health services for support with stress, anxiety, and burnout.
Most participants indicated that sharing best practices across the sector would be a helpful way to support organisations, managers, and staf throughout the pandemic. Two also suggested that the best practices developed during the pandemic should be incorporated in infection control training more generally.

Managing Expectations and the Impact of Conficting
Messages in Media. Tree managers and one care worker in residential care reported that having to manage expectations from families and to some extent staf and people receiving care have been an important part of their work. Tey indicated that this situation was created in large part by conficting and vague information in the news and media on the guidance for social care. For example, a care manager commented as follows: It is trying to manage the staf, the members and the families' expectations that when they hear the news, you're constantly having to say to them "Yes, I know they are saying that but it is not happening" and I have felt that we have had to do this all the way through (Residential Care Manager 4).
Interviewees reported that the overall guidance for the public was diferent and at times at odds with that for residential and domiciliary care, but this was not reported clearly enough in ofcial and news messaging. Tey also relayed some examples of conficting or unclear messages that were given to the public. For instance, from April 2020, visits to care homes from family and friends were either banned or heavily restricted (e.g., only possible if socially distanced). In July 2020, regular testing for care staf and residents was rolled out, but some restrictions on visits to care homes remained. However, the way this was reported in the news made families think they could now visit residents without socially distancing.

Improving Communication with Hospitals.
Tree participants discussed concerns relating to communication and coordination with hospitals. Te frst relates to the discharge of non-tested patients from hospitals to care homes contributing to outbreaks within care homes in the frst weeks of the pandemic, with one care-home manager indicating that a similar incidence still occurred in October, 2020.
A second issue concerns the discrepancies in the information provided about testing and isolation for discharged from hospital to care homes. Some participants indicated that hospital staf were telling patients who tested negative upon discharge, and that they would not need to self-isolate once discharged to a care homes. Tis was at odds with current infection control practices in care homes, resulting in distress, unmet expectations, and general confusion for the resident and their family. Participants also relayed concerns for the lack of clarity on whose responsibility was to liaise with other services (e.g., ambulance services).

Problems
Early in the Pandemic. Nine participants discussed specifc difculties that arose in the early COVID-19 phase and early in the frst national lockdown (March-May 2020). Tese confrmed the fndings from previous research [3,4,[6][7][8][9]20] and included limited supply of PPE, sparse testing, NHS test and trace contact tracing service that are not working, and difculties in getting advice on infection control.

Discussion
Te experiences of domiciliary and residential care workers and managers around the implementation of COVID-19 infection control guidance in this study can be summarised as follows: (a) frustration with the marginalisation, neglect, and negative image of the sector in policy response and public discourse; (b) concerns over the dissemination and quality of infection control policy; and (c) the sector's own proactive agency in developing best practice and adapting inadequate or missing guidance.
Te fndings reveal the widespread feelings of marginalisation and lack of support perceived by care workers and managers working during the frst wave of the pandemic. Similar fndings are emerging in recent studies exploring the experiences of social-care workers during the pandemic [9] and have strongly resonated with sector stakeholders with whom we have discussed the results of this study. Longstanding challenges and shortcomings of the COVID-19 policy response entangle in producing adult social care as a marginalised sector during this pandemic. As many have argued, the adult social care sector has been facing systemic problems for decades, and these have been exacerbated by the pandemic [24,38,39]. Most of these problems afect the workforce directly and include, along with market fragility and lack of adequate funding, low wages and underpaid staf, increased casualization, high turnover and vacancy rates, and unclear career development and pathways, especially for those in direct-care jobs [18,39].
Policy responses to emergencies, including economic measures that support the workforce both in the short and long term [40], benefts and incentives to work during an emergency, and clear, consistent, and timely communications [41], are important tools that governments have to recognize the essential work of social-care workers. Te experiences relayed by the interviewees show important shortcomings on each of these levels and reveal how the lack of work incentives and benefts for care staf, inadequate dissemination and quality of infection control guidance, and inconsistent communication between social-care providers, NHS settings and the public translated into added work and pressure for managers and care workers.
Another issue that strongly emerges from this study is the lack of understanding of the settings within which COVID-19 guidance would be implemented due to an overreliance on a clinical model of infection prevention and control. Tis has important and immediate consequences on the safety and working conditions of domiciliary and residential care staf, while also contributing to deepening feelings of neglect and marginalisation. By a clinical model of infection prevention and control, we mean an approach similar to that of the National Health Service (NHS) hospital settings, characterised by clearly defned roles, responsibilities, and structures that overlook the implementation and evaluation of guidance throughout the whole system, and where routines and procedures are consistent across settings.
Adult social care is characterised by fragmentation and a sparser workforce than the NHS [15], with multiple channels and sources of policy dissemination, authority, and accountability, including local authorities, independent care providers, residential and domiciliary care managers, and independent employers. On the contrary, within the NHS there is a clear structure to streamline and comply with infection control policies. Unlike hospitals or other secondary care settings, the social care settings within which the policy is implemented also vary enormously in size and infrastructure, whether residential or where is provided in people's own homes. Tey range from small family or locally run businesses through to those run by a large national company [18]. Tis makes a "one-size-fts-all" approach to infection control highly impracticable and unviable in residential and domiciliary care settings. For example, guidance on donning and dofng PPE should use simple and straightforward messages, taking into account that the setting may be domestic and waste disposal may need to be adapted compared to a customised hospital ward or operating theatre.
Moreover, it is important to consider the social and relational aspects of care labour [42] and care settings [14,43]. Adult social care is labour intensive and characterised by physical proximity between care workers and the people who receive care. COVID-19 measures like social distancing prove difcult or impossible to implement in these settings, with important repercussions on care staf's working life and fears around infection. Care homes as well as people's homes are social and domestic spaces where seemingly straightforward infection control measures like wearing PPE can be seen "antithetical to notions of a homely environment" [14], p. 9.
Drawing on recent works on health and safety [44], particularly on ethnographic works around "tacit knowledge" [45], "workarounds" [46], and the spatial and material aspects of infection control [47], we suggest that these could be used to rethink infection control guidance for residential and domiciliary care. To cope with the absence or inapplicability of guidance, care staf drew on multiple sources of knowledge of infection control, including previous experiences of viral outbreaks and guidance produced in diferent settings, and adapted it to the specifc needs and logistics of domiciliary and residential care settings. Tese endeavours could be used to reframe understandings of infection control in nonclinical settings such as domiciliary and residential care more broadly. Tey represent "necessary and practical adaptations" of general guidance "based on the context of the task and the worker" ( [46], p. 108) and a testimony to the importance of tacit knowledge, that is, the cognitive and "practical skills required for the performance of a task" ( [44], p. 80), for the implementation of infection control in variable settings such as care homes and home care.

