NHS managers' use of nursing workforce planning and deployment technologies: a realist synthesis

Initial scope of the literature

To guide the development of the initial programme theory, an exploration of the literature was conducted to help delineate how WPTs are supposed to work in particular contexts and to identify potential barriers to, and enablers of, their successful implementation. The study engaged with an information scientist (BH) with previous experience of realist synthesis, to help clarify the breadth of the resources available to support this synthesis. Scoping searches were conducted between May and September 2016. This scoping involved targeted searches of PubMed, MEDLINE and the Cumulative Index to Nursing and Allied Health Literature (CINAHL), a search for major reviews by The Cochrane Library and NICE and a search of the grey and allied literature included in key areas relevant to the expertise and networks of the research team (e.g. from UK mental health services). In addition, several health human resources and human resource management journals were searched. Further texts were gained from citation tracking. References were drawn on that were cited in the Royal College of Nursing guidance on safe nurse staffing levels in the UK and the Shelford Group Safer Nursing Care tool implementation guidance11,12 and a search was conducted for citations in identified articles via the Web of Science citation indices. Similarly, references were searched from an evidence review completed for NICE, Effectiveness of Management Approaches and Organisational Factors on Nurse Staffing Sensitive Outcomes;49 this resulted in the identification of six key articles that facilitated an understanding of the system complexity for this evidence synthesis. 118–123 Further scoping searches identified a further 11 sources of evidence from the literature. 124–134 Some of this literature was also identified in later targeted searches.

Conceptualising a model of nursing workforce planning and deployment

To illuminate the complexity of the contexts of the systems in which WPTs are used, soft systems methodology was used within the stakeholder co-production workshops. The format of the workshops, guided by soft systems methodology, was designed to understand the systems in which workforce deployment and planning operate through the co-production of a root definition. Soft systems methodology offers an epistemological approach for analysing complex, real-world issues, which may integrate multiple cognitive, social and cultural perspectives. 135 The methodology is appropriate for the examination of implementation challenges. 136 In addition, it is compatible with a realist approach for mapping complex programmes. 137 This approach accepts that issues that surround complex interventions and their implementation may be contested because of multiple stakeholder perspectives, which makes sense-making difficult. The relationship and synergy between people, systems, WPTs and resources around nurse workforce planning and deployment, can therefore be described as a soft, human adaptive system, which is open to differing perspectives and values about its purpose and impact. Data were explored using the six dimensions of a soft system (worldview, transformation, customers, owners, environment and actors), culminating in a narrative of the system of nurse workforce planning and deployment.

The two co-production workshops with NHS managers comprised a purposive sample of, in one case, 14 participants and, in the other, 17 participants from across NHS organisations and professional bodies within the UK, and combined a range of discussion and practical activities. Using techniques from LEGO^® SERIOUS PLAY^® (The LEGO Group, Billund, Denmark), individual models of the system of workforce planning and deployment of nurse staffing were constructed and important aspects from each individual model were then combined to create one overall collaborative nurse workforce planning and deployment model. Participants were required to offer explanatory accounts of the components they brought to the collaborative model; in this way, a root definition of the system was created that articulated multiple perspectives (see Appendix 2). Using the same technique, the initial PPI stakeholder groups (of seven participants) were asked to convey their ideas of what makes a good nurse manager (Figure 3). This knowledge contributed to the development of a root definition of a nurse workforce planning and deployment system. A second workshop explored the perceptions of staffing within the NHS. Seven PPI stakeholders participated and discussions centred on how staffing levels made visible by the organisation influence patients’ and their relatives’ perception of care, particularly when there are deficits (see Appendix 3).

FIGURE 3.

Patient and public involvement collaborative LEGO model of a ‘good’ nurse manager. This represents that nurse managers need to have good communication skills (represented by the bridge and chain) and take a high-level view of staffing needs. They should have compassion, courage and the leadership skills required for safe staffing. They need to value people, develop good ideas and remember the past (this is represented by being caught in the web).

Semistructured telephone interviews

Ten interviews were audio-recorded and built on the co-production workshops to explore variations in workforce planning systems across organisational settings and health services. The interview schedule consisted of 11 key themes, structured around the theory areas from the co-production workshops and the initial scoping review of the literature (see the interview material document at www.journalslibrary.nihr.ac.uk/programmes/hsdr/1419420/#/). Several themes intended to build an understanding of how WPTs are assumed to work, the impact of WPTs in different health-care settings and their predictive reliability. Interviews lasted approximately 30 minutes. Participants were recruited from NHS organisations from three sites, two of which were in large cities and one of which was in a smaller city within a rural area. Table 2 provides details on participants’ roles. An e-mail invitation to take part in the study was sent out to potential participants with a participant information sheet (see www.journalslibrary.nihr.ac.uk/programmes/hsdr/1419420/#/). Written consent was sought prior to participation (see www.journalslibrary.nihr.ac.uk/programmes/hsdr/1419420/#/). The data from the interviews further contributed to the emerging specifications of the systems of nurse workforce planning and deployment and a functional typology of WPTs.

Identification of theory areas

Based on the information from the stakeholder work gathered in phase 1, a complex, multifaceted system of nurse workforce planning and deployment emerged with explanations of the role and functions of WPTs. From this, the project team identified eight theory areas for the main searches of this evidence synthesis; these were operationally defined (see Appendix 4). These theory areas comprised the elements described by stakeholders as having an important influence on NHS managers’ use of WPTs within the system of nurse workforce planning and deployment.

Purposive searches

In total, 11 iterative searches were conducted. These evolved as the programme theory emerged from the review of the literature (Figure 4). For each search, bibliographic databases that were relevant to the topic were selected, and searches were translated for each database to use relevant keywords and make use of available thesauri. Search terms were identified from early scoping work, theory development and stakeholder engagement, and were built on and developed iteratively as common terms were identified from the selected literature (see Appendix 5).

FIGURE 4.

The iterative search process selection and appraisal of documents. ASSIA, Applied Social Sciences Index and Abstracts; DARE, Database of Abstracts of Reviews of Effects; HTA, Health Technology Assessment; NHSI, National Health Service Improvement; QNI, Queen's Nursing Institute; RCNI, Royal College of Nursing, (RCNi) is a wholly owned subsidiary of the Royal College of Nursing. a, Google Inc., Mountain View, CA, USA.

During the early scoping searches, a broad range of databases were used including The Cochrane Library, NHS Centre for Reviews and Dissemination, the Database of Abstracts of Reviews of Effects, the Health Technology Assessment database, PubMed, MEDLINE via EBSCOhost (1950 to March 2017) and CINAHL. Later, the purposive searches on nursing managers’ use of workforce tools and staffing management concentrated on the use of the MEDLINE via EBSCOhost (1950 to March 2017), CINAHL and PubMed databases. Additional specific databases were added for the searches of social care staffing management and for the police sector.

Only one search limit was applied: a date limit of 1983 to the March 2017. In 1983, the NHS Management Inquiry was commissioned to evaluate methods of estimating staffing levels and the classification of workload analysis approaches. 139 Backward reference searching was used to identify any works of relevance to this study that were cited in the selected articles. Forward reference searching was used to identify any newer works of relevance to this study that cited the selected articles.

Initially, the study sought to identify any studies that evaluated the effectiveness of nursing staffing models in the UK (iterative search 1). As the programme theory developed, the literature searches were designed around the complexity of nurse staffing (iterative search 3), a number of searches focused on the nurse managers themselves and how they make decisions (iterative search 4) and take risks (iterative search 8) and the support available to them (iterative search 9). Iterative search 5 was designed to focus on the practicalities of allocating staff to staffing structures, and iterative search 6 was designed to gather evidence on the different tools and information technology (IT) used for staffing management. Iterative search 7 was designed to capture any evidence on how information about the capacity and capability of staff is considered when allocating shifts and planning a rota (see Appendix 6 for an example of a search strategy).

Iterative search 2 focused on the grey literature, such as workforce planning project reports relating to national and local initiatives and evaluative information about these initiatives that is held in the public domain. The project team searched examples of development programmes across multiple services (adult, mental health, learning disability and community) on the evaluation of the use of WPTs. The grey literature search also explored policy, evidence to government, NHS organisational strategies and the reports of workforce technology projects and programmes. None of the pieces of evidence identified in this search was included in the final synthesis; however, the review of the grey literature informed the team’s understanding of the system’s complexity and highlighted contextual influences on the implementation and impact of WPTs.

In order to investigate WPTs in use in the social care sector, a search was designed for the Applied Social Sciences Index and Abstracts database on the ProQuest platform. For evidence from the police sector, the National Police Library at the College of Policing was contacted and the librarian provided the team with some references from a search of their library catalogue. This was also supplemented with a search of Google Scholar (Google Inc., Mountain View, CA, USA) and the databases available via the Bangor University ProQuest subscription.

