Discrimination, feeling undervalued, and health-care workforce attrition: an analysis from the UK-REACH study

There are increasing concerns about health-care staff leaving the National Health Service (NHS) workforce, and the substantial adverse knock-on effects that attrition has for patient care, which the COVID-19 pandemic is likely to have exacerbated. In July, 2022, a report by the UK Health and Social Care Committee stated that “The NHS and the social care sector are facing the greatest workforce crisis in their history”,1House of Commons Health and Social Care CommitteeWorkforce: recruitment, training and retention in health and social care. House of Commons, London2022Google Scholar with estimated shortages of 12 000 hospital doctors and more than 50 000 nurses and midwives,1House of Commons Health and Social Care CommitteeWorkforce: recruitment, training and retention in health and social care. House of Commons, London2022Google Scholar while demand for services increases and waiting lists grow. NHS staff data indicate that the numbers of staff leaving since 2021 vary by region, professional group, gender, age, and country of professional qualification;2Palmer B Rolewicz L The long goodbye? Exploring rates of staff leaving the NHS and social care.https://www.nuffieldtrust.org.uk/resource/the-long-goodbye-exploring-rates-of-staff-leaving-the-nhs-and-social-careDate: Feb 9, 2022Date accessed: October 28, 2022Google Scholar, 3Palmer B Rolewicz L Peak leaving? A spotlight on nurse leaver rates in the UK.https://www.nuffieldtrust.org.uk/resource/peak-leaving-a-spotlight-on-nurse-leaver-rates-in-the-ukDate: Sept 30, 2022Date accessed: October 28, 2022Google Scholar however, little information exists on the reasons staff from different groups are leaving. Furthermore, data from the 2021 NHS Staff Survey showed that more than half of respondents were considering changing jobs, but it is uncertain why, and crucially, what would encourage and enable them to stay.4National Health ServiceNHS Staff Survey National Results.https://www.nhsstaffsurveys.com/results/national-resultsDate: 2021Date accessed: October 27, 2022Google Scholar The General Medical Council workforce report published in October, 2022, called for “Workforce planners [to] consider the data regarding leaving rates and what lies behind them so that methods for improving retention can be found.”5General Medical CouncilThe state of medical education and practice in the UK: the workforce report 2022. General Medical Council, London2022Google Scholar A previous study conducted in the USA early in the pandemic, which collected data from July to December, 2020, reported that health-care workers (HCWs) who feel valued by their organisation are less likely to reduce their working hours or leave their jobs than those who do not feel valued.6Sinsky CA Brown RL Stillman MJ Linzer M COVID-related stress and work intentions in a sample of US health care workers.Mayo Clin Proc Innov Qual Outcomes. 2021; 5: 1165-1173Summary Full Text Full Text PDF PubMed Google Scholar A pre-pandemic systematic review identified feeling undervalued by an employer and experiencing discrimination at work were negatively associated with job satisfaction and retention in the NHS.7Bimpong KAA Khan A Slight R Tolley CL Slight SP Relationship between labour force satisfaction, wages and retention within the UK National Health Service: a systematic review of the literature.BMJ Open. 2020; 10e034919Crossref PubMed Scopus (17) Google Scholar Considering the current staffing crisis facing the NHS, and to inform interventions, we sought to identify the proportion of HCWs who are considering or have acted on intentions to change or leave their health-care role as a result of the COVID-19 pandemic. We also sought to investigate whether such intentions are associated with feeling undervalued (ie, by the UK Government, the general public, and their employer), experiences of discrimination at work, and some sociodemographic and occupational parameters. We conducted a cross-sectional analysis using questionnaire data from the third wave (October–December, 2021) of The UK Research Study into Ethnicity and COVID-19 Outcomes in Healthcare Workers (UK-REACH) longitudinal cohort study.8Woolf K Melbourne C Bryant L et al.The United Kingdom Research study into Ethnicity And COVID-19 outcomes in Healthcare workers (UK-REACH): protocol for a prospective longitudinal cohort study of healthcare and ancillary workers in UK healthcare settings.BMJ Open. 2021; 11e050647Crossref Scopus (14) Google Scholar HCWs or ancillary workers in a UK health-care setting aged 16 years or older and living in the UK were recruited via all major UK health-care regulators and hospital trusts (further details on inclusion criteria and recruitment are shown in the appendix p 2). Our outcome was binary and derived from the questionnaire item “Has the COVID-19 pandemic made you consider or act upon any of the following in relation to your work? (select all that apply)”. Participants could select “No”, “Yes, considered”, or “Yes, acted upon” in relation to the following options: “1. Reducing the hours you work in your current job”; “2. Changing the field in which you work (e.g. changing speciality)”; “3. Leaving your healthcare role entirely”; “4. Reducing clinical duties”; “5. Taking early retirement”; “6. Other (please specify)”; and “0. None of the above”. Responses to the questionnaire item allowed participants to be coded as either having considered or acted on making any changes to their role in response to the COVID-19 pandemic (ie, 1) or not (ie, 0). Our primary exposures of interest were answers to questions about whether an HCW felt their work was valued (ie, by the Government, by their employer, and by the public) and experiences of discrimination at work (ie, from colleagues, patients, or both). We used multivariable logistic regression to establish the association between