The Relationship Between Employee Engagement and Organisational Outcomes in the English National Health Service: An Analysis of Employee and Employer Data in 28 Healthcare Organisations

Abstract

There have been concerns around low productivity, poor health and wellbeing and staff engagement for well over a decade in the English National Health Service (NHS). Our analysis of survey data in 28 healthcare organisations builds an understanding of what drives employee engagement and how it impacts organisations. Our findings are that: better staff engagement in a healthcare organisation is associated with better financial and care outcomes; some professional groups in healthcare organisations (e.g. healthcare assistants) have substantially worse employee engagement than others; and employee participation in wellbeing programmes, effective leadership and line manager training are associated with better staff engagement. The key message for those involved in healthcare delivery is that there is a strong business case for improving employee engagement as it likely improves a range of organisational outcomes. The analysis also points to some specific initiatives that organisations focused on improving employee engagement may wish to consider.

1 Introduction

There have been concerns around low productivity, poor health and wellbeing and staff engagement for well over a decade in the English National Health Service (NHS). In 2009 the Department of Health commissioned an independent review into the health and wellbeing of NHS led by Dr. Steven Boorman. The Boorman Review as it became known gave an indication of potential savings (mostly through reduced staff absenteeism) that could be made by organisations from adopting more effective ways of managing the health and wellbeing of staff. Especially, it looked at whether hospital boards and senior managers discussed staff health and wellbeing in executive and board meetings, whether they showed role modelling behaviour and whether line managers were trained to look after employee health. Savings to the NHS alone were estimated at £500 million a year (see for instance the Interim Report of the Boorman Review, Boorman & Fellow, 2009). The report also highlighted interesting associations between staff health and wellbeing in the NHS and a range of outcomes such as quality of care lower, mortality rates, and reduced rates of hospital acquired infections. Building on this, the Keogh Review (2013) into patient safety reviewed the quality of care provided by a number of Trusts,¹ and recommended a number of actions to improve patient outcomes. One of the actions was that all NHS organisations seek to understand the positive impact that happy and engaged staff can have on patient outcomes, including mortality rates, and that this should form a key part of their quality improvement strategy. This focused mainly on involving staff more in decision-making in the health service and indeed decisions regarding the delivery of care.

Wider research on the health service has been seeking to understand the associations between employee experience, health and wellbeing, performance and organisational outcomes in the NHS (Dawson, 2014). Seminal work undertaken by Michael West and Jeremy Dawson (2012) outlined the link between employee experience and positive organisational outcomes (including better financial results, lower rates of hospital acquired infections, and lower mortality rates) and positive patient perceptions on care as measured in patient surveys. While some reviews in the US did not find a clear relationship between employee health and wellbeing and quality of care (Tawfik, Scheid, Profit, et al., 2019), others find that staff wellbeing is also linked to patient outcomes (Maben, Adams, Peccei, Murrells, & Robert, 2012). Dawson also undertook a review of the literature on staff experience and patient outcomes. This study concluded that there is evidence that the experiences of staff, particularly in the form of support received from supervisors and line managers, and staff engagement, are associated with quality of care, patient satisfaction, health outcomes, and ratings of quality of care, as well as staff absenteeism and retention (Dawson, 2014). Other research shows that health professionals who show higher engagement are less likely to make mistakes and produce better patient outcomes (Laschinger & Leiter, 2006). In addition, in the NHS, higher levels of staff engagement are associated with better health outcomes, lower levels of stress and lower presenteeism (the state of being at work but in suboptimal health) (Admasachew & Dawson, 2010). Recent work around presenteeism hints at associations between presenteeism and wider organisational outcomes mainly through lower productivity and reduced engagement with colleagues and the work environment (Lohaus & Habermann, 2019).