Inclusion and exclusion criteria

A systematic process developed by the team in a recently completed realist synthesis was followed to determining relevance. 89 Consistent with Pawson’s suggestion,94 the test for inclusion involved:

• linkage with the programme theory and explanatory potential

• discernible ‘nuggets’ of evidence within the source material

• evidence of trustworthiness.

Reports of WPTs, such as workforce planning, workforce measurement, workforce management, patient acuity and dependency, staffing ratios, professional judgement and skill mix, were included. Evidence on settings was also searched, recognising the shifting patterns of health care and the importance of enabling patient flow and quality across systems of care. As the programme theory developed, a functional typology of WPTs used in safe staffing was developed to encompass tools that (1) help to summarise and aggregate information, (2) aid communication and (3) support standard setting and quality assessment (see Table 4).

In a realist synthesis, evidence is excluded only if it does not relate to, or inform the development of, the programme theory; however, in this review, evidence was not included that had limited transferability to the NHS, such as nursing workforce issues within low-income countries. Evidence was included only if it was generated from different international contexts in comparable health systems in high-income countries. Discrepancies in opinions on the relevance of evidence were resolved through discussion among the project team. Title sifting was cross-checked across four team members (SD, AJ, LW and AM). Levels of agreement among reviewers were scored for 10% of the total titles.

Data extraction, analysis and synthesis process

This aspect of the review process is resource intensive and is reliant on discussion and deliberation, including consultation with a wider group of stakeholders. In a realist evidence synthesis, bespoke data extraction forms are developed to guide the decision-making process for extraction. 94 Based on the initial programme theory of WPTs, a bespoke extraction form was developed, based on the eight theory areas, to interrogate the literature and data were extracted only if the evidence met the test of relevance for the programme theory (see Appendix 7). The theory areas and their sub-elements, specified within the data extraction form, were transposed into ATLAS.ti code manager version 7.5.10 (ATLAS.ti Scientific Software Development GmbH, Berlin, Germany). A total of 90 sources from the literature were entered into ATLAS.ti for coding. Additional codes were also identified (see Appendix 8). The coding of a selection of included data was validated by three members of the project team (SD, CRB and AJ). Three sources of evidence were not able to be coded; therefore, the final selection was 87 sources of literature. This incorporated many types of published evidence, including qualitative interviews, quasi-experimental studies, surveys, evaluations and audits, case studies, frameworks and guidance, theoretical conceptualisations, literature and evidence reviews (see Appendix 9). Commentary evidence was included because it was felt that it was credible and critical to theory development, as it was provided by those directly engaging with WPTs in the field. Following data extraction, the eight theory areas guided the data analysis process. Within each of the eight theory areas and their coded sub-elements, the data were organised into themes (see Appendix 10). Appendix 11 offers an example of the thematic development within theory area 6.

The relationships between CMOs were analysed from the extracted information. This involved organising extracted data into evidence tables to represent the different bodies of literature. Using abduction and retroduction across the evidence themes,140 WPTs were reconceptualised from different angles to identify the underlying structures and emerging demiregularities (patterns) around plausible CMOs, seeking confirming and disconfirming evidence. Linking these demiregularities to develop the initial programme theory provided an explanation of the implementation, utilisation and impacts of nurse WPTs. The resultant seven hypotheses act as synthesised statements of findings around which a narrative can be developed that summarises the nature of the CMO links and the specific characteristics of the evidence underpinning them.

Advisory group

The initial CMOs were developed by two members of the project team (CRB and LW) and were reviewed extensively in an advisory group meeting in May 2017. This process sought to clarify, develop and refine the CMOs.

Think-aloud interviews

To further enhance the trustworthiness of the resultant hypotheses and to facilitate the development of a final review narrative, 11 audio-recorded telephone interviews were conducted with NHS managers between July and August 2017. Participants were purposively sampled to obtain different perspectives that were relevant to the review question, including different national contexts and service settings (Table 3). Participants were recruited from four sites (three large cities and one smaller city surrounding a rural area). Interviews lasted approximately 30 minutes. In addition to exploring the synthesis findings, the interviews included a think-aloud technique, whereby NHS managers were given information and asked to describe their ‘thinking work’ around a nurse staffing scenario, validated by the project advisory group (see the interview material document at www.journalslibrary.nihr.ac.uk/programmes/hsdr/1419420/#/). This is a recognised technique for determining the information behind decision-making and judgements,141 which was considered to be a key focus for the operationalisation of the emerging programme theory. The technique illuminates the cues, priorities and strategies that are being considered at a particular point in time. It can identify the rationale and inference being drawn in that moment and can be used to test hypotheses;142 specifically, it was used to test and refine the seven CMOs.

WeNurses Twitter chat

A Wechat#WeNurses Twitter chat was held in July 2017 to capture the stakeholder perspective on the use of tools and technology for nurse staffing deployment. This opportunity facilitated the refinement of the understanding of the complexity of NHS managers’ roles in managing the staffing resources day by day and tested the CMOs (see Appendix 12).

Changes to the protocol

Amendments to the protocol were granted in phase 3 of the review. These amendments related to the process of audio-recording verbal consent for telephone interviews with NHS managers in place of written consent and changes to the interview process for the second stage of the telephone interviews, in which the submitted interview schedule was replaced by a think-aloud scenario.

Ethics approval

Ethics approval was granted by the Bangor University Healthcare and Medical Sciences Academic Ethics Committee to conduct up to 10 semistructured telephone interviews in phase 1. Following amendments to the protocol, the team also gained ethics approval for the think-aloud interview technique, to seek audio-recorded verbal consent in phase 3 and to increase the number of interview participants.

Understanding the system

NHS managers were aware that they were working within a complex system and that they needed to understand its demands and resources. The delivery of patient safety and good-quality care was their key priority, underpinned by the evidence base for safe staffing and focused by national standards for good-quality care. NHS managers acknowledged the increasing pressures from expanding demand and financial constraints. They understood the requirements to ‘flex up’ (increase staffing levels) because of unscheduled care and the numbers of patients generally, with higher acuity and complex needs:

So, my remit is to make sure that they [nurses] escalate that to the right people, to the right agendas, so that procedurally, when there are unscheduled care pressures, everybody has to change the pace of work not just, operational nursing, medical or therapy staff.

Site 1, participant 3

NHS managers discussed how the system also had to contend with a reduction in RNs, as nurses were retiring but not being replaced in sufficient numbers. This often prompted the dilution of skill mix through substitution with non-registered staff. NHS managers acknowledged the development of new roles, such as the associate nurse role in England, and, in some instances, welcomed this as a new way of potentially delivering good-quality care:

Everybody knows the staffing pressures in the organisation and the NHS as a whole. We know that we are working tremendously hard to improve those staffing numbers, and equally we’re looking at diverse roles as well, at trying to support . . . so thinking outside the box really.

Site 3, participant 3

The NHS managers’ work in safe staffing

NHS managers understood that their role with regard to staffing deployment is complex. They were conscious of their responsibility as decision-makers and acknowledged that this could sometimes be stressful, as they were accountable for balancing risk and safety:

What we can see is where we have increases in personal demand, we have reductions in core staffing, what you find then is . . . that . . . potentially even an increase in throughput, what you find then is there’s decision-making going on which is to increase staff, close the intake, move some things through, alter some activity, and it’s that decision-making which is really the thing that creates the level of safety.

Site 1, participant 4

NHS managers felt that they needed to be comfortable in justifying their staffing decisions in the light of patient requirements. They understood patients’ clinical needs and anticipated how these related to staffing requirements. They discussed their local knowledge of the organisation, clinical areas and staff capacity and capability:

So it’s important to know your staff really . . . and what skills they have, and how, probably, they sometimes work with each other.

Site 3, participant 1

Teamwork and planned meetings to determine demand and resource allocation, for example, bed meetings, were features of the nurse manager’s staffing work:

Staffing is always discussed at the end of bed meetings as well, just to say look, this is where . . . from an operational point of view, this is where our pressure points are, have we got the staff in place to support these? So, it’s knitted together very, very tightly.

Site 3, participant 3

The importance of communication

You’ve got to juggle the staff, you’ve got to juggle the patients, and you’ve got to juggle upward and downward messages, you know?

Site 2, participant 2

Communication with a range of staff, and for the escalation of concerns to senior managers, was an essential factor in safe staffing. NHS managers at all levels (e.g. matrons, lead nurses) had a physical presence in clinical areas, although for some lead nurses this was less often. This was to liaise with staff, make a first-hand assessment of the situation and assess the nurses’ ability to cope. This presence also enabled them to offer support and determine if redeployment was required:

And you can get the feel of the place . . . And it does take some footfall, so I’ll walk 15,000 steps on a late shift. But then I know that everyone is safe. And they appreciate it, they appreciate you going in and saying hello.

