Abuse and Wellbeing of Long-Term Care Workers in the COVID-19 Era: Evidence from the UK

: The UK long-term care workforce has endured difficult working conditions for many years. During the pandemic, the sector faced unprecedented challenges, which further exacerbated these conditions and brought concerns about workplace abuse and violence. Such experiences can vary by personal and work characteristics, particularly affecting minority ethnic groups. They can subsequently impact workers’ wellbeing and the sector overall. Drawing on the first wave of a UK longitudinal workforce survey, this article examined the impact of COVID-19 on social care workers’ working conditions, general health and wellbeing, and intentions to leave the employer and sector altogether. The analysis is based on both quantitative and qualitative responses 1037 valid responses received between April and June 2021. The respondents were predominantly female, working in direct care roles and mainly serving older adults (including those with dementia). The findings highlighted worrying experiences of abuse in relation to COVID-19, which differed significantly by nationality, ethnicity and care settings. The analysis further showcased the negative impact of experienced abuse on work-life balance and intentions to leave the current employer or the care sector altogether. The findings emphasise the need for targeted measures that promote workers’ physical


Introduction
The COVID-19 pandemic created significant challenges for the UK's long-term care (LTC) workforce. Government policy prioritised the protection of the National Health Service (NHS), while the risk for the LTC sector was largely unrecognised [1,2]. It is estimated that around 25,000 older people were discharged from hospitals to LTC settings to 'free up capacity' without testing or quarantine and the provision of adequate personal protective equipment (PPE) for staff in the first months of the pandemic [3] (p. 11). As a result, COVID-19 rapidly spread in care settings, leading to a significant rise in deaths and excess mortality, especially in care homes [4]. Meanwhile, the Government's response was perceived as slow and indecisive [5], managers had to keep up with continuously changing guidance [6], and care workers had to take on tasks usually undertaken by registered nurses [7,8]. The increased volume of tasks during the pandemic resulted in higher workloads among existing staff [9]. In addition, lack of adequate PPE, staff shortages exacerbated by sickness-related absences, and a workforce with a high proportion of lowpaid and precarious workers [10] created a 'perfect storm' for adverse working conditions are particularly at risk of workplace violence and abuse [24]. For minority workers, abuse incidents can also often be racially motivated [29,32]. Race and ethnicity can influence the nature and frequency of physical and verbal abuse against LTC workers, although further research on these groups is needed [33][34][35]. In the context of COVID- 19, there have been concerns about mistreatment of LTC staff in the form of abuse by service users, their families, and the public, likely triggered by restrictions such as limited visitations in care homes, increased workload, perceived lack of safety measures with potential impact on care practices, and inability to travel [36]. Evidence from qualitative interviews with LTC stakeholders in the UK emphasised the growing concerns about the abuse of workers during the pandemic calling for further research on this topic [37].
While it is crucial to examine the prevalence and nature of abuse experienced in LTC in relation to COVID-19, it is equally important to consider the consequences of such experiences for workers' wellbeing and the sector more broadly to sustain a dynamic and resilient workforce. Abuse of care workers negatively impacts their physical and mental wellbeing, reduces job satisfaction, increases burnout, absenteeism and turnover, and reduces retention at the organisational level [19,[38][39][40][41][42][43]. Such experiences ultimately impact care quality, safety and efficiency [18,[44][45][46][47].
This study aims to examine the impact of the first wave of the COVID-19 pandemic on LTC workers' working conditions, general health and wellbeing. It investigates the differential experiences of workplace violence by nationality, ethnicity and care settings. The study further explores the association between the experience of abuse and mistreatment with workers' wellbeing, and their intention to quit their current employer or the care sector.

Survey Design
The data used in this article stem from the first wave of a longitudinal LTC workforce survey conducted as part of the Retention and Sustainability of Social Care Workforce (RESSCW) project. The project team designed the survey in consultation with the steering group and the funder of the project (The Health Foundation). The longitudinal survey design was informed by the findings from an earlier 'pulse survey' conducted in the summer of 2020, followed by qualitative interviews with stakeholders as part of the same study [12,37]. The longitudinal survey was designed to be completed online. It included several questions taken from validated and widely used workforce surveys (e.g., Workplace Employment Relations Survey, World Health Organisation: Workplace in the Health Sector), adapted where appropriate to social care. It also included open-ended freetext questions. The survey was piloted with three LTC workers to ensure clarity in early April 2021 and was then revised with changes in wording, response options and order of some questions to address the feedback received.