our outcome and these exposures. We constructed a base model of age, sex, ethnicity, and occupation and added each of our primary exposures separately to the model. We present results as adjusted odds ratios (aORs) and 95% CIs. We investigated interactions between demographic or occupational covariates with each of our primary exposures of interest by fitting models with and without the interaction and comparing model fit by use of likelihood ratio tests. Detailed methods are shown in the appendix (pp 4–5). We excluded data from people who did not provide information on the outcome and primary exposures of interest. As questions about whether an HCW felt their work was valued were asked to only those who indicated that they were currently working, people who indicated that they were not working in any capacity were excluded from the main analysis. We determined the reasons given for not currently working in this group and stratified the group by our outcome measure. Finally, because individuals who left the health-care workforce and took up a role outside of health care could have answered questions about whether they felt their work was valued with respect to their current role (rather than their health-care role), we completed a sensitivity analysis excluding people who indicated they had left their health-care role or had taken early retirement (further details are shown in the appendix pp 4–5). Formation of the analysed sample is shown in the appendix (p 6). Recruitment began on Dec 4, 2020, and continued until Feb 28, 2021. In total, 17 891 HCWs were recruited into the study, and 15 199 responded to the baseline questionnaire. 5892 of 15 199 HCWs who had completed the baseline questionnaire also completed the third questionnaire. 4916 respondents provided information on the primary exposures and outcome of interest and were included in the main analysis. A description of the analysed sample is presented in the appendix (p 7). Overall, 2358 (48·0%) of 4916 staff considered or acted on changing or leaving their role (1668 [33·9%] considered and 690 [14·0%] acted on). After adjustment for age, sex, ethnicity, and job role, the groups most likely to report making changes to, or leaving, their health-care role were women versus men (aOR 1·45, 95% CI 1·25–1·67; p<0·0001); people who self-categorised as being from mixed or multiple ethnic groups of White and Black Caribbean, White and Black African, White and Asian, and any other mixed or multiple ethnic backgrounds versus people who self-categorised as White (1·47, 1·09–1·98; p=0·011); people aged 50–59 years versus those aged 40–49 years (1·32, 1·13–1·54; p=0·0004); and those in nursing or midwifery roles versus those in medical roles (1·25, 1·03–1·50; p=0·022). Health-care scientists were less likely than medical staff to report attrition intentions (aOR 0·61, 95% CI 0·46–0·82; p=0·0010), as were allied health professionals (0·84, 0·70–0·99; p=0·041; figure). Overall, 1041 (21·2%) of 4916 staff reported having experienced discrimination in the past 6 months (403 [8·2%] participants reported discrimination from patients, 449 [9·1%] from colleagues, and 189 [3·8%] from both patients and colleagues). 2338 (47·6%) staff strongly disagreed or disagreed that their work was valued by the Government, 1009 (20·5%) strongly disagreed or disagreed their work was valued by their employer, and 869 (17·7%) strongly disagreed or disagreed that their work was valued by the public (appendix p 7). After adjustment for demographics and job role, attrition intentions or actions were strongly associated with experiencing discrimination, with higher odds of attrition intentions if an HCW had experienced discrimination from colleagues (aOR 2·84, 95% CI 2·29–3·51; p<0·0001), patients (2·06, 1·66–2·56; p<0·0001), and colleagues and patients (2·96, 2·14–4·08; p<0·0001) than if an HCW had experienced no discrimination. Compared with people who neither agreed nor disagreed, participants were far more likely to report attrition intentions or actions if they strongly disagreed that their work was valued by the Government (aOR 2·49, 95% CI 2·10–2·95; p<0·0001), their employer (1·83, 1·39–2·42; p<0·0001), or the public (2·07, 1·52–2·81; p<0·0001). The only interaction that improved model fit was between age and feeling valued by the public (appendix pp 8–9). Reasons given by those not working at the time of data collection are shown in the appendix (p 10). Proportions of participants who were considering or had acted on changing or leaving their role were similar when those not working at the time of data collection were included (appendix p 11). Nearly half of the HCWs in this study reported intentions to change or leave their health-care role. This result is highly concerning given that the NHS is already short of 103 000 full-time equivalent staff, with shortages projected to grow to 179 000 in the next 2 years.9The Health FoundationNHS workforce projections 2022.https://www.health.org.uk/publications/nhs-workforce-projections-2022Date: July, 2022Date accessed: October 31, 2022Google Scholar Such staff shortages will put increasing burden on remaining staff, most likely exacerbating attrition and ultimately risking patient safety. Additionally, we have identified several important factors associated with intentions to change or leave a health-care role as a result of the COVID-19 pandemic, including feeling undervalued, experiencing discrimination at work by colleagues or patients, and belonging to particular demographic and occupational groups. Our study has several limitations. This study is a cross-sectional analysis, and some of the associations reported could be bidirectional. The analysis might be affected by selection bias but, given that the study was not advertised as specifically relating to workforce attrition, it avoids the framing effects that might be seen in studies specifically investigating this topic. As questions used to derive information on whether HCWs felt their work was valued were asked only to those currently working, we could have underestimated the proportion of people acting on attrition intentions (because people who had left the health-care workforce entirely and not taken on another role would have been excluded); however, the proportions of those who had considered or acted on changing their role were similar when the non-working cohort were included. The UK-REACH study initially aimed to over-recruit HCWs from Asian, Black, mixed (ie, self-identifying as from mixed or multiple ethnic groups of White and Black Caribbean, White and Black African, White and Asian, and any other mixed or multiple ethnic backgrounds), and other (ie, self-identifying as Arab or from any other ethnic group) ethnicities, as defined by the Office for National Statistics. 