Site 3, participant 2

Many participants felt that they offered an independent, objective view to balance more emotional responses to system pressures. They had to understand the environment and culture to implement the staffing decisions they made, particularly if they were asking ‘difficult questions’ or negotiating the redeployment of staff. They related their own management skills to their experience:

Probably what I’ve got better at is at asking people to move and encouraging them to move, and . . . being able to charm people are probably the things I’ve got better at, rather than being able to ascertain where needs more support.

Site 3, participant 2

Valuing nurses

There was an overwhelming sense that nurses should be valued, respected and supported. Maintaining staff satisfaction was important, but this had to be balanced against patient needs. NHS managers wanted to ensure that junior staff were developed, particularly if they were in management roles. Staff development was also an important part of recruitment and retention:

I’ve started some forums for nursing assistants. I’ve started forums for junior staff nurses. I’m starting a band 6 development programme; starting surgical skills training. Since I’ve got to the division I’m on a mission to . . . I want to keep them engaged, want to keep them interested.

Site 3, participant 2

Patient and public satisfaction

NHS managers understood the importance of listening to staff and patient concerns, particularly in dealing with patients and the public when complaints were linked to staffing. Here, focused communication was used to determine a satisfactory outcome; often this meant managing expectations and setting realistic goals:

Because, unfortunately, that’s where a number of complaints come from, because (1) we haven’t communicated properly with families, and (2) we just haven’t had the right staffing infrastructure in place to deliver on what was expected from that family.

Site 3, participant 3

We try and avoid complaints getting to a formal stage, so at the point of a family complaining, going to X [formal complaints system], either the ward manager or the matron would go in and meet with the family and try and resolve those issues.

Site 3, participant 4

Seeing the bigger picture

NHS managers set great importance on seeking out appropriate information about demand and resources within the organisation as a whole; this was used alongside their professional judgement for effective staffing decisions:

So, the information goes to the senior management team for the hospital and the Trust so that they’ve got an understanding of what’s happening within the hospital, as well as whoever the on-call manager may be, and then the department managers as well also receive a copy of that. So, we’re going to the computer three times a day to input that information anyway and to send out.

Site 3, participant 4

The use of workforce planning and deployment tools and technologies

A range of components were considered for staffing decisions, for example the bed state, patient acuity levels and the number of nurses. NHS managers valued WPTs if these made data on patient need and resources transparent for decision-making; for example, WPTs were useful when they could match patient acuity and dependency data (e.g. early warning scores) with staffing requirements. In some instances, they could track both patients and nurses through the system. WPTs linked important components of staffing data, such as annual leave, sickness and payroll. They facilitated comparisons of costs and budgets. Some rostering technology also provided rules on, for example, skill mix and numbers on annual leave:

It [electronic rostering] does bring an awful lot of benefits around HR [human resource] management, and just having all the information in one place about your staff, annual leave management, time owing management, it’s all so much better . . . and completely traceable; whereas before it was quite difficult to manage.

Site 3, participant 4

Senior managers scrutinised data from WPTs for evaluation, to formulate evidence for establishment setting. Comparison and benchmarking enabled learning from other areas. Data could also be used for reviewing staffing levels where there had been complaints or clinical incident reporting. WPTs enhanced communication in many instances by making needs and resources visible:

And then that gets uploaded into a report, which is very clever and it allows us to view fill rate, overall fill rate, how many agency staff we’ve got on duty at any one time. And then it picks up, it allows us to pick up comments that the wards give us. So, it’s real time, it’s ward to board, anyone in the Trust can access it . . . from the Chief Exec down to a nursing . . . HCA [health-care assistant] on a ward.

Site 2, participant 1

However, some NHS managers indicated that WPTs were not used to full capacity for evaluation and suggested that this was an area for development. Senior NHS managers also found that technology highlighted whether or not managers were compliant and engaged with electronic staffing tools. NHS managers also had a role in encouraging their staff to engage with WPTs to ensure compliance with data input.

Real-time data

NHS managers generally embraced the use of WPTs, but only when they felt that the data were credible and reliable. Data were valued if these were real-time data. NHS managers often valued data from personal communication above data from WPTs and used paper records or computer printouts of current data on staffing and patient requirements, admissions and discharges, but checked these for accuracy by walking around the wards. They suggested that occasionally they were frustrated with technological difficulties with WPTs and how inputting data could be time-consuming. Some senior NHS managers reflected that it had been difficult to embed their use into practice, as other nurse managers were initially reluctant to engage and learn the skills required. In some instances, data had to be accessed in terminals away from the clinical area; NHS managers valued data that could be accessed anywhere, particularly via handheld devices.

The NHS manager’s role in facilitating the use of workforce planning and deployment tools and technologies

NHS managers recognised that they had a leadership role to engage staff to utilise WPTs effectively, and to develop the nursing staff’s skills for effective decision-making. Many felt that this was about setting standards and indicating expectations. This was particularly noted in ensuring that data input was a true reflection of the staffing requirements. Some participants suggested that it was important to liaise with nurses in areas where WPTs were to be introduced, so that they were user-friendly. Organisational support was important for the provision of dedicated staff with IT and nursing expertise to lead on training and education. This needed to be readily available, either virtually or in person. In addition, in order to ensure effective nurse development, all NHS managers reflected that they needed to know their staff and identify gaps in their skill set, via performance review. Nurturing staff was a key theme. This was facilitated through spending time with them, role modelling, mentoring, coaching and promoting access to leadership programmes and peer support:

So it took us about 6 months, really, to get people confident in the work of the systems, and just working with them while they were working the systems, just to give them the confidence and the knowledge.

Site 1, participant 3

Context and mechanism

The complexity of establishing staffing requirements has been characterised as subjective, contentious, political and time dependent. 143

It appears that, when NHS managers are encouraged to be responsive to staffing demands, it is vital that they understand the different levels of the staffing system and how proximal and distal aspects are in synergy. When a complex staffing system is working well because it responds effectively to real-time demand (proximal factor), this can create a positive reinforcement of distal aspects, such as upholding organisational reputation and enhancing recruitment (distal factors), creating a positive cycle so that nurses can be matched to demand. For responsive decision-making, NHS managers need to understand the essential factors of the staffing system and the interactions between them. Certainly, knowledge of the clinical environment, patient needs, organisational structures and processes is important for effective rostering and patient–nurse allocation. 129,131,144–152 Wilson and Weiss153 indicated similarities in police staffing and suggested that staffing must also account for broad local factors, such as geography and organisational changes. The NHS manager stakeholder group feedback indicated that there is little control of the work coming in, but that it is important to understand the system as a whole and recognise different needs.

Organisational factors and patient flow

Understanding the staffing system’s organisational processes and context is essential for NHS managers’ effective responsiveness to real-time and future need. The synthesis of the literature indicated how various organisation process factors had an impact on staffing resources and care delivery. 124,128,131,143,145,149,154–162 For example, nurse managers need knowledge of patient procedures and hospital/unit policies. 149 Nurse managers also need an understanding of specific organisational processes and organisation of work, for decisions on staffing. 161 These may be unique to each clinical area. Berkow et al. 152 assessed Chief Nursing Officers’ (CNOs’) staffing decisions in four unit types (paediatrics, the emergency department, general medicine and intermediate care) and found that CNOs from similar hospitals prioritised staffing for clinically similar units differently, suggesting that there are clinical and operational differences between similar organisations. 152

The NHS manager stakeholder group feedback highlighted that managers need to understand the system and how to effectively and efficiently move people through it. NHS Education for Scotland gave the example of knowing how many elective cases the ward area will receive and indicated the importance of attending to local context; including the local models of care, the turnover/throughput and the ward climate. 131 The advisory group emphasised the importance of being able to respond to planned and unplanned care. A key factor is admissions and discharges to ensure that there are enough beds. 131,145,151,161,163 The allocation of staff based on organisational processes is also related to the time and day of the week. 158 For example, Kontio et al. 151 found that managers in a cardiac unit made staffing decisions that were dependent on task prioritisation related to booked and emergency procedures; these were also based on the peaks and bottlenecks in the system.

Harris and McGillis Hall133 emphasised the importance of consideration of nursing structures, processes and nursing roles for staffing decisions. Decisions on skill mix should be determined not only in relation to patient need but also in the light of a ward’s or unit’s objectives. 131 Nurse licensure, which is linked to competence and the level of support staff, is also an important consideration for nurse-to-patient allocation for nurse managers. 158,164 Decisions about staffing need to relate the complexities of care to staff competencies, plus support within the environment. 144 However, Flynn and McKeown165 suggested that nurse managers believe that an appropriate skill mix has a high ratio of RNs to health-care assistants. Berkow et al. 152 found that RN experience and nurse-to-patient ratios were the most important factors for CNOs in units characterised by high rates of admission, discharge and transfer. 152

Evidence from the synthesis of the literature therefore emphasised the importance of responsive decision-making to ensure that staffing meets workload, as a result of patient flow. 151,156,158 This suggested synergy between proximal parts of the system, including the flow of patients, staffing and the nature of the work and its organisation. For example, Allen158 illustrated how this differed between clinical areas and offered the example of episodic care units in which task-orientated nursing work fitted with workload completion. In these contexts, workflow maintenance was only identified in episodic care units, reflecting the high volume of patients that are seen each day. A focus on making nurse-patient assignments that assist in the movement of patients through the unit is required.