In this chapter, we want to look in particular at the concept of employee engagement in the English NHS and build on the Boorman Review and the work of West and Dawson. An issue with some of the reviews has been that they have used secondary data such as NHS administrative data and the NHS staff survey to draw inferences. These sources are not necessarily set up to build a holistic view on the employee experience, health and wellbeing, and the engagement of staff. We will draw on two large surveys that included large NHS employers (mostly acute hospital trusts). These are respectively the Britain’s Healthiest Workplace (BHW) and the NHS Healthy Workforce survey administered by RAND Europe in 2016. We expand on how these surveys are conducted below. The main advantages of using these surveys are that they include a wide range of variables in one data set including variables on health and wellbeing, employee engagement and indeed outcome measures. Overall, this study is complementary to the seminal work by West and Dawson (2012) in that it looks at the determinants and outcomes of employee engagement within the NHS, but using a different data source and with a focus on different organisational level interventions (e.g. health and wellbeing interventions, leadership training) beyond the common human resource management (HRM) practices. They offer both an employer as well as an employee perspective by surveying both an employer representative as well as employees. We can also dig a bit deeper below the surface and understand which groups are particularly affected by low staff engagement and the circumstances in which they work. We will discuss the data sources and analysis below.

In this research, we want to investigate a number of hypotheses on the basis of the data.

• There is a relationship between employee engagement and organisational outcomes (H1)

• There is a relationship between employee engagement and specific human resource practice (H2)

• There is a relationship between employee engagement and ‘good work’ (H3)

• There is a relationship between employee engagement and the health and wellbeing of staff (H4)

2 What Is Employee Engagement?

It has proved difficult to conceptualise what we mean by employee engagement. At times, concepts such as health and wellbeing and employee experience are intertwined. This is because some of the reviews mentioned earlier show associations between the health and wellbeing of staff, life satisfaction, job satisfaction, perceptions of the work environment, and employee experience.

The topic of employee engagement gained particular prominence in 2008 when the Secretary of State for Business in the UK commissioned David MacLeod and Nita Clarke to take an in-depth look at employee engagement. The report, ‘Engaging for Success’ also known as the MacLeod Review explored the potential role of employee engagement to improve organisational competitiveness and productivity in the UK (MacLeod & Clarke, 2009).

This chapter does not aim to provide a comprehensive overview of the literature on employee engagement. There are varying definitions and concepts used. We can broadly distinguish between five approaches. Firstly, Kahn (1990) focuses on personal engagement and sees employee engagement as an expression of the physical, emotional, and psychological self at work. There is an interplay between the self, the role, specific job and environment that a person is in and as such personal engagement is likely to change over time and vary. Higher engagement is then linked to certain number of physical, emotional, and psychological characteristics. Secondly, another body of literature sees employee engagement as the commitment or involvement an individual employee has to or with the objectives and values of an organisation. Commitment and involvement in this sense refer to a psychological state that the individual is in (Robinson, Perryman, & Hayday, 2004). Thirdly, engagement can be seen as a performance construct. So, we would look to a certain set of organisational or social behaviours in an employee that would indicate engagement (West & Dawson, 2012). Within this sit a wide range of concepts: including showing proactive behaviour (Crant, 2000); taking personal initiative (Frese & Fay, 2001); and organizational citizenship behaviour (Organ, 1988). A fourth concept combines the last two and looks at both commitment and involvement as well as behaviours or effort that the employee shows or puts in. Finally, we need to acknowledge some differences in the conceptualisation between practitioners and psychologists. There is a literature that defines engagement as a set of organisational practices and behaviours that are associated with high engagement. This focuses on organisational aspects rather than the individual. It can for instance include involvement in decision-making or processes like consultation (Dickinson & Ham, 2008).

Overall, the characteristics of an engaged staff combine both the individual’s psychological state as well as behaviours and effort. It can be described as motivation, satisfaction, involvement, commitment, meaningful work, initiative and pride. There remain key challenges with how engagement is operationalised and different constructs exist to measure staff engagement. In this chapter, we use two concepts of employee engagement that combine both psychological state and behaviours. We provide some more detail below. Our main interest is to show the important associations between the concept of employee engagement and a wider range of variables. Though we draw inferences from the literature, we cannot necessarily be clear about the direction of causality in our analysis.