Recruitment and Data Collection
The lack of a care workers' registry and the fact that many might be working in several settings reduced the effectiveness and feasibility of targeting worker groups within specific care settings. Instead, to maximise survey coverage, the project team opted for a more inclusive recruitment strategy. First, the survey details were shared with the project's steering group members (who represented large care workers and providers representatives), who distributed them to the relevant contacts. Additional recruitment channels involved formal and informal groups of care workers through social media and newsletters and individuals or organisations who had previously engaged in research, interested in the LTC workforce in the UK. Qualtrics was used to implement the survey optimised for mobile devices. Ethical approval was gained from the School of Social Policy, Sociology and Social Research SRC Ethics Committee (SRCEA ID 240) at the University of Kent. Participation in the survey was voluntary and anonymous, and participants could withdraw at any time. Participants were asked to read a brief study information sheet about the benefits and potential risks of taking part, and the data protection and privacy notice before proceeding to the questions. A prize draw was offered (online retail vouchers) to incentivise uptake. Respondents could opt-in to the prize draw by leaving their contact details. All personally identifiable information was removed from the dataset before analysis. The survey was open from 13 April to 28 June 2021.

Measures
The survey included multiple-choice questions with pre-recorded and free text options to best capture the workers' views and characteristics. The questions covered basic demographic information, working conditions, support, general health, wellbeing, intention to leave employer or sector, experience during COVID-19, such as mistreatment and abuse, and COVID-19-specific topics (including vaccination, infection of staff and clients, and isolation). The nature and incidents of abuse and mistreatment in relation to COVID-19 were captured by responses to several questions, including (a) type of abuse (verbal abuse, bullying, threat, physical violence-for a definition of the different types as presented to the respondents, see Appendix A, Table A1); (b) perpetrator(s) (manager/supervisor, colleague/staff member, service user/client, service user's/client's family, general public) if indicated experiencing at least one type of abuse; and (c) subsequent action(s) taken by the respondent (took no action, reported it to a manager/supervisor, told a colleague/staff member, sought help from a union). A dummy indicator of any abuse incident was generated if responded positively to at least one type of abuse. A severity measure (for abuse) was created based on the number of positive responses to any abuse type (i.e., single or multiple). In addition, respondents were asked to indicate how much of the time in the past few weeks their job has made them feel (a) tense, uneasy or worried; (b) calm, contended or relaxed; (c) depressed, gloomy or miserable; and (d) cheerful, enthusiastic or optimistic, with five response options ('All of the time', 'Most of the time', 'Some of the time', 'Occasionally', 'Never'). A composite index based on responses to these feelings questions was created as a proxy wellbeing measure. The second wellbeing indicator used in this article reflected responses on a 5-point Likert-type work-life balance satisfaction scale ('Very satisfied', 'Satisfied', 'Neither satisfied nor dissatisfied', 'Dissatisfied', 'Very dissatisfied'). Intentions to leave the current employer or the sector altogether in the next 12 months were captured by responses on 4-point Likert type scales ('Very likely', 'Quite likely', 'Not very likely', Not at all likely'). Open-ended free text questions where participants could add more information or explain their answer choices (for example, about abuse and wellbeing support received) were also included.

Analysis
Data from the survey was downloaded (in Excel) and split into a qualitative and a quantitative dataset. Qualitative data-free text responses to open-ended questions-was transferred to NVivo version 12 and analysed thematically [48]. All quantitative data were analysed in Stata SE 15.1 [49]. Several steps, including identifying hard and soft red flags, were undertaken during the data cleaning process to ensure only valid responses were included in the analysis (for more details, see Appendix A). Missing data was minimal (up to 2%) and was imputed with the mean to retain the total sample [50]. We examined the demographic characteristics of the overall sample and identified differences among specific population groups. We focused on differences by nationality, ethnicity, and care setting, as these have been found to be important in previous research. Following [12], we combined responses to nationality and ethnicity questions and created three subgroups: White British, White Non-British, Black, Asian and Minority Ethnic (BAME). For care settings, we distinguished between residential care (with or without nursing) and domiciliary care or other to capture groups with very few respondents (including, for example, daycare centres and supported living/extra care housing). Independent t-tests were used to ascertain differences between the different subgroups. Finally, regression analyses (i.e., ordinary least squares and probit) were performed to identify how the experience of abuse (any; type; severity) impacted workers' wellbeing and intention to leave, using a set of controls (i.e., age, gender, ethnicity and nationality, employer type, care setting, client group, job role, tenure, contract type, union membership, regional COVID-19 cases and deaths, north-south dummies), most of which have been used in the literature to identify such relationships (see, for example, [19,51]). All descriptive statistics are reported as proportions or percentages, and the lower statistical significance level was set at 0.01.