Data for in-the-moment decision-making

When NHS managers understand the complexity and dynamicity of the staffing system, they can use data from WPTs for responsive decision-making and to enable better forecasting. Interview participants and evidence from the synthesis of the literature emphasised the importance of real-time data for decision-making on staffing:155,156,166–170

Having a real-time tool to stay on track on a daily and even hourly basis also makes a big difference in how well staffing matches demand.

Barton168

Evidence strongly suggested that WPTs can provide real-time data for decision-making. 151,161,168,169,171,172 Interview participants also indicated that real-time staffing resources were made visible by WPTs, so that staff could be redeployed to meet patients’ needs:

So, if you come on and you’re a nurse short, . . . you can connect the dots and see where, and then you’ll e-mail that ward sister direct and say is there any chance I can borrow a member of staff. And it works quite well . . . but we do use it all the time, people always look at each other’s [staff levels].

Site 2, participant 1

However, evidence sources emphasised that WPTs are effective only if the data input is up to date:167,173,174

It is inevitable that changes will be made to the active roster and essential that these changes are made in ‘real time’. Real-time data maintenance of the IRS [integrated rostering system] facilitates Senior Nursing Managers to accurately identify actual staffing levels throughout the hospital.

p. 99, Kerr and Timony167

The NHS managers’ role is intrinsic to real-time data collection. Wilson et al. 150 found that managers were actively involved and engaged with the workflow activities, so they knew the staffing information status of the unit by checking on patients and staff nurses:

This behavior helped the effective charge nurses to be prepared for the next admissions, know which nurse can take the next patient, be aware of the acuity level of the patients, and comprehensively have a hand on the pulse of the unit.

Wilson et al. 150

There was strong evidence from the interviews that NHS managers also did ‘walk-arounds’ in hospital settings to check that the data from WPTs were up to date:

You would go on and you would find the centralised data with all of it on there, I would then print that off and I would then walk the wards and double check the numbers match up with what is on there.

Site 3, participant 2

One participant indicated how inputting data on staffing levels had become embedded practice:

The thing is, if somebody hasn’t done it another person will spot it straight away, they’ll say ‘Well it looks like you’ve got five today’, ‘Oh no, she’s off sick’. So, it tends to be a habit for people to keep it up to date.

Site 2, participant 1

Data from patient dependency/acuity classification tools or technology are vital for forecasting real-time staffing requirements. Many sources of evidence indicated that managers use these scoring tools to match up staffing effectively; these scores are linked to workload. 124,125,127,129,130,133,144,147,151,175–180 Nagy and Nathan181 were able to use scheduling technology to identify appropriately skilled nurses, as it articulated experience and credentials. Botter173 proposed that a patient classification system was useful to predict real-time and future staffing levels, particularly when decision-makers were familiar with the daily activities of the clinical area. Data may come from other sources; Diers et al. 154 found that nurse managers kept a ward document or ‘brain’, which detailed patient factors (e.g. leaving the ward for tests), and this information was not in the computerised system. These nurse managers could not be dissuaded from using this information source, so it was eventually encouraged alongside the tool. Interview participants also indicated how NHS managers used knowledge of patient flow and acuity to determine staffing need; this also prompted the redeployment of staff:

. . . it fluctuates every day. And then what we do is we offset that with acuity of the ward. So, for example, X ward may have four registered nurses on this morning, but actually they’ve not got that many sick patients today, they’ve got delayed discharges, so they don’t need four registered nurses. Whereas the diabetes ward, for example, actually they’ve had a lot of sick patients overnight, they need a bit more help, so let’s move a registered nurse from that ward to that ward to try and help them.

Site 3, participant 3

The NHS manager stakeholder groups highlighted that there is often a wealth of skills on site; they emphasised the importance of a flexible workforce, in which staff are happy to be redeployed and have a good understanding of how each ward/clinical area works. Interview participants indicated the importance of the provision of staff in accordance with extra need, for example one-to-one supervision, and how technology could be used to specify for codes to ensure that agency or bank nurses had specific skills in line with patient requirements; extra staff were then distributed in accordance with need:

So if we need to book agency or check shifts or whatever, we tend to use our [handheld] devices to do that. So, everybody will put up their agency shift on the device and know exactly what position they’re in at that moment in time. Because sometimes we do need agency, but sometimes . . . we’ll share the agency round.

Site 2, participant 1

So, that’s a person who’s been booked a special, and we try to do it out of our own staff, you know, so that the patient is familiar with the member of staff. So, we might book agency, but they might go and work on the floor . . .

Site 3, participant 3 (talking about the one-to-one supervision of patients with dementia)

The interviews therefore suggested that NHS managers use WPTs to identify immediate need and use their professional judgement to distribute staff for effective patient care. van Oostveen et al. 178 argued that a computerised decision support system can never fully replace professional insight and decisions may need to be adjusted by nurse managers.

Data for forecasting future demands

When NHS managers understand the complexities of the staffing system and how proximal and distal aspects inter-relate, WPTs can articulate future requirements. There was evidence of how WPTs could be utilised to forecast the need for long-term workforce planning. 168,182 Ruland171 indicated how WPTs can provide prognostic data. This can also enable trend analysis. 131,154 WPTs can also predict costs associated with their staffing decisions,134,151,171,172,174,183 or they can be useful for the development of ‘what-if’ staffing scenarios, to determine potential financial and patient outcome feedback. 134 Therefore, computerised decision support systems can:

. . . facilitate the evaluation of the impact, adequacy and relative benefits of alternative nurse organization processes and structures.

Maenhout and Vanhoucke184

Anderson et al. 134 illustrated this by indicating how the use of a dashboard meant that managers would be able to forecast:

. . . the impact of various staffing patterns on patient outcomes and costs rather than staffing to keep costs down on a shift-by-shift basis. Managers will be able to predict the ‘true’ costs associated with staffing decisions and their impact to the bottom line.

Anderson et al. 134

Workforce planning and deployment tools and technologies may therefore assess multiple dimensions of the staffing system to determine future need. van Oostveen et al. 124 suggested that the Professional Assessment Of Optimal Nursing Care Intensity Tool (which assesses the optimal nursing care intensity for a clinical area) can highlight non-patient factors that may affect workload (e.g. organisational issues). The NHS Education for Scotland toolkit131 suggested how to determine numbers and skill mix in the light of demand and budget to ensure good-quality care. It identified the importance of workload measurement, to assess activity, capacity, demand and capability, and highlighted the professional judgement approach, acknowledging subjective judgements from nurse leaders in particular areas. Furthermore, Dent130 suggested that part of staffing planning was a focus on recruitment and retention and highlighted the importance of incorporating fatigue management guidelines when forecasting staffing:

These guidelines provide a strategy that recognizes and manages the potential negative consequences of sleep deprivation and sustained work hours on patient outcomes and staff wellbeing models of care on nursing outcomes.

Dent130

Workforce planning and deployment tools and technologies effectively drive the management of staffing establishments, for example through rostering. 129,185 Barton168 described how an acuity application technology linked to national benchmarks was used to set overall staffing parameters by unit and helped to adjust skill mix. Kane179 used technology for caseload analysis in the community for skill mix and numbers. There is some indication that establishment setting using WPTs has to consider the local organisational needs of the staffing system. Local knowledge of the clinical area/care setting is also deemed to be important when implementing WPTs. 127,175,178 Fenton and Casey147 emphasised that the Safer Nursing Care tool must relate to local factors and be used in combination with nurses’ professional judgement. 147 Taylor et al. 127 illustrated this point:

Each unit must define its needs using standardized approaches, and learn to argue the importance of the recommendations that they, as experts in the micro-system in which they provide care, consider optimal levels of staffing for safe patient care.

Taylor et al. 127

When WPTs forecast future need, responsive decisions can result from this. One interview participant reflected how the application of the Safer Nursing Care tool identified an increase in patient dependency and acuity, and thus establishment figures were adjusted accordingly. Interview participants associated the achievement of staffing establishments with the provision of good-quality care, but noted the importance of maintaining this level:

Now when I’m at template it’s three trained on a long day, so they get that better quality, and, safety in an afternoon with the extra trained, so that we have benefitted there, but that’s fine while you’ve got (pause) all your vacancies filled, isn’t it?

Site 1, participant 1

The budgeted establishment includes an allowance for the fact that staff hours are not necessarily just to fulfil demand, and this is called headroom; this is usually 22–25% of the budgeted establishment. 129 Drake further explained how staffing establishments must allow for headroom or time out; this includes:

. . . additional staff to allow for staff unavailability due to annual leave, sickness, study leave, parenting, etc.