3 Data and Analysis

The NHS Health Workforce Survey was conducted by RAND in 2016 and collected responses from NHS health organisations and their employees through the Organisational Health Assessment (OHA) and the Employee Health Assessment (EHA) respectively. Participating organisations returned the OHA, including general organisational characteristics such as the size and nature of the organisation, the work environment, and information on the organisation’s approach to health promotion and well-being interventions. Subsequently, employees were invited to respond to the EHA, which collected socio-demographic information (such as age, gender, income, general background); lifestyle and behavioural and clinical risk factors (including weight, diet, exercise, smoking, alcohol intake, stress, cholesterol and blood pressure); data on the work environment and culture; and information on how often people participate in organisational health and well-being interventions. The EHA had about 150 questions that aimed to provide a holistic picture. We stipulated that a minimum number of employees per organisation had to participate in order to have confidence in the sample. We calculated this number on the basis of the total employees and we wanted a 90% level of confidence and accepted a 5% margin of error. In total, 19 organisations took part in this survey. These were mostly organisations that were participating in a NHS health and wellbeing initiative or similar organisations that were approached for participation in the survey. In terms of the former, it stands to reason that these 11 trusts or primary care organisations were motivated participants as they were already included in a central initiative conducted by NHS England to improve the wellbeing of staff. In terms of the latter, we identified eight trusts that had similar outcomes and profiles to the other 11. The purpose here was to see whether there was something distinct about the 11. Most of our descriptive survey data suggested that there was nothing particularly exceptional about the 11 trusts when compared to the eight or general staff survey. However, on a number of organisational outcomes, the 19 trusts and primary care organisations performed slightly better than average on organisational outcomes when comparing results seen in administrative data and the NHS staff survey.

The organisations have a combined headcount of 105,838 employees and the survey was distributed to 91,872 staff and a complete survey was collected for 7246 employees, resulting in a response rate of about 8%. This response rate is lower than the NHS Staff Survey, which offers a representative sample across all NHS organisations, but our sample was broadly similar across the main demographic variables including age, gender, job category and income levels. We are over-represented in having a large group of mid-career NHS workers and slightly under-represented in the lower income groups and younger and older workers. This is similar to the sample bias in the annual NHS Staff Survey. So despite a lower response rate than in the annual NHS staff survey, we are confident that there is nothing particularly distinct about the bias in our sample. Where we can compare to the general NHS staff survey, our survey shows similar data.

In addition to the NHS Healthy Workforce survey, RAND Europe has conducted Britain’s Healthiest Workplace (BHW) survey on an annual basis since 2014. In contrast to the NHS Healthy Workforce Survey, the BHW survey is generally open to all companies, government organisations and other organisations in the UK. For instance, in 2016 it includes nine NHS organisations as well, for which we use data in this study. In this research we combine both surveys. The surveys are about 90% identical and broadly use similar variables. The response rates are also similar (8%) and as before the BHW sample is similar to the NHS Healthy Workforce survey. In total the data sample includes 28 organisations with a combined number of respondents of 9375.

It is important to note that while the participating organisations represent a diverse range of NHS organisations, generalisations of the findings to the wider NHS should only be made with some caution as the sample of participating organisations was not intended to be representative of the entire NHS. The surveys are overweighting secondary care acute trusts. Some community and mental health trusts participated but we have no primary care participation other than one clinical commissioning group. This is not exceptional in NHS surveys as the primary care population is fragmented and hard to survey. Often, the primary care population attracts dedicated surveys. As a result, our findings are mostly relevant to larger health organisations that operate as trusts.

In order to analyse the association between employee engagement and a variety of outcome measures, we complement the information from the NHS Healthy Workforce and BHW surveys with information at the trust level provided by the NHS. Specifically, we use administrative sickness absence rate data, patient satisfaction surveys for acute and mental health trusts by the Care Quality Commission (CQC)² and NHS account data which can be merged at the trust level to the NHS Healthy Workforce survey.