Survey Respondents
The survey received 1037 valid responses from staff working in the LTC sector across the UK. The demographic and work-related characteristics of survey respondents are summarised in Table 1. Like the overall profile of the LTC workforce in the UK, most respondents were female (82.2%), and almost a third (29.0%) were aged 45-54 years old, which is consistent with the national picture [9]. Over two-thirds (83.5%) were White British, and a tenth (11.1%) were from a BAME background. Nationally, BAME workers constitute 21% of the adult LTC workforce, suggesting that this group was under-represented in the survey. However, difficulties with engaging participants from this population group in research have been highlighted elsewhere [21]. Survey respondents had, on average, at least six years of experience in the LTC sector, which is broadly consistent with the national average [9]. Nearly three-quarters of all respondents were employed on permanent or temporary contracts with guaranteed hours (70.0%), and more than half worked in the private sector (53.7%) or were currently/had been in the past a member of a trade union or staff association (57.8%). Most workers were in direct care roles (e.g., care worker, support worker, care assistant), serving mainly older adults, including those with dementia (58.4%). In terms of care settings, there were slightly more respondents working in domiciliary than residential care (38.3% and 36.2%, respectively), which is similar to what is observed nationally [9]. At the time of the survey, less than a quarter (22.9%) reported fair or poor general health.

Self-Isolation, Working Hours and Pay
Thinking back to the beginning of 2021, 330 respondents (31.8%) indicated increased workload without additional pay, while 212 (20.4%) had increased their paid working hours (for more information, see Appendix B, Table A2). This was also reflected in the free text responses, emphasising the lack of staff with suggestions for improvement.  Table 2 shows that a quarter of all respondents (25.6%) reported having experienced some form of abuse in relation to the pandemic. Among all abuse types, verbal abuse and bullying were most mentioned (19.8% and 11.4%, respectively), followed by threat (8.0%) and physical violence (5.1%). The free text responses expanded on some negative experiences, with social media portrayed as exacerbating these experiences. While about half (46.8%) of those verbally abused reported it to a manager or supervisor, over a quarter (26.3%) took no action. For bullying, nearly a third (29.7%) said it to a manager or supervisor, less than a third (27.1%) told a colleague/staff member and, interestingly, over a tenth (15.3%) sought help from a union. Notably, it was not safe to report the incident in some settings, but there were also cases where it was said to management with no further action.

Abuse Differentials by Individual and Organisational Characteristics
White British were the least likely to have experienced any type of abuse about COVID-19 compared to White Non-British and those from a BAME background. Specifically, BAME respondents were twice as likely to have been abused than the White British (40.0% vs. 23.4%, respectively), and the difference is statistically significant. All remaining comparisons by race and ethnicity were not statistically significant (see Table 3 for more details). Regarding care settings, respondents working in residential care were significantly more likely (31.5%) to have reported at least one type of abuse than those working in domiciliary care or other settings (22.2%).  Care staff also highlighted the structural stigma associated with this work compared to working in the NHS. For example, NHS workers received discounts and queue jumps at the beginning of the pandemic before social care workers were acknowledged. This was mentioned frequently by the respondents.