1276 (26%) of 4916 people in the analysed sample of this study were from these ethnic groups, which is a similar proportion to that of the NHS (337 725 [25·7%] of 1 315 562 people with known ethnicity).10UK GovernmentNHS workforce.https://www.ethnicity-facts-figures.service.gov.uk/workforce-and-business/workforce-diversity/nhs-workforce/latestDate: April 13, 2023Date accessed: July 22, 2023Google Scholar Because of this similarity, we do not feel that the difference between our aim and the number of people recruited with these ethnicities will have substantially affected the generalisability of our results. This study greatly adds to the little information in the literature concerning attrition in the health-care workforce during the COVID-19 pandemic. Our results are concerning and suggest that policy makers need to find and implement solutions at both national and organisational levels to reduce discrimination, improve staff satisfaction and wellbeing, and improve retention to prevent the workforce crisis from worsening. MP conceived of the idea for UK-REACH and led the application for funding with input from KW, LBN, KK, and the UK-REACH Study Collaborative Group. The questionnaire was designed by CAM, KW, LBN, KK, MP, and the UK-REACH Study Collaborative Group. CAM, KW, and MP formulated the idea for the analysis and contributed to the analysis plan with input from AM, MG, LT, and LBN. CAM analysed the data with input from LT, KW, and MP. CAM and MP have accessed and verified the underlying data. CAM and KW drafted the Correspondence with input from MP. CAM, AM, MG, LT, JN, DP, SC, KK, KW, and MP edited and approved the final version of the Correspondence for publication. All authors had full access to all the data in the study and had final responsibility for the decision to submit for publication. KK is Director of the University of Leicester Centre for Black Minority Ethnic Health, Trustee of the South Asian Health Foundation, and Chair of the Ethnicity Subgroup of the UK Government Scientific Advisory Group for Emergencies. MP reports grants from Sanofi, grants and personal fees from Gilead Sciences, and personal fees from QIAGEN, unrelated to this Correspondence. MP reports consulting fees from Pfizer, unrelated to this Correspondence. SC is Deputy Medical Director of the General Medical Council. KW has received grants from the NHS Race and Health Observatory and was in receipt of a National Institute for Health and Care Research Career Development Fellowship Award during this study, and a National Institute for Health and Care Research Skills Development Award. KW has received payment for PhD examination at Southampton University and for lectures at Leicester University and Hampshire Hospitals NHS Foundation Trust. KW has also received payment from the Ministry of Education United Arab Emirates for a review of a degree programme. All other authors declare no competing interests. To access data or samples produced by the UK-REACH study, the working group representative must first submit a request to the Core Management Group by contacting the UK-REACH Project Manager in the first instance ([email protected]). For ancillary studies outside of the core deliverables, the Steering Committee will make final decisions once they have been approved by the Core Management Group. Decisions on granting the access to data or materials will be made within 8 weeks. Third party requests from outside the project will require explicit approval of the Steering Committee once approved by the Core Management Group. Note that should there be significant numbers of requests to access data or samples then a separate Data Access Committee will be convened to appraise requests in the first instance. UK-REACH is supported by a grant from the Medical Research Council-UK Research and Innovation (MR/V027549/1) and the Department of Health and Social Care through the National Institute for Health Research (NIHR) rapid response panel to tackle COVID-19. Core funding was also provided by NIHR Biomedical Research Centres. JN is an NIHR Academic Clinical Fellow. DP is an NIHR Doctoral Fellow (NIHR302338). KW is funded through an NIHR Career Development Fellowship (CDF-2017–10–008). LBN is supported by an Academy of Medical Sciences Springboard Award (SBF005\1047). KK and MP are supported by the NIHR Applied Research Collaboration East Midlands. KK and MP are supported by the NIHR Leicester Biomedical Research Centre. MP is funded by an NIHR Development and Skills Enhancement Award. This work was carried out with the support of BREATHE-The Health Data Research Hub for Respiratory Health [MC_PC_19004] in partnership with SAIL Databank. BREATHE is funded through the UK Research and Innovation Industrial Strategy Challenge Fund and delivered through Health Data Research UK. The funders of the study had no role in the design, data collection, data analysis, data interpretation, or writing of the report. Members of the UK-REACH Study Collaborative Group are listed in the appendix (p 1). Download .pdf (.41 MB) Help with pdf files Supplementary appendix