Similar to the study by West and Dawson (2012), we conduct multilevel regression analysis using two engagement constructs across all employees in our sample (from 9 BHW trusts and 19 NHS Healthy Workforce trusts) and examine variables associated with employee engagement. The two constructs that we use are the Utrecht Work Engagement (UWES) nine point scale (Schaufeli, Bakker, & Salanova, 2006) and a composite construct of engagement. The UWES scale focuses on dimensions such as absorption, vigour and dedication. Each dimension has three questions. The use of two constructs was necessary because the surveys used slightly different engagement questions. For the second measure, we followed the approach taken by West and Dawson (2012) that focuses on psychological engagement and involvement in decision-making. For psychological engagement we use the standardised value of job satisfaction and for involvement in decision-making the amount of involvement an employee has at his work as measured through participation in meetings and having knowledge of initiatives, organisational guidance and other organisational dissemination. The correlation coefficient between the engagement indicator and the total UWES scale is 0.58, which suggests a non-perfect but significant correlation. Hence, we are confident that our findings hold across the two samples.

As a first step, the overall engagement score and the UWES-9 overall score are included as dependent variables. In these models, the independent variables include the following:

• Demographic: education, ethnicity, age, gender, income;

• Occupational: type of NHS occupation, length of tenure;

• Work environment: workplace related stress as measured by the Health and Safety Executive management standards (control, relationships, time pressure, etc.), management support, leadership, bullying;

• Health and lifestyle: musculoskeletal and chronic health conditions, mental health, sleep and BMI³;

• Personal: children, providing unpaid care, financial concerns;

• Human capital: HR practices, provision of health and wellbeing interventions.

In a second step, the overall engagement scores are used to examine the link between engagement and outcome variables, holding other drivers that may determine engagement and the dependent variables simultaneously. Here we are mainly interested in the relationship between employee engagement and the following outcome variables:

• Absenteeism: individual level, reported by the employee using the Work Productivity and Activity Impairment scale (WPAI)⁴;

• Presenteeism: individual level, reported by the employee using WPAI scale;

• Sickness absence rates: organisational level;

• Overall patient satisfaction: organisational level⁵;

• Staff turnover: organisational level⁶;

• Financial situation: organisational level⁷;

• Operational surplus/deficit: organisational level.^8,9

The statistical analysis is conducted using STATA Version 15.¹⁰ All results as is common in this type of analysis are presented at the 10% significance level or lower.

4 Findings

We present the findings here alongside the hypotheses that we introduced in our introduction. The first hypothesis focuses on the relationship between staff engagement and organisational outcomes. Our analysis consistent with West and Dawson (2012) finds an association between engagement and wider organisational outcomes. Here we combine absenteeism and presenteeism as measured in our surveys using the WPAI with administrative data reported by NHS organisations through administrative data sets (see Table 8.1). These include data on absenteeism and staff turnover. Finally, we look at a wider set of organisational outcomes such as financial situation, patient satisfaction and operational surplus (see Table 8.2). We find that higher staff engagement is associated with better organisational outcomes across the board. These findings hold for variables collected through our surveys or when introducing administrative data. In some cases, we have similar data from multiple sources such as for instance on staff absenteeism. Absenteeism data is collected in the employer survey, the employee survey (using the WPAI) and in NHS administrative data. The relationship holds across the data sources. Importantly, our analysis shows an association between presenteeism and engagement.

Table 8.1 Employee engagement and the association with absenteeism, presenteeism and turnover (^*,^**, and^***) denoting statistically significant outcomes with level of significance increasing with the number of stars)

Table 8.2 Employee engagement and the association with patient satisfaction and financial performanc e (^*,^**, and^***) denoting statistically significant outcomes with level of significance increasing with the number of stars)

The second hypothesis looks at the relationship between engagement and human resource practice. Here, we are mainly interested in variables around the health and wellbeing programme or employer offer to employees and leadership training. These are the variables below in Table 8.3 with a # symbol next to them. Our analysis finds that certain components of the offer are associated with higher staff engagement. Those NHS organisations with a greater number of physical health interventions and those that have health checks in place for staff have higher staff engagement. At the same time, we see an association between leadership (manager) training on employee health and wellbeing and greater staff engagement. Leadership is often seen as a key enabler as it also signals intent to staff that the organisation is willing to change and to engage with the needs of staff. We do not find the same association between a greater number of mental health interventions and staff engagement. The latter is perhaps surprising and an indication we need to understand the relationship between the specific components of a health and wellbeing offer and engagement better.