Implications of Abuse/Mistreatment on Individual Workers and the Organisation
Over a third (39.0%) of respondents reported that their job in the past few weeks made them feel tense, uneasy, or worried, while almost half (44.1%) occasionally or never felt depressed, gloomy or miserable. Nevertheless, less than a third (27.6%) were dissatisfied or very dissatisfied with their work-life balance overall (for more information, see Appendix B, Table A3). The frequent negative feelings could be related to the nature of the role itself and the extra challenges created by the pandemic, as mentioned in some quotes.
'It has been a hard year and fear to the job working all the way through the COVID  Looking at the relationship between experiencing abuse and wellbeing, the findings show that, irrespective of the measure used (i.e., proxy or work-life balance; different abuse indicators), abuse negatively impacts workers' wellbeing (see Table 4). More severe (i.e., multiple types) abuse has a more adverse negative effect on wellbeing. When focusing on the impact of each abuse type, while the relationship remains negative and strongly significant for all types except for threat, the adversity of the effect differs by the wellbeing measure used. Specifically, bullying impacts the most on the proxy wellbeing measure, whereas physical violence affects work-life balance (the full estimation results are included in Appendix B, Table A4).  Note. Robust standard error in parentheses. All models include controls for personal and work characteristics. The omitted group is no abuse in all cases. *** p < 0.01, ** p < 0.05, * p < 0.1, ns not significant.
A staggering 43.3% reported that they were likely to leave their current employer in the next 12 months (see Appendix B, Table A5 for more details), which could have been exacerbated by the experienced abuse. Indeed, according to the estimation findings reported in Table 5, the experience of at least one type of abuse increases the probability of wanting to leave, with multiple types of abuse having a more substantial negative effect than a single abuse type. Among the different abuse types, the picture is mixed; verbal abuse influences the intention to leave the sector, whereas bullying has the strongest influence on the intention to leave the current employer (the full estimation results are included in Appendix B, Table A6). Note. Marginal effects at mean. Z-scores based on robust standard errors in parentheses. All models include controls for personal and work characteristics. The omitted group is no abuse in all cases. *** p < 0.01, * p < 0.1, ns not significant.

Discussion
The LTC sector in the UK is facing long-standing challenges centred around chronic underfunding, workforce shortages, poor pay and the low status of the care workforce. The ongoing challenges in the sector were exacerbated by the COVID-19 pandemic, with severe implications on the workforce. The wellbeing of care workers was put to the test, with minority ethnic workers more at risk of challenging experiences [29]. For many, the workload and responsibilities increased directly due to the pandemic. LTC workers, while usually short-staffed, were required to cover for significant increases in absenteeism while managing infection control measures to contain the pandemic. The latter measures were not always easy to implement. Examples include contradicting guidance in the first wave of the pandemic around safety measures, such as social distancing, isolation and limiting visitations by family members in care homes [52]. Since the onset of COVID-19, care workers were often perceived by family members as the ones not allowing visitations, further exposing staff to potential abuse and bullying incidents. The additional workload associated with limited guidance led to higher levels of stress and staff burnout. The uncertainty brought in by the pandemic could have only intensified the effects of abuse and mistreatment incidents. The lack of support mechanisms to mitigate these negative behaviours adversely impacted care workers' health and physical and mental wellbeing. This study confirmed that feelings of neglect, depression and stress combined with increased workload featured during the pandemic.
This study focused on the impact of the COVID-19 pandemic on LTC workers' working conditions, general health and wellbeing. Using data from the first wave of a UK longitudinal LTC workforce survey, it was found that the workload of staff increased substantially during the pandemic. In most cases without extra pay. While the increased workload could be, for example, due to a lack of staff and resources, or additional tasks created by the pandemic, it raises concerns about its sustainability and the long-term impact on workers' health and wellbeing. It is essential to highlight that despite organisational and governmental processes in place, some workers who took sick leave, self-isolated or were stopped from working as a measure to control the infection reported that they did not receive any pay, adding to the financial challenges already faced by low-paid staff. Crucially, receiving no pay while sick, in a sector characterised by very low wages, is likely to negatively impact all wellbeing aspects of the LTC workers, especially their financial wellbeing [29].
Furthermore, our study found that a quarter of respondents experienced some form of abuse in relation to the pandemic. While the overall abuse prevalence is lower than reported in other studies (see, for example, [15,16,35]), the context is different given the focus on the pandemic. Nevertheless, the dominance of verbal abuse observed in our study was also present in the pre-COVID era (see, for example, [19,53]). Our study findings showed a higher prevalence of abuse incidents among minority ethnic workers, perhaps explained by the lack of empowerment and deeply entrenched negative attitudes, including structural racism, towards minority workers [29,54]. As in other studies [19,21], we found significant differences in abuse experiences between care settings, with such incidents more common in residential care. This could be explained by the nature and needs of clients served in each setting, the job demands and stressors, and specific restrictions pertinent to the pandemic. The structural stigma of social care work, particularly in the context of the COVID-19 pandemic, was also likely to exasperate these feelings [55].
The negative experiences of abuse identified in this study stress the importance of considering the broader implications for workers' health and wellbeing, the organisation and the sector. For instance, our results showed a negative association between abuse and work-life balance irrespective of abuse type. Consistent with similar studies, we found that many LTC workers frequently felt tense, uneasy, depressed and gloomy because of their job, with further implications on their general health and work-life balance [55,56]. In addition, workers' intention to quit the sector was found to be about half of that quitting the current employer voluntarily in the next year. The direction of these findings showcases the strong motivations of the care workforce for joining the sector and the hope that a change of employer will result in better working conditions. Despite all the challenging experiences faced before and during the pandemic, the workforce remains highly committed to the sector.
This study is not without limitations. First, the study sample was self-selective, focusing only on one survey mode (i.e., online). This potentially prevented some groups of workers from participating, including live-in carers and personal assistants, where we observed a handful of respondents. Furthermore, our findings-especially from the estimations-reflect associations and cannot be interpreted as causal relationships. Future analysis, using longitudinal data on abuse, wellbeing and working conditions, which attempts to establish causal relationships between these measures, could be a natural extension of this study. As too would be a further investigation into the nature of changes to working conditions, including a focus on the intention to quit, during the challenging, uncertain and multi-faceted COVID-19 era.