Drake143

Headroom may also factor in meal breaks and seasonal pressure. In addition, making provision for headroom is a key performance indicator in England. 129 Many authors have highlighted how headroom must be factored in for staffing allocation. 129,159,186 Local flexibility associated with this headroom is an important proximal aspect of the staffing system associated with staff satisfaction, but with plausible repercussions for distal aspects, such as recruitment and retention. Murtola et al. 186 indicated how headroom must be incorporated into technology. Drake143 noted how headroom allowances are often lost because of budget when there are nurse-to-patient ratios.

Outcomes

Workforce planning and deployment tools and technologies can be used by NHS managers to forecast real-time and future demands so that they can respond effectively to deploy staff. This can have a positive impact on nursing outcomes, when there is positive reinforcement of proximal and distal aspects of the staffing system. Nursing outcomes, such as nurse satisfaction, recruitment and retention, appear to be linked to the impact of organisational factors on staffing. The data from the synthesis of the literature mainly examine nursing outcomes within hospital settings, with some preliminary evidence on long-term care and a lack of evidence in primary care. 133 The evidence suggested that organisations need to be responsive to staffing needs in order to have a positive impact on staff outcomes. The structural features of the organisation, organisational processes and outcomes appear to have an impact on staff outcomes. 184 When there is reinforcement of responsiveness to staffing needs, as part of the proximal staffing system, this appears to have a positive impact on recruitment, retention and satisfaction. Fairbrother et al. 121 found reduced vacancy rates when the model of care was team nursing, although the results were not significant (U = 10.0; p = 0.20). Team-based models of care have also been linked to increased job satisfaction. 120,121 In addition, Magnet hospitals associated with high levels of nurse support increased nurse satisfaction. 119,123 Educational support appears to be fundamental for the facilitation of nurse satisfaction with organisational factors. ‘Train the trainer’ programmes also increased nurse levels of satisfaction with teamworking. 120 Furthermore, nurse managers’ responses appear to strongly influence the environment or climate of care to improve nurse satisfaction. 149,187,188

The nurse manager’s ability to creatively problem solve and develop relationships with staff can positively influence the practice environment and nurses’ job satisfaction.

Zori et al. 188

The interview participants noted that they strove to make clinical areas attractive to potential staff and invested in current staff, often through development and education programmes, to promote retention. Some participants reflected that models of nursing care are changing with the development of new roles, such as the nurse associate role in England, prompting staff to be more innovative in their approach to care provision and mitigate nurse dissatisfaction with staffing deployment, through developing and reviewing different roles.

Understanding the proximal and distal aspects of the staffing system meant that there had to be a balance between staff satisfaction and ensuring safe staffing. There was an understanding of the synergy between proximal and distal staffing systems so that nurses did not feel disempowered. Both the PPI and NHS manager stakeholder groups highlighted the importance of valuing and respecting staff with regard to staffing decisions. The NHS manager stakeholder groups and interview participants emphasised the importance of working with staff to ensure flexibility. One participant indicated the level of engagement work needed across one clinical directorate to respond to real-time and future needs for a flexible workforce:

Staff were very much ‘This is my ward and I’m not leaving it’. It took them a while. So, . . . to overcome that, we’ve done quite a lot of work. So, we did a lot of time-out days, right at the beginning, but we also did all our training together.

Site 2, participant 1

The NHS manager stakeholder groups and interview participants also emphasised how effective communication helped to prevent nurse dissatisfaction when redeployment was required:

We’re very lucky as a team that we’re constantly talking to the staff so we’re making them aware of the . . . you know, the organisational pressures, what . . . you know, we’re not moving these staff unnecessarily, there is clinical need and priority to have to move them. So, to be fair, they’re very, very good.

Site 3, participant 3

NHS managers’ proximal decision-making around staffing deployment may have a positive impact on nursing outcomes. Good nurse–patient allocation/assignments improve organisational reputation, making the hospital a better employer. 178 Allen158 indicated that an equitable nurse-to-patient workload decreases burnout, and this influences satisfaction and reduces the nursing shortage. Furthermore, the manager’s role in rostering may have a direct impact on hospital performance,185 and rostering can have an impact on staff welfare and satisfaction. 131,155,181 The quality of management decisions therefore appear to affect staff outcomes related to job satisfaction and satisfaction about care standards. 131

The interview participants strongly indicated how recruitment, rostering and nurse–patient allocation was a priority to ensure both good patient outcomes and good nurse outcomes:

Well I think if you get your staffing right you’re 99% there, aren’t you? . . . To me, the two most important things of my job is recruitment and rostering. Patient safety is embedded in those two things. So, if you can get those two basics right, you know, if you can get your recruitment right in the first place and get your rostering right, and make sure it feels safe, because half the battle is your staff feeling safe.

Site 2, participant 2

The impact of tools and technology on nursing outcomes

Successful data information management tools may affect staff satisfaction. 151 Pruinelli et al. 175 suggested that the Nursing Management Minimum Data Set (NMMDS) can result in improved patient and staff outcomes:

These data are instrumental in supporting effective work-force management practices and in developing effective staff recruitment and retention strategies.

Pruinelli et al. 175

Crist-Grundman and Mulrooney169 indicated that technology can have a positive impact on nurses’ work–life balance. Maenhout and Vanhoucke184 suggested that electronic nurse scheduling should avoid unhealthy shift patterns (e.g. too many consecutive nights or too few days off):

The quality of a nurse roster is also measured in terms of personnel job satisfaction which is related to the working atmosphere, the fairness among nurses, the compliance with nurses’ wishes and requests and the healthiness of nursing schedules.

Maenhout and Vanhoucke184

Self-rostering has emerged as a result of shortages in nurses and difficulties in recruitment. 157 Davidhizar et al. 189 suggested that staff satisfaction increases when staff members control the assignment/allocation structure. Self-rostering technology enables staff to see which shifts are available online and select suitable shifts to work. Both Valentine et al. 190 and Brown191 reported increased staff satisfaction, recruitment and retention with self-rostering technologies. They found that it improved staff satisfaction through offering flexibility and choice in staff schedules. Brown191 described this as ‘empowering’ staff. Valentine et al. 190 indicated how accessing shift availability online at home was useful, as nurses wanted to work in specific areas, usually their own unit; the results were positive, as improved staff satisfaction resulted in increased staff retention.

In relation to positive reinforcement of the distal staffing system, NHS manager stakeholder groups highlighted that organisational reputation is important for recruiting staff and for people within the organisation. All interview participants emphasised how the retention of staff and recruitment were important priorities and how WPTs articulated data for understanding:

We’ve been doing quite a lot of work in terms of retention and trying to understand what gives us our turnover, you know. We’re no different in terms of our turnover; we’re no different from any other organisations, any large organisations of the same size and amount of nursing staff. So, we have a turnover at 10% or 11%, you know, but given the numbers of staff that we do employ, that’s quite considerable every month, you know, or in a year, you know. So, we’re doing quite a lot of work to try and understand trajectories and what does that . . . the impact that that does have on morale and wellbeing of our staff, you know.

Site 2, participant 2

In addition, WPTs can improve operational efficiencies and reduce staffing and agency costs, with Brown191 reporting a reduction of 50% in agency costs and better-quality care. Junttila et al. 192 indicated how nursing satisfaction could be captured with technology to enhance decision-making on staffing. They note that measurements on job satisfaction are rarely stored in databases, and these data could form part of a nursing management information data set.

When there may be misalignment

Proximal aspects of the staffing system, particularly management responsiveness, can be adversely affected without real-time data. Sources of evidence indicated how a lack of real-time data had an adverse impact on decision-making on staffing. 151,159,162 Kontio et al. 151 discussed the outcome of a lack of data on participants in their study:

This information would have helped them to estimate patient flow and thereby allocate staff better. They also lacked information about the bottlenecks in the process; therefore, they were not able to manage the process in real time. Furthermore, information on patient priority was not available in real time.

Kontio et al. 151

Some interview participants also noted how data from WPTs were not always up to date:

. . . so if somebody’s phoned in sick overnight . . . it won’t have been updated so it won’t be correct. So, the ring round is really important. With the best will in the world, the data . . . is only as good as the people that are inputting it.

Site 3, participant 1

Interview participants also highlighted when poor data input may impede the use of WPTs:

. . . to be honest, it doesn’t matter how good your systems are, if you put rubbish in you’ll get rubbish out.

Site 2, participant 1

Furthermore, Hockley and Boyle166 pointed out that any reliance on manual data input carries risks arising from time pressures on staff, human error, or the deliberate manipulation of data. Other barriers to effective real-time forecasting were apparent. NHS managers have little control of the distal environment (e.g. workforce availability) in which staffing decisions are made, and this causes demand-driven roster changes. 143 Some evidence from the literature and interview participants noted that, rather than being able to respond to need, managers had to make do with the staffing resources available. 158,164 Rischbieth160 cautions that rapid decision-making with limited resources may result in adverse patient-, staff- or system-related incidents. Allen,158 on interviewing nurse managers, uncovered a degree of pragmatism associated with making the best use of available staffing resources against a sliding scale of the attainment of ‘best care’.