Table 8.3 Full regression table of all socio-demographic, occupation, work environment, health and wellbeing, and HR practice variables and employee engagement (^*,^**, and^*** denoting statistically significant outcomes with level of significance increasing with the number of stars)

The third hypothesis considers whether ‘good’ work is associated with better staff engagement. We first need to consider what good work looks like. Here, we look at proxy indicators (indicated by the^* symbol in Table 8.3) such as Health and Safety Executive (HSE) management standards, bullying, flexible working, financial wellbeing and working hours. The HSE management standards focus on relationships at work, control, role, time management, and unrealistic demands. Financial wellbeing is measured by the proportion of individuals who report having financial concerns. Of course, these indicators do not provide a holistic picture of the work environment but it stands to reason that good working environments have lower levels of bullying, allow flexible working, and indeed score better on the HSE management standards. Again, our analysis finds that across the board such proxy indicators are associated with better staff health and wellbeing. Those staff who can work from home and report having flexible hours have higher staff engagement. Clearly, working from home is not available for all NHS staff. All staff can request flexible working but relatively few are aware of the right to flexible working. Management functions have the greatest flexibility followed by clinical groups such as doctors. All HSE management standards are associated with staff engagement. This means that employees who report having a lack of control (control) and are not sure about their role (role) report lower levels of engagement. The same applies for employees reporting high levels of workplace demands (demand), employees that have a lack of peer support (peer support) at the workplace, that are bullied (bullied) and have strained relationships at work (strained relations). These are some of the most significant relationships in our analysis.

Two variables that we may have expected to have an association with staff engagement proved insignificant. Working hours and financial wellbeing were insignificant. This may reflect on the nature of work in the NHS and also the reasons why people join the NHS. Generally, it appears that employees do not join the NHS for financial gain and appear proud of the work they do. Antisocial working hours (including shift work, long hours) are perhaps seen as part of the job.

A fourth hypothesis focuses on the link between employee health and wellbeing and employee engagement. Here, we can look at a range of self-reported proxies collected through the EHA including obesity, quality and quantity of sleep, risk of mental health, chronic conditions (any long-term illnesses such as high blood pressure, cancer, diabetes, etc.) and musculoskeletal conditions (@ variables indicated in Table 8.3 below). There exist a number of significant associations between these conditions and engagement. In particular, the risk of mental health is strongly associated with employee engagement in our analysis. We capture this risk through two measures, the Warwick-Edinburgh scale¹¹ and the Kessler scale.¹² In addition, the self-reported assessment of quality of sleep where respondents rate their quality of sleep from very poor to very good on a Likert scale is associated with engagement. It is notable that few other variables show a significant relationship with engagement. There is for instance no relationship between engagement and musculoskeletal conditions or obesity in our analysis.

Finally, there are a wider range of demographic variables that have a positive association with engagement. Certain demographic and staff groups have lower staff engagement compared to others. For instance, social care workers, administrative support staff and managers tend to have lower engagement than other groups. We do not discuss those associations here in detail. However, it is obvious that healthcare organisations employ different types of staff and as such also have different sub-cultures. These need to be considered when looking at the issue of staff engagement.

5 Conclusions

Our evidence suggests that staff engagement is slightly below average in the NHS compared to other organisations that participate for instance in BHW. However, employee engagement is broadly similar in these NHS organisations when compared to other large employers in BHW and especially when adjusted for age, gender and income. It tends to be lowest in the healthcare assistant, nursing, midwives, and ambulance staff groups. So, employee engagement varies across staff groups and the issue tends to be more pronounced in certain staff groups. These groups also often have other significant wellbeing challenges such as worse mental health, higher rates of obesity and higher incidences of bullying compared to other staff groups.