Recommendations.
Infection control in social care is complex and these fndings suggest that guidance would greatly beneft from a better and deeper understanding of residential and domiciliary care settings and how they difer from clinical settings, such as hospitals. It is worth noticing that prior to COVID-19, government guidance on infection control for care homes was last updated in 2013.
We recommend that policymakers explore more fexible approaches to infection prevention and control to attend to the variability of care settings and activities. Best practices and responses produced during the COVID-19 crisis by care staf would be a productive starting point for this exploration and to value the expertise of the adult social-care workforce.
Coproduction [48,49] with adult social-care stakeholders, including those in direct-care roles, would help to ensure that guidance is relevant and applicable in domiciliary and residential care settings. Along with a better understanding of the organisational and practical aspects of infection control in care settings, issues around working conditions (e.g., low pay, precarious contracts, and devaluation of care work) and feelings of marginalisation need to be addressed as a matter of priority.
We also recommend that policymakers should consider domiciliary and residential care as separate settings and diferent groups of workers. Infection control guidance for domiciliary care should be designed around the specifc needs, procedures, activities, and relations of this service. It should also be a priority as much as that produced for NHS and care homes.
Finally, we suggest that further research is needed to explore infection prevention and control in care settings and improve guidance and outcomes for the adult social care sector, service users, and policymakers. Domiciliary care, in particular, remains underrepresented both in research and policy.

4.2.
Limitations. Te small sample is a clear limitation of this study. Tis was mainly due to a challenging and slow recruitment process infuenced by three main factors: (a) complex sponsorship and ethical approval due to the need of adapting data governance to online data collection and recruitment procedures to changing conditions while the pandemic progressed; (b) working with a population heavily impacted by the pandemic; and (c) the beginning of the second wave of the pandemic during data collection.
We are also aware that there is an over-representation of the experiences of managers in residential care (n � 5) compared to those of care workers in domiciliary (n � 3) and residential (n � 1) care and managers in domiciliary care (n � 1). Tis is most likely due to the channels used for recruitment that brought care providers and managers to have the role of gatekeepers.

Conclusion
We conducted semistructured interviews with domiciliary and residential care staf to understand their experiences of implementing infection control guidance during the frst wave of the COVID-19 pandemic. Te following fndings were revealed: (a) the inadequacy and inapplicability of guidance produced for these settings; (b) widespread concerns for the marginalisation of the sector in policy response and public discourse; and (c) care staf's resourcefulness and proactivity in developing best practices to provide care safely and overcome challenges due to inadequate or missing guidance.
Infection prevention and control in adult social care is complex and paramount to efectively respond to COVID-19 and future infections. Policymakers need to understand and address the practicalities of care work and care settings to produce applicable guidance. Tis might involve moving away from prescriptive and clinical models of infection control, in favour of more fexible approaches that better address the variable and at times unpredictable contexts within which care is provided (e.g., people's own homes), the varied and multiple activities and procedures carried out, and the interpersonal and intimate qualities of care. Tis shift calls for policymakers to prioritise adult social care alongside the NHS and to recognize and value care staf's expertise.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request.
(equal) and reviewed and edited the manuscript. Jo Middleton supervised the work and reviewed and edited the manuscript. Rebecca Sharp was involved in dissemination and implementation (lead) and reviewed and edited the manuscript. Wendy Wood was the project administrator (lead). Daniel Roland contributed to methodology (supporting) and reviewed and edited the manuscript. Julien Forder was involved in funding acquisition. Jackie Cassell contributed in conceptualization, funding acquisition, supervised the study, and reviewed and edited the manuscript.