. . . when you’ve got wards on minimum numbers it can be quite challenging at times. And, like I say, the only impact I can do on that is to try and do the best I can with the staff I’ve got. I’m not a magician, I can’t create staff out of . . . do you know what I mean?

Site 2, participant 2

Poor rostering practice can lead to staff dissatisfaction, absenteeism and a high level of nurse turnover. 143,151,156,163,178,185,189,193 Changes in the stability of nursing personnel, skill mix or decreasing nursing numbers may also affect patient outcomes:133

. . . poor rostering decision making can deleteriously affect the key performance areas of service delivery both clinical care and the patient experience), resource utilization, and staff morale and retention.

Silvestro and Silvestro185

Davidhizar et al. 189 indicated how changes in shift patterns or length, and moving staff to another area when there are staff shortages, increase unsafe patient care conditions. They indicate that the requirement placed on nurses to work shifts and unsocial hours is a major contributor to job dissatisfaction and burnout. Low RN ratios may have an impact on satisfaction levels. Tran et al. 122 found few differences in satisfaction levels with nursing care delivery models, but satisfaction with co-workers was reduced in a shared care in nursing model with increased proportions of second-level RNs and nursing assistants. van Oostveen et al. 132 found that staff dissatisfaction was linked to a desire for more autonomy and perceptions of a lack of authority. This may also be related to shift patterns:

Nurses are particularly frustrated when they feel a lack of input into decisions about their work and their schedules.

Davidhizar et al. 189

Drake162 emphasised that requests can be rejected if they conflict with the needs of the ward, but equally found that when a large number of requests were assigned to the roster before other duties, this reduced the number of consecutive shifts available, making rostering some shift patterns more difficult.

In addition, allocation does not always take into account nurses’ preferences for specific areas; many nursers consider themselves to be specialists despite the assumptions of others that ‘a nurse is a nurse’. 189 Interview data indicated that the redeployment of staff to cover for sickness and absence is a source of nurse dissatisfaction, particularly when staff are moved to different specialties that are short-staffed. Staff redeployment to cover staff shortages worsened the cycle of staffing deficits:

There is no doubt that there is some kind of link between the number of staff moves that occur and the sickness that is happening, because . . . staff are employed to work with one team, and they like to work with one team . . .

Site 3, participant 3

van Oostveen et al. 178 found that nurses were less satisfied about their patient assignment following the implementation of a computerised decision support system tool, as the choice of allocation to patient was lessened. However, Drake,162 on reviewing an electronic roster system, found that many areas avoided using an auto-roster in order to honour staffing requests. One interview participant indicated that the technology could be developed further, for example to incorporate agency into e-rostering or a staffing skills function.

Poor resource management may lead to the inefficient use of resources, such as overstaffing of the ward, not using staff members’ full contracted hours or mismanaging annual leave so that shifts are not covered;185,194 it may also result in excessive bank and agency costs. 171,185,193 Furthermore, poor management may have an adverse impact on organisational reputation, with a negative effect on recruitment. 185 Certainly, poor nurse satisfaction and morale appear to lead to high rates of nursing turnover134,185 and absenteeism. 185,193 There are organisational costs associated with high rates of nursing turnover. 144 Subirana et al. 128 proposed that in their realist logic model of nurse staffing and nurse outcomes, nurses’ perception of staffing inadequacy can lead to emotional exhaustion, burnout, job dissatisfaction and poor retention. High rates of turnover and absenteeism are associated with inadequate staffing. 134 Poor perceptions of staffing levels also reduced staff morale in police organisations. 153 However, Flynn and McKeown165 suggested that there is a small evidence base for staffing levels and adverse nurse outcomes.

Many stakeholders and all of the interview participants acknowledged problems in the distal elements of the system, which affected recruitment and retention. In semi-rural areas, nurses were unlikely to apply from outside the area; conversely, in urban settings within small geographical areas, there were high levels of competition for available staff, with several organisations trying to attract staff to work for them. Participants did not anticipate improvements in staffing levels:

But we still have vacancies; we still have staff turning over. You know yourself, the retirement age . . . particularly, for example, district nursing, we know that almost half the workforce is going to retire in the next 5 years.

Participant 3, site 3

One participant noted that retention had an impact on recruitment:

And there’s a correlation on wards that have higher amount of vacancies and higher turnover, so you know, you’d argue what comes first, it’s chicken and egg stuff isn’t it, you know? But we do have some wards that are unable to retain staff because of the vacancy rate, you know.

Site 2, participant 2

When recruitment and retention is a problem, the effectiveness of WPTs may be hampered. Taylor et al. 127 found that the staffing methodology recommendations for staffing and filling vacancies could not be realised because of recruitment and retention problems.

Data to negotiate for change

Where and when the alignment of organisational strategies occurs, nurse managers are able to use the data from WPTs to demonstrate the reality of practice in comparison with strategic expectations. 126,127,129,131,144,169,192 These staffing data enable NHS managers to advocate and negotiate for a change in systems or resources, to ensure that staffing deployment achieves the organisational goals. The advisory group highlighted that, within organisations, many people can influence staffing decisions, such as doctors, financial managers and bed managers, and these powerful others may adversely affect nursing decisions. Eduardo et al. 148 indicated that nurse managers have to confront organisational barriers, owing to conflicting political goals from disparate groups:

The major forms of political intervention in decision-making arise from hierarchical power, financial control needed to implement the strategic decisions, and the influence on and the power of veto.

Reproduced from Eduardo et al. 148 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See http://creativecommons.org/licenses/by/4.0/

The evidence suggested that WPTs may empower NHS managers, as they provide additional weight to managers’ arguments to justify changes to staffing. 129,161,168,179 Eduardo et al. 148 suggested that decision-makers respond to policy and strategic aims. Therefore, it would appear that when WPTs provide data for the achievement of policy and strategy, they confer leverage to the NHS managers’ opinions. One manager with a strategic remit indicated what he had observed when NHS managers came to understand the utility of data from WPTs:

But it’s something else, you know, when you talk about it, it’s actually quite . . . quite invigorating, when you see a group of ward managers spend 3 hours together going through data and making decisions to change how they’re doing things tomorrow.

Site 1, participant 4

NHS managers need data for different purposes (e.g. strategic, financial). 180 Murtola et al. 186 indicated how managerial decision-making in hospitals occurs at strategic, tactical and operational levels. The strategic level is concerned with long-term goals and the tactical level is concerned with the achievement of short-term gaols, such as resource distribution, operational concerns and daily activities. Murtola et al. 186 suggested that nurse managers are generally involved in tactical and operational decisions. The evidence from the data also reflected that there are different information needs at different levels within the organisation. 131,180,182,186,197 Kivinen and Lammintakanen182 found the need for multifaceted information and more detailed information for different user groups with a management information system (MIS).

The interview participants also articulated the need for different information depending on their role. Ward managers and matrons were interested in real-time and retrospective staffing numbers in relation to patient acuity and dependency; they had a strong awareness of the financial data related to staffing in their specific areas. Matrons and lead nurses also reviewed this information, but from a broader perspective; in addition, they needed to link these data to other strategic areas, such as patient experience and workforce retention. This reflected the findings of Murtola et al. 186 that upper levels of management are generally interested in information about the quality of care and costs, with middle managers needing information about care processes and resources.

Workforce planning and deployment tools and technologies provide data to justify and rationalise the utilisation of resources and to convey the reality of managing the service related to resource availability. 131,144,161 The data from WPTs may make nursing care visible, providing objective evidence of staffing needs and how staff may best be deployed throughout the organisation. 168,179,192 The interview participants indicated how negotiation for tactical and operational decisions for changes, to ensure safe staffing in the short term, happen on a day-to-day basis, particularly the need to move staff. When using data from WPTs, interview participants indicated that they use their professional judgement and skills to negotiate staff moves:

Well obviously I negotiate it, we work as a team on the ward so moving people around on the ward if need be is not a problem.

Site 3, participant 4

The evidence suggested that data from WPTs can be used to achieve long-term strategic goals in quality care. 168,192 Junttila et al. 192 indicated how information from different tools within a nursing MIS was found:

. . . to be useful in redefining care profile of ward units and reallocation of nursing resources.

Junttila et al. 192

Barton et al. 168 discussed how one organisation used technology to determine staffing numbers and patient acuity, which provided data to set staffing parameters linked to national standards; this enabled the adjustment of staff numbers and skill mix, in alignment with budgetary information:

These data allow us to ensure each unit’s direct care staff stays right-sized during the course of the year, and it helps us justify the need for additional nurses.

Barton168

The interview participants also indicated that data from WPTs provided evidence for change to achieve strategic goals on good-quality care:

It provides evidence and backup that is so needed, that is saying actually, ‘look, this care of the elderly ward had this number of dependency 12 months ago, over a period of 12 months we’ve gone to this number of dependency’. So, we know that we’ve got to review the establishment of this ward because there’s been such a significant shift.

Site 3, participant 3

Therefore, through the use of WPTs, and the data they collate, NHS managers may be empowered to make changes or make the case for changes. van Oostveen et al. 132 found that the data from nursing workload measurement tools:

. . . creates possibilities for better positioning of nurses in hospitals and further professionalization in terms of enhanced authority and autonomy.

van Oostveen et al. 132

Kane179 found that team leaders who engaged with the Electronic Caseload Analysis Tool (eCAT) started to challenge the caseload data, as the tool enabled an objective and systematic review of equity within caseload allocation. On introducing a staffing technology that integrated patient acuity with staffing availability, Barton found that:

Nurse managers were hungry for data, and when they saw how the information generated by the system could help them improve staffing, they became enthusiastic supporters.

Barton168

Therefore, data from WPTs, such as rosters, can influence change129,164 through the reallocation of resources,169,192 and redefining care profiles of clinical areas. 192 Several interview participants commented on the need for innovation, as the functionality of existing staffing systems did not reflect the reality of staffing availability. The development of new roles was a key area for change:

Obviously, we’ve recently commenced the nursing associate role, and we also have assistant practitioners on the wards. So, it was about kind of thinking about your nursing team differently, what skills can other people bring into that team, rather than the old, traditional registered nurse, non-registered nurse.

Site 3, participant 3

Outcomes

Many sources of evidence linked staffing to quality and safety. 128,132,133,158,161,165,166,176,180,184,198 Effken et al. 145 indicated the challenges that nurse managers face in solving problems related to patient safety, quality and satisfaction and finance and staffing. All stakeholder feedback strongly emphasised that patient-centred care is the priority. When there is alignment of strategies in support of safe staffing, this facilitates the nurse managers’ use of data from WPTs to make an impact on patient care quality and safety outcomes:131,132,149,159,187

At the end of the day, if the ward needs it, we have to have it and I can justify why I used, why I spent it.

Site 1, participant 1

When there may be misalignment

Information is power, but only if there’s a process in place to interpret the data and then make policy and procedure decisions based on the evidence the system provides.

Crist-Grundman and Mulrooney169

The evidence highlighted areas in which there was misalignment between financial and other organisational strategies related to budget restrictions; this had the potential to adversely affect the effective use of WPTs;126,127,160 this was also the case for police staffing. 153

Robinson et al. 126 and Taylor et al. 127 reviewed the implementation of the Veterans Health Administration’s mandated staffing methodology. They categorised organisations on their implementation (by scoring on integration with budget, implementation of directive elements, engagement of leadership and use of data to support the process) and found that organisations with a low level of implementation did not have the flexibility within the budget to employ the recommended increase in staffing numbers. Conversely, if the budget provision is not sufficient to staff a unit as required then units could go over budget. 129 Given the strong links between staffing and patient outcomes, misalignment between financial and other organisational strategies may present:

. . . a danger that bottom-line fiscal improvements may occur at the expense of poor patient outcome in the longer term, thereby negating the presumed cost efficiencies.

Rischbieth160

Misalignment of financial strategy with staffing requirements was also articulated in the interviews:

And we get frustration from the ward managers, and they know what to do, but they have to go through in their own organisations to get . . . additional staffing, or to provide a business case for replacing the nurse that they’ve always had in their establishment who’s currently vacant. It takes 12 months to get a new nurse in post after one leaves. So, all these kinds of things, the organisations are almost set up to do exactly the opposite thing from what the ward managers know, but can’t quite articulate.

Site 1, participant 1

Furthermore, Murtola et al. 186 suggested that nurse managers are generally involved in tactical and operational decisions; however, most WPTs focus on strategic and tactical decision-making, with little evidence of support for operational decision-making. 186 Possibly, WPTs do support strategy when it is aligned, but do little to help articulate day-to-day operational requirements. Finally, when strategic requirements are ‘top down’, this may impede the implementation of WPTs because of a lack of ownership:

I think that maybe the deployment stuff, for me, I think having things that are thrust upon us from a governmental point . . . you know, so things like, say, staffing was . . . was very much thrust upon us, and so I don’t know whether the embedding, or the implementation was as nurse-driven as it should have been, and I think if you have things that are more driven . . . that are all accepted and better driven because of need, not just because we have been told to, I think we would have got better engagement, and I think that’s been a challenge.

Site 3, participant 2

Support and skills development

Several sources of evidence indicated the importance of the investment of time and resources to support nurse managers to successfully use WPTs, for effective and efficient staffing deployment decisions. 126,127,134,162,174,199 Examples that demonstrated this investment, and which may support NHS managers to develop the knowledge and skills required to use WPTs, are summarised in Box 3.

It is important for managers to feel that they are supported in their decision-making on staffing deployment. 128,163 The evidence showed that support for managers in their decisions on safe staffing was visible in multiple ways. The evidence suggested that a key method of support was the presence of a dedicated support team. 124,126,154,179,191,194,200 This team could be strategic, encompassing managers and finance and information specialists. 174,195,199 A support team could also provide more practical advice on WPTs. Junttila et al. 192 highlighted the need for an information analyst with clinical expertise to produce reports, identify trends and recognise new information needs. Many interview participants commented on the helpful role of e-roster support teams; often this was online or via telephone:

I think the organisation invested very well into e-rostering, it had a good team, a good strong team putting it in, and they’ve sustained it by keeping a good helpdesk and a good turnover. If you’ve got a problem they’ll sort it for you quite quickly.

Site 2, participant 3

He can access our rosters remotely from there, so if there is a problem and we can’t do it he’ll do it then and there and then we just click refresh and carry on then.

Site 1, participant 1

The evidence indicated that support via personal contact is useful. Robinson et al. 126 found that the presence of a dedicated and knowledgeable staff was an important factor in support for the Veterans Health Administration staffing methodology processes. Staff had a specific remit to support the staffing methodology and acted as a resource for managers:

Our SM [staffing methodology] Coordinator . . . did an outstanding job with it . . . she found ways of taking stuff and making it to the point where the staff who lacked the expertise or the knowledge or skill set could really . . . digest it.

Robinson et al. 126

Interview participants acknowledged that face-to-face support was also very useful:

When I came into post we were still doing the daily drop-in sessions, teasing out problems really, snag lists; that sort of thing. I ended up going for a one to one session with X, . . . I booked it in, just so I knew what I was talking about.

Site 3, participant 4

If the ward manager or deputy is struggling with something and can’t remember how to do something, they can just pick up the phone and somebody’ll say ‘I’ll pop down there now and do it with you’ so next time they’ll be able to do it.

Site 1, participant 3

Also related to the Veterans Health Administration staffing methodology, Robinson et al. 126 and Taylor et al. 127 used unit and organisational expert panels for support. Taylor et al. 127 reported the use of expert users from other organisations to help with training. Organisations with a good reputation for e-roster such as Portsmouth Hospital, were chosen to demonstrate practice and tools to others. 129 Diers et al. 154 used a ‘buddy’ system; one interview participant also highlighted the importance of such a system:

I mean it’s a way of working together isn’t it, and to bringing people on. So, a new ward sister would always be buddied up with another ward sister, a more experienced one, to make sure they’re okay.

Site 2, participant 1

Research evidence suggested that it is vital that NHS managers have support to develop the skills and knowledge to equip them to deal with the challenges within their role, including decision-making around safe staffing. Shirey et al. 187 found that stress and organisational complexity has a negative impact on managers’ cognitive decision-making and recommended various means of support, including formal programmes for professional development and the creation of a supportive environment. The evidence suggested that supportive environments can be fostered by an organisation. Magnet hospitals appear to increase nurses’ perceptions of support; these are organisations in which professional models of nursing are promoted to enhance innovation and nursing autonomy. 118,119 One interview participant explained how her organisation facilitated support for NHS managers around safe staffing decisions:

So they’ll have support from their managers, and we’ve recently gone through an organisational restructure, so last year, that makes sure that we do have adequate nursing support, and we’ve put in extra posts to do that.

Site 2, participant 2

Peer support within the organisation also appears to be important for NHS managers. All of the stakeholder feedback highlighted how NHS managers should have compassion and empathy for each other and that there should be open channels of communication at all levels. One interview participant’s comments illustrated this:

My role is about maintaining positivity at the end of the day and, as I say to the matrons, you know, come in the office by all means, close the door, rant and rave and shout and say whatever you like, but when you go out there, you go out to the coal face, you know, you put your smile on your face and we deliver that positive encouragement to the teams as best we can. We know it’s hard, we know it’s not perfect by any stretch of the imagination, but what they want to see is, is they want to see the senior nurses out there on the wards supporting and helping, doing their best to try and get where they can.

Site 3, participant 3

Visible support is linked to investment in the development of NHS managers’ skills and knowledge. Feedback from the NHS manager stakeholder groups indicated that it is essential for NHS managers to understand the organisational model of staffing deployment and its impacts. Interview participants indicated how teamwork on staffing helped to develop a common understanding:

I think everybody knows what the expectations are to be honest.

Site 2, participant 2

The availability of training and education on WPTs was highlighted by many sources as being essential, so that nurse managers could input, collect and analyse data and understand how to use the technology effectively. 124–127,129,131,143,147,150,154,156,162,171,173,174,179,182,191,192,195,198–203

The evidence also suggested that training and education is important for the development of nurse managers’ skills for decision-making processes related to their role generally148,149,187,188 and in relation to staffing decisions. 153,174 Training and education can promote skills in organisation, critical thinking149,188 and human relations, such as communication and team-building,149 negotiation and conflict mediation148 and skills in technical competence. 149 Zori et al. 188 suggested that emotional and social intelligence skills can be developed through educational strategies. Interview participants also explained how some areas used mentorship programmes and action-learning sets to develop NHS managers’ skills. Leadership and the ability to work as a team are essential competencies for the nurse manager149 and are vital for staffing decisions. 163,167,186,189,199,202 The development of leadership and teamwork skills can be facilitated through train-the-trainer programmes. 120 Interview participants also found formal leadership programmes useful:

Well it’s given us a bit of ownership, and helped us with our leadership skills, maybe, to realise what our skills are and how to sort of improve them and how to, staff engagement and discussing and doing practical things like that. We’ve had practical sessions on, appraisals, managing the roster was another one, all very practical tools that we can use day-to-day, and of course it helped the networking with other managers, other hospitals and sites as well, and you realise you’re not the only one we all have the same problem.

Site 1, participant 1

Targeted training and education on WPTs appears to be important for all levels of the organisation. 126,127,174,179,195,200 The evidence suggested that nurse managers also have a role in the training and education of their staff; Lammintakanen et al. 197 found that in the introduction of electronic information systems:

Learning and using the information systems was supposed to take place in a positive and encouraging atmosphere. The nurse managers’ duty was to promote such an atmosphere, which also allowed learning by trial and error.

Lammintakanen et al. 197

Some interview participants who were senior managers stated that they mentored or coached more junior managers:

I’ll sort of show somebody three times, you do it on the fourth time; it’s that sort of coaching really just to get the confidence going.

Site 1, participant 3

Furthermore, Allen164 highlighted the importance of educating staff nurses on what to expect about how their patient allocation will be made. Bonner et al. 156 also suggested that staff need education in order to understand the constraints under which the decisions are made.

Consistency and standardisation in training for WPTs also had an impact on staffing deployment. 147,164,167,200 In addition, the training may clarify what can be expected from their use; Flynn et al. 203 commented on the need to:

. . . explain through the education and training sessions that the purpose of the workload and workforce planning tools was not necessarily to create additional capacity, but to obtain the most out of existing resources.

Flynn et al. 203

There were examples of the need for continuous repeated training and motivational sessions for some WPTs, for example the RAFAELA system (a workforce planning tool),124,125 and also for automated rostering. 167 Training took many forms for both managers and nurses. Anderson et al. 125 used exercises for a patient classification tool and NHS Education for Scotland used work-based activities within their toolkit. 125,202,203 Wallace and Pierson195 found that managers were positive about a scheduling programme of resources and tools, and found workshops helpful. They requested more scheduling guidance and an online discussion forum. Kane179 described regional user forums for the development eCAT, the community caseload technology. Diers et al. 154 and Ruland171 both noted the importance of financial training. There was some evidence that training aimed to provide additional technical skills or focus on those who lacked IT skills. 174,179 Kane179 found that those who had better IT skills were initially more likely to engage with a staffing technology, but:

Some users who perceived themselves ‘not good with computers’ were anxious at the start of the training, but most became so interested in the reports that they were inspired to learn more.

Kane et al. 179

Many interview participants indicated that they had received training in use of WPTs, particularly e-roster. However, some studies’ findings suggested that education and training alone are not sufficient and other factors are inter-related, such as the credibility of the support team and good leadership. 126 van Oostveen et al. 124 also concluded that although training and education are essential, preparatory work is needed on the social context to facilitate nurses’ acceptance of staffing tools.

Leadership

Strong nurse leadership appears to be important for effective nurse staffing. 126,129,134,143,146,169,199,200 All stakeholder groups highlighted the need for strong leadership and felt that a good leader will offer a shared vision and people will follow them. NHS manager stakeholder group feedback referred to leaders as ‘flying the flag for nursing’. This may be linked to the use of data from WPTs as political levers for change or negotiation. Drake143 further illustrated how leadership was important for roster management:

Directors of Nursing explained their relationship with ward staff in terms of patronage, often citing examples of, ‘fighting for resources’ or, ‘confronting non-nursing staff’ on behalf of the ward staff.

Drake143

Szumlas204 also suggested that effective leadership is important for recognition of nurse capacity to align staffing skills and competence with patient need. Supportive leadership also appears to be important to drive staff engagement with WPTs. One interview participant indicated how she gave practical help:

We’ve had a day training to learn how to e-roster. To be honest, I, as a ward sister at the time, as a matron now, I will still go and spend time with a ward manager on e-rostering, to teach them some of the tricks.

Site 2, participant 1

The importance of the nurse executive and other corporate-level roles in driving staff engagement with WPTs, through the provision of time and resources to embed their use, was also evident. 126,129,134,143,147,169,200 Douglas199 highlighted how WPTs should be at the top of the nurse executive’s priorities and that their experience and wisdom could influence success. Robinson et al. 126 found that organisations that were categorised as being at a high level for implementing a staffing methodology all had nurse director support, which ensured consistency and staff development at all levels. Senior nurse managers can also influence the type of training and education that managers receive. 150

The evidence suggested that leadership should be shared at different levels of the organisation to promote collaboration in the use of WPTs. 126,127,129,147,163,166,182,190,197,200 An aligned approach appears key to the facilitation of local clinical and human resource ownership. 129 This may also facilitate political leverage, negotiation and change. 166

A standardised team approach appears to be important in implementing new technology. 200 NHS manager stakeholder feedback, interview participants and many sources from the literature indicated how leadership delegation, collaboration and teamwork is essential:124,126,127,130,131,134,169,174,179,200,205

Obviously, we’re very lucky in my division that I have X care as well, that sits in my division, so the lead nurse for X care, we’re constantly talking, saying ‘Do you need staff? Do we need staff? How do we support each other? What do we do?’ So, the matrons will do that kind of at grass-roots level. They will move the staff and they will make the decisions based on the acuity of the ward at the time.

Site 3, participant 3

Integration of data for sense-making

The value of technology is its ability to capture and process large amounts of data and to display data in meaningful ways that optimize communications, support collaboration, drive informed decision making and maximize resource use.

Crist-Grundman and Mulrooney169

The integration of data from WPTs can empower managers to understand the link between the nursing workforce and the delivery of care on patient outcomes and identify potential problems. 206 Data can be integrated from multiple sources and tools,129,180,183 and can be used for multiple purposes. 180 Many sources noted that in order for nurse managers to make effective decisions, data needed to be presented in a format that they could understand. 175,180,183 Hyun et al. 180 suggested that data presentation should be in a standardised and user-configurable way, to allow for sophisticated processing and data-mining. 180 Pruinelli et al. 175 noted that unless data are integrated and extracted in a meaningful way:

. . . nurse leaders will struggle to achieve better patient, nurse, and financial outcomes.

Pruinelli et al. 175

There are many examples of the integration of different types of staffing data through the use of WPTs. Some approaches to data integration followed a data warehouse model, which is a single source of information relevant to staffing deployment collated in one tool or visual dashboard. 134,174,175,192 The evidence demonstrated how multiple sources of data had been integrated; for example, Kerr and Timony167 integrated a rostering system with patient dependency, the hospital information system and the payroll. 167 Junttila et al. 192 integrated nurse-sensitive indicator data with staffing and rostering. Robinson et al. 126 found that a staffing methodology was implemented effectively when integrated into the budgeting process. Data have also been integrated on admissions and discharges,192 caseload179 and the environment of care. 175 Many sources of evidence indicated the importance of having information on nurses’ characteristics (qualifications, skills competence and training requirements). 128,133,144,151,152,157,158,160,164,174,175,180,181,184,189,191,198 Rostering can also integrate data on annual leave and sickness. 129 Valentine et al. 190 integrated data on shift availability, tracking and incentives when introducing an ‘Openshift’ (self-rostering) technology. Similarities were also noted in other organisational staffing, with the need for the integration of an extensive series of factors for complex police staffing decisions, such as current police workload based on individual and group performance, and demand and performance indicators. 153

Technology is reported to be important for identifying data for effective nurse–patient allocation. 151,170,175,181,184,186,192 The integration of data appears to facilitate NHS managers’