The NHS in England has been under severe budgetary pressure since the financial crisis of 2008. In this climate, it has proved challenging making substantial progress on the employee experience in the service. Some initiatives have tried to move the debate forward. The Five Year Forward View, published by NHS England in 2014, underscores the importance of staff health and well-being as a crucial factor in improving the performance of the NHS (NHS England, 2014). This led to guidance and frameworks being rolled out across NHS health organisations with the aim to improve the health and wellbeing of staff. In the same year, Public Health England launched the Workplace Well-being Charter, which, for the first time, contains a set of evidence-based national standards for workplace health.¹³ In 2015, the National Institute for Health and Care Excellence (NICE) issued guidelines on workplace health (NICE, 2015). The guideline ‘covers how to improve the health and well-being of employees, with a focus on organisational culture and the role of line managers’.

The analysis in this chapter is important as it continues to build an understanding of what drives employee engagement and how it impacts organisations. By showing the associations between engagement and a wide set of organisational outcomes it also adds to the business case for health organisations to focus on their staff and work environments. The limitations of our work are several. Our sample focuses mostly on secondary care acute trusts and large NHS organisations. Therefore, we can say little about primary care. We are also limited by the data that we collect as part of the wider survey. As such, there may be variables or concepts that we do not ask about and that may be relevant for further in-depth analysis. Finally, some measurements are not identical across surveys. As such, we had to create some measures to compare outcomes. In terms of bias, we have mostly motivated NHS trusts participating in the surveys as our surveys rely on voluntary participation and dedicated HR staff rolling out the surveys in their organisations. Respondents of course are under no obligation to participate. However, the results and sample that we obtained are not particularly distinct and mostly similar when compared to other NHS data sources such as the NHS staff survey and NHS administrative data.

This chapter offers some concrete recommendations. The first is that it is important for senior executives in the NHS to know their numbers and understand the issues that they may have in their health organisations. Our surveys are an example of an approach but it is still striking from our experience how few healthcare organisations are aware of employee engagement and health and wellbeing and discuss these issues at board or senior executive level. Our findings also appear to suggest that a more holistic approach, which moves beyond single initiatives or interventions, may be important as work environment and culture variables show a positive association with staff engagement as conceptualised in our analysis. Of course, changing a work environment or culture is a difficult undertaking. The analysis also shows some specific relationships between human resource practice and engagement as well as mental health, quality of sleep and engagement. This offers some entry points for decision-makers and practitioners looking to improve employee engagement in the workplace. In particular, we want to emphasise the interesting associations between the size of the health and wellbeing programme and some specific components and employee engagement as well as the positive relationship between training leadership and line managers to look after the health and wellbeing of staff and employee engagement. Flexible working is also associated with better employee engagement but there remain issues in the NHS with general awareness of flexible working policies and how the wishes of staff are accommodated in working patterns. Bullying remains a toxic issue in the NHS, which on average is significantly higher than in most other organisations that we survey. We also need to consider different sub-cultures in an organisation. Our findings suggest some important differences across staff and demographic groups. So, improvements in staff engagement across a service may also require more tailored or targeted engagement with specific staff and demographic groups. Finally, some of these groups show a range of other challenges including worse mental health and higher rates of obesity than in other groups.

The key message for those involved in healthcare delivery is that improving employee engagement makes business sense as it likely improves a range of organisational outcomes. We suggest some entry points for managers and executives in the health service. Our work hints at some critical components of a strategy that tries to improve employee engagement. However, further research is required to show what aspects of HR practice or changes in work environment are particularly associated with improvements in employee engagement.

Key Messages

For healthcare professionals

• Better staff engagement in a healthcare organisation is associated with better financial and care outcomes

• Even where staff engagement across employees is good as measured on existing scales, there can be substantial differences in specific professional groups

• Employee participation in wellbeing programmes and leadership and line manager training are associated with better staff engagement

For researchers

• Looking at staff engagement in healthcare organisations contributes to a wider understanding of what drives performance in healthcare

• There is evidence on the association between staff engagement and organisational outcomes

• The evidence base around the effectiveness of health and wellbeing programmes is emergent

• As a result, our analysis of what human resource management practices are associated with better staff engagement points to some practices but more research is required