Conclusions
In an era of continued challenges, including austerity, a new post-Brexit immigration system and changes in practices to allow for a more flexible work model following the global COVID-19 pandemic, there is a call for targeted interventions to focus on more supportive jobs that will improve workers' wellbeing at work [29]. It is not just about maintaining adequate standards and building relationships with the care workforce but also emphasising all aspects of wellbeing-physical, mental and financial-that can pave the way forward.
This study highlights issues of considerable policy importance. The first is acknowledging the implications of COVID-19, including the increased levels of abuse and mistreatment, on the LTC workers' wellbeing and ensuring adequate interventions to mitigate them in place. This is of high policy relevance to ensure a sustainable workforce. However, a recent government inquiry report [57,58] assessed the government's response to the health and social care workforce wellbeing as inadequate. It highlighted the differentiated experiences across different groups of workers and between health and social care settings in England. This lack of recognition of the impact of COVID-19 on the LTC workforce in the UK was reflected in minimal support measures during the pandemic compared to other countries such as Australia and Canada [2]. The second policy implication is the linkage between the exposure to mistreatment and abuse and the intention to quit. With current workforce recruitment challenges and high turnover rates, it is crucial to implement interventions aimed at reducing incidences of bullying and abuse to minimise the haemorrhage of talents and wasted training and skills.

Data Availability Statement:
The data used in this study are managed by the authors and cannot be accessed by others outside of the project research team.

Acknowledgments:
We would like to thank all participants for taking the time to complete the online survey. We also thank the RESSCW research team for valuable feedback on the development of the survey. The views expressed as well as any errors are entirely of the authors.

Conflicts of interest:
The authors declare no conflict of interest.

Appendix A. Data Quality Checks and Exclusions
The survey was completed by 1530 respondents. Several fields were used separately or in combination to identify if the responses were valid. These included: (a) IP address, (b) survey completion time, (c) phone number (provided for prize draw), (d) (first half of) work postcode, (e) longitude and latitude (where available), (f) email and name (provided for prize draw), (g) free text responses. The following exclusions were made: duplicate formatting of contact details and/or unrecognisable free text (n = 19), invalid email address or name (n = 30), non-UK longitude and latitude (n = 16), invalid work postcode and/or phone number (n = 28), completion time under five minutes and/or invalid name/phone number (n = 339), all aforementioned fields invalid (n = 61), the latter including n = 17 who did not consent. In total, 493 responses were deemed invalid and excluded from the analysis.

Verbal abuse
Behaviour that humiliates, degrades or otherwise indicates a lack of respect for the dignity and worth of an individual.

Bullying
Repeated and over time offensive behaviour through vindictive, cruel, or malicious attempts to humiliate or undermine an individual or groups of workers.

Threat
Promised use of physical force or power resulting in fear of harm or other negative consequences to the targeted individuals or groups. Physical violence (attack/assault) The use of physical force against another person or group, that results in harm.
Note. Adapted from World Health Organisation: Workplace Violence in the Health Sector.