Working Conditions and Well-Being across the COVID Pandemic in UK Social (Care) Workers

Prior to the COVID-19 pandemic, social work and social care practitioners had some the worst working conditions of any sector in the UK. During the pandemic, data revealed that social care occupations had higher COVID infection and mortality rates than the general population. The article reports the changing working conditions (measured via the Work-Related Quality of Life scale) and well-being (measured via the Short Warwich– Edinburgh Mental Well-being Scale) of UK social care and social workers across three timepoints between May 2020 and July 2021 through an online cross-sectional survey of working conditions and well-being. Analysis of variance demonstrated that both well-being and working conditions were significantly poorer in July 2021 (phase 3 [ n ¼ 1,606]) than the previous two phases ( n ¼ 2,523 and n ¼ 2,424, respectively), suggesting that both working conditions and well-being worsened within the social care and social work workforce across the pandemic. Furthermore, each of career satisfaction, working conditions, control, general well-being and home–work interface predicted poorer well-being


Introduction
In the first four months of 2022 through to 14 April, care homes in England alone registered 2,372 deaths in which Covid-19 was at least a contributory factor. Between the 10 April 2020 and 14 April 2022, there were 33,096 covid-related deaths in care homes across the country (Office for National Statistics, 2022), illustrative of higher mortality rates than that within the general public. However, whilst much attention has been paid to the impact of the pandemic on healthcare services, there has been less coverage of the impact on those working within the care sector. Both the social work and social care sectors were increasingly affected across the first two and a half years of the pandemic (Comas-Herrera et al., 2020). UK-wide data revealed that those employed in social care occupations had a significantly higher COVID-related death rate than the general population, even when taking into account age and sex (Office for National Statistics, 2021). Similar disproportionate mortality rates have been found internationally (World Health Organization, 2020), all demonstrating the difficult and, at times, dangerous jobs of social care workers and social workers during the pandemic. This increased likelihood of morbidity and mortality, combined with continually poor working conditions, may have ongoing impacts on the psychological health and well-being of these key workers.
Social care in the UK is an umbrella term which describes the organisational sector that provides care and support to children, their families and adults. Social care has distinct differences to healthcare, for example, with the UK National Health Service (NHS) providing clinical care which is free at the point of use for UK residents, whereas social care has country-specific rules on service provision, access and funding. Within this article, we focus on social care in two separate but related roles: social care workers and social workers. The social care workforce is large, consisting of some 1.5 million employees (Kingsfund, 2021). There are numerous roles within this sector, although the majority work in care homes, domiciliary/home care and/or provide day services for adults. All social care workers are skilled but most not professionally qualified. Also coming under the umbrella of social care are social workers. These are professionally (degree) qualified and provide a largely statutory role and mainly (but not exclusively) employed by local government in all parts of the UK apart from Northern Ireland, where they are most likely to be employed by integrated Health and Social Care Trusts. In the UK, social work is a title protected by law. In all parts of the UK, social care is also provided by private, community and voluntary sectors. Whilst social care workers and social workers often work with similar populations, their roles and functions are distinct. The main differences are the statutory function of social work, including service provision, professional assessment and safeguarding duties. Whereas social care is associated with 'social' needs and personal care. In the UK, social work is a title protected by law and is a regulated profession with legal oversight from professional bodies such as Northern Ireland Social Care Council, Scottish Social Services Council, Social Care Wales and Social Work England. In Northern Ireland, Scotland and Wales, social care is also regulated by the respective social work regulators, whereas in England social care is currently unregulated.
Chronic stress is defined as a negative psychological state which lasts over an extended period of time (Ravalier, 2019). There is an increasing wealth of evidence which demonstrates that the working conditions and chronic workplace stress can have a negative impact on the psychological and physical health of employees. For example, a large meta-analysis by Niedhammer et al. (2020) found that low control at work combined with poor social support were associated with cardiovascular mortality. Rosengren et al. (2004) also reported similar outcomes when looking at the impact of psycho-social risk factors on cardiovascular disease in nearly 25,000 people across fifty-two countries, and a prospective cohort study of over 10,300 participants by Chandola et al. (2006) found that chronically poor working conditions were related to the development of metabolic syndrome, a risk factor for conditions such as Type 2 diabetes. Furthermore, in social work, Ravalier (2019) and Ravalier et al. (2021b) found that numerous psycho-social working conditions (such as demands, control, managerial support, role understanding and the way in which change is communicated) impact stress-related outcomes such as psychological well-being and turnover intentions.
These individual impacts also have knock-on effects on employing organisations. For example, stress and common mental illnesses such as depression and anxiety are the primary cause for long-term sickness absence (that which lasts four weeks or more) in the UK, responsible for 54 per cent of all health-related sickness absences in 2018/19 (Health and Safety Executive [HSE], 2020). Furthermore, chronic workplace stress and poor working conditions have implications for patients and service users in front-line Health and Social Care occupations. For example, West and Dawson (2012) demonstrated that UK NHS Trusts which had the highest levels of employee engagement were associated with better patient outcomes such as satisfaction and mortality, as well as reduced turnover and sickness absence. Similarly, in social work, there is dissatisfaction with the service, gleaned from a lack of time given to service recipients (Gaskell, 2010). Scanlan and Still (2019) confirm the relationship between burnout and intention to leave due to high demands for mental health practitioners. Whilst there are organisational and individual factors at play, the impact is felt at service user levels which is unacceptable (McFadden et al., 2015).
The UK Health and Safety Executive (Cousins, 2004) define working conditions are work-related factors which, if left in a chronically poor state, can lead to poorer health and organisational outcomes. Examples of working conditions include demands, support and control. Health and social care workers in the UK have amongst the worst working conditions (Ravalier, 2019;Ravalier et al., 2022) and highest levels of stressrelated sickness absence (HSE, 2021) of any occupational sector. This is typified by high levels of demands (quantitative and qualitative workload), low levels of support and low levels of autonomy (Ravalier, 2019). Across the first year of the COVID-19 pandemic in the UK (April 2020 onwards), social workers moved to primarily working from home with some still conducting in-person visits to service users deemed to be higher risk in terms of safeguarding concerns. The term 'glocal' (local impact of global issue) became used to describe the same concerns in different global contexts (e.g. see Harrikari et al., 2021). As everywhere, there was limited personal protective equipment and support from both management and government more widely. This resulted in working conditions becoming increasingly poor during this time . These working conditions mean social care workers may be more likely to want to leave the job, have poor levels of job satisfaction and high levels of presenteeism (Ravalier, 2019). Consequently, in addition to impacting upon individual workers and their employers, there will also be an impact upon outcomes for service users. For example, Flower et al., (2005) demonstrated that children who had multiple social workers, as opposed to a consistent social work figure, were 60 per cent less likely to be found a permanent placement, which has long-term life consequences for those affected.
The job demands-resources (JDR) model of workplace stress (Demerouti et al., 2001) suggests that a dispropionate level of demands at work, which are not counterbalanced with appropriate resources, can lead to employee stress and burnout (Taris and Schaufeli, 2015). Demands are conditions under which employees work which add to the physical or psychological load experienced, such as qualitative and quantitative workload. Resources are conditions which detract from the potentially deleterious effects of demands, and may include improved developmental opportunities and managerial/peer support (Broetje et al., 2020). The model therefore suggests that a chronic outmatching of demands to resources can lead to sickness absence, dissatisfaction and burnout (Bakker and de Vries, 2021). Whilst the model has been critique for being too broad and simplistic, this is also one of its key strengths-it allows the inclusion and encorporation of a wide range of working conditions which may be missed within other models such as the job demands-controlsupport theory. The model also therefore provides a good platform for which to conceptualise the findings of the presented project. As demonstrated above, these negative outcomes not only impact on individual social care workers but are likely to have additional impacts on multiple levels, for example, for the organisations for whom they work, efficacy of services and (as previously discussed) directly impact on the clients/service users whom they work with (Scanlan and Still, 2019).
The aim of this article therefore is to investigate how working conditions may have changed across the COVID-19 pandemic in UK social care workers and social workers, and the influence that this may have had on their psychological well-being.

Methods and participants
This article presents data from a three-wave non-probability crosssectional online study collected across approximately eighteen months to look at the changes in working conditions and well-being in UK social work and social care practitioners through the Covid-19 pandemic. Data were collected between May and July 2020 (Phase 1), November 2020 and January 2021 (Phase 2) and May 2021 to July 2021 (Phase 3) as part of the wider 'Health and social care workers' quality of working life and coping whilst working during the Covid-19 pandemic' project. This article will present the outcomes for social care staff and social workers across the UK. Recruitment of participants was via an opportunity and snowball sampling, promoted via social media, emails to contacts and newsletters. The study was conducted according to the guidelines of the Declaration of Helsinki and approved by Filter Ethics Committee in the School of Nursing at Ulster University.
Demographic questions were split across those specific to the individual (age, gender, disability and ethnicity), with Table 1 demonstrating the most frequent responses for each of these categories across all three phases of the project (Table 1). Across the three phases profiles are relatively homogeneous, with the majority of respondents being White, female and without a disability. This is similar to the profile of UK social workers and social care workers who tend to be female (86 per cent and 82 per cent, respectively) and aged 40-49 years (Skills for Care, 2020; Gov.uk, 2021). However, our sample is over-represented by White respondents, with approximately 78 per cent of UK social workers (Gov.uk, 2021) and 82 per cent of social care workers (Skills for Care, 2020) having White ethnicity.
To collect occupational demographics, we asked about individual's role within social care and social work (i.e. whether they work with children, adults, people with a learning disability or older people's teams), number of years' experience in the role, whether holding a full-time, permanent contract and whether they were a carer outside of the workplace. These questions were not compulsory to answer and some did not respond to all demographic questioning. Table 2 shows the occupational demographics of respondents across the three phases of the project, outlining the area that they worked in, how long they had been in the role and whether they worked a full time, permanent role or other. Across all respondents, the majority were employed under full-time, permanent contracts and had between 11-and 20-years' experience in their role. Most respondents worked in either Children's or Older People's services, although social workers were more likely to work with children and social care workers with older people.

Materials
Psychological well-being was measured by the short Warwick-Edinburgh Mental Well-being Scale (SWEMWBS) and working conditions by the Work-Related Quality of Life (WRQoL) scale. The short version of the WEMWBS is a seven-item measure of psychological well-being, measures on a five-point Likert scale (from 1, none of the time, to 5, all of the time). Survey respondents were asked to state how often they had particular thoughts or experiences over the previous fortnight, and higher scores are indicative of better well-being. Example questions are 'I've been feeling optimistic about the future' and 'I've been dealing with problems well'. SWEMWBS scores were calculated by summating all seven item scores into a total score in the range of 7-35, where higher scores indicate better well-being. SWEMWBS has a mean of 23.5 and a standard deviation of 3.9 in UK general population samples (Ng Fat et al., 2016) with cut-off categories defined as low (7-19), medium (20-27) and high (28)(29)(30)(31)(32)(33)(34)(35).
Various studies have demonstrated that the shorter WEMWBS is inherently valid (e.g. Ng Fat et al., 2016) and reliable (e.g. Ng Fat et al., 2016, demonstrated Cronbach's alpha of 0.84 amongst the general population) and sensitive to changes in well-being across population over time. The WRQoL measure consists of twenty-three items split across six subscales, and assesses quality of working life and working conditions. The six sub-scales are: home-work interface, career satisfaction, stress at work, general well-being, working conditions and control at work. Total scores also indicate overall WRQoL. Scoring is also on a five-point Likert scale from 1 (strongly disagree) to 6 (strongly agree). Reliability measures for each subscale have been calculated between 0.75 and 0.88 and overall scoring of 0.91 demonstrated reliability (Easton and Van Laar, 2018). Higher total scoring indicates better WRQoL, with higher scoring on all sub-scales apart from stress at work also indicating better WRQoL. Low scores on stress at work indicate higher stress as these items are reverse scored.

Analytical strategy
Quantitative data were analysed using IBM SPSS 24.0 (IBM, Armonk, NY, USA). Demographics are presented through mean, standard deviation and most frequent responses. Mean and standard deviation scoring were determined for all respondents, separated by phase of response and separated by job role (i.e. social worker or social care worker). ANOVA was undertaken to look for differences in scoring across the three phases of response for all respondents, and separated by job role, with Bonferroni post-hoc analysis to compare across time on any significant findings in the ANOVA. Finally, three regression analyses were undertaken to investigate the impact of working conditions on well-being at phase 3 for all respondents as well as separated by job role. Table 3 presents the mean and standard deviation scoring on mental wellbeing and quality of working life for all respondents across phases 1-3, as well as separated by job role. Mean scoring on the WEMWBS across all respondents indicates highest reporting psychological well-being at phase 1, followed by phase 3, with those in social care following a similar trend. Social workers again demonstrated highest scoring at phase 1, with little difference between phases 2 and 3. However, when compared with UK-wide norm data, across each phase of the study scoring was much lower than that of the general population (Ng Fat et al., 2016). Similarly, WRQoL was higher (i.e. scored better) at phase 1, decreasing across time through to phase 3. The only exception to this across all respondents was in the control at work sub-factor. When looking at individual occupation responses, a similar trend was found for most factors. Total WRQoL, career satisfaction, general well-being, home-work interface and working conditions all decreased across the three study phases. However, work stress was slightly higher in phase 3 for social workers than at phase 2 (highest in phase 1), and similarly control at work amongst social care workers was highest at phase 1, reduced at phase 2 and then slightly improved at phase 3.

ANOVA
A series of one-way ANOVA analyses was conducted to see whether there are statistical differences in scoring on each WRQoL sub-factor and total WRQoL and WEMWBS scoring across the three phases of the project, with post-hoc Bonferroni testing to compare mean scoring across the three timepoints. For each sub-factor and overall score, a significant difference in scoring was found across the three phases.

Regression
Finally, a regression analysis (Table 4) was undertaken to investigate the impact of working conditions on psychological well-being. The regression model demonstrated good fit (p < 0.001), and accounted for 60 per cent of the variance. Each of job satisfaction, working conditions, control, well-being and home-work interface significantly predicted well-being scoring, with stress at work the only non-significant factor. For each factor, tolerance scores above 0.2, and variance inflation factor (VIF) less than 10, indicating no collinearity (Field, 2013).

Discussion
The aim of this article was to illuminate the changing conditions under which social workers and social care workers were working across three phases during the COVID-19 pandemic. This is amongst the first articles to compare working conditions across multiple phases of the pandemic for UK social workers and social care workers, with the majority of previous studies being individual snapshots in time. Analysis demonstrated that both psychological well-being and WRQoL, as well as all of the sub-factors (career satisfaction, stress at work, working conditions, general well-being, control, and home-work interface), were significantly poorer in phase 3 than they were in phase 1. This suggests that both working conditions and psychological well-being significantly worsened between November 2020 and July 2021 (i.e. across the three phases of the project). Compared with the national average, scoring on psychological well-being at phase 1 (21.10) suggests much poorer well-being than the national average of 23.61 (Ng Fat et al., 2016), with this scoring significant worsened at phase 2 (19.96) before recovering slightly at phase 3 (20.06), but still being worse than both phase 1 and the UK national average. As such psychological well-being has gotten worse across the pandemic, and was maintained at a chronically poor state. Additionally, analyses demonstrated that each of career satisfaction, working conditions, control at work, general well-being and home-work interface were significantly influencing the experience of poorer psychological well-being. This is consistent with the wider literature on impact of working conditions during the pandemic, on service delivery, practice challenges and on and worker well-being and coping (Baginsky and Manthorpe, 2021;Harrikari et al., 2021;Manthorpe et al., 2021;McFadden et al., 2021;Gillen et al., 2022;Ravalier et al., 2022).
Pre-COVID 19 pandemic, social workers had amongst the worst working conditions of any occupational sector in the UK (e.g. Ravalier, 2019;Ravalier et al., 2021a). However, this article reveals that perceived working conditions and well-being continue to be poor, and have worsened across the pandemic. Reddington et al. (2021) reported that the social care workforce was under severe pressure during the early stages of the pandemic as requests for social care support increased, whilst the number of social work and care workers was static, albeit with increasing COVID-19-related sickness absences and mortality. Changes in working practice within the sector (in particular for social work) were implemented, with many working exclusively from home, and in-person visits restricted (Abrams and Dettlaff, 2020). Social care workers were much less able to work from home, with this ultimately contributing to greater levels of COVID-19 sickness absence and mortality (Office for National Statistics, 2022). Changing working practices, associated with chronically poor working conditions, ultimately affected the stress experienced in the sector, with already high demands and limited resources (Ravalier, 2019) worsening across the pandemic. Indeed, these may be the reason that both psychological well-being and working conditions worsened across the three phases outlined in this article. These findings are consistent with the JDR model, indicating that job demands had increased whilst resources were not (sufficiently) available to counter the negative effects of increased work demands. The increasing (perceived) mismatch between demands and resources therefore explains increased stress and lower psychological well-being across both occupations investigated here.

Strengths and limitations
This is amongst the first studies to investigate the ongoing impacts of the COVID-19 pandemic on two of the sectors described as 'key workers' by the UK government : social workers and social care workers. This article therefore adds significantly to the literature. Furthermore, we used well-established, valid and reliable tools which have been successfully used in the sector to measure similar outcomes, with data collected from a relatively large sample. The cross-sectional design also allowed several outcome and predictor variables to be examined, a further strength of the project's methodological approach (Wang and Cheng, 2020). VIF and tolerance scores also suggested no collinearity within the model. However, limitations persist.
Collecting data using a self-report survey methodology may increase the likelihood of recall bias, selection bias and/or desirability bias (Rosenman et al., 2011). Similarly, we used a snowball sampling approach to participant recruitment. This can often lead to a sample which is unrepresentative and unbalanced because of the self-selection methodology (Sadler et al., 2010). However, a cross-sectional survey methodology was the most appropriate and ethical way of collecting large amounts of data from a population which is key to the pandemic response across the UK and beyond.
Furthermore, despite the strengths outlined above, this cross-sectional survey cannot be used to determine causality. The survey methodology is also limited in that it can only find what it was set out to find and therefore further environmental and working conditions impacting upon wellbeing have not been measured. Therefore, findings need to be viewed within this light. Finally, across all three phases of data collection, our sample was predominantly female (85 per cent þ), aged 40 and over, 90 per cent white ethnicity, with three quarters of respondents describing themselves as not having a disability. These demographic descriptions may therefore suggest that the findings of the current project are not representative of either male, younger, non-white social workers and care workers, or those with a disability. Despite this, these demographics are broadly representative home of the social work and social care professions (Batra et al., 2020;Holmes et al., 2021). Thus, the authors do not make generalisability claims from the data. Rather we see the research as snapshots across time during different timeframes of the pandemic as it developed in the UK.

Implications and future research
Chronically poor working conditions can lead to poorer individual psychological and physiological health outcomes (e.g. Ravalier, 2019). The findings from this study suggest that working conditions for social workers and social care workers have been poor for some time, and worsened across the first year of the pandemic. The major implication therefore is that stressors (and thus working conditions) need to be improved in order to support the workforce, and ultimately to ensure the best possible outcomes for service users.
Future research would also be useful in understanding social care and social worker beliefs towards working conditions, and potential interventions which can be implemented to support the working conditions of social care and social workers. Projects should seek to develop and robustly evaluate support interventions, focused on both primary (organisational) and secondary (individual employee) interventions to support working conditions (and ultimately individual psychological health and well-being). There is also a distinct lack of high-quality evidence demonstrating the impact of psychological health interventions on service user and client outcomes in social care. Indeed, whilst such research is more common in the healthcare sector (e.g. West and Dawson, 2012), it is less so in the social care sector. As such, as well as developing and evaluating interventions within the sector, the impact of these interventions on outcomes for those people that workers support should also be investigated.

Conclusions
To summarise, evidence has demonstrated that the UK social care system has amongst the worst working conditions of all occupations, although many maintain a strong level of engagement in their work. Results from the present study have demonstrated that these working conditions have worsened across the pandemic, with psychological wellbeing also worsening. These results provide timely and convincing evidence of the pressures experienced by social workers and the social care workforce across the pandemic. They also indicate that future research and innovations which seek to improve working conditions should be able to show the impact on the psychological well-being of these key workers. Doing so would not only improve individual employee wellbeing but may also improve outcomes for employing organisations and, perhaps most importantly, for the clients and service users.

Research ethics
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by Filter Ethics Committee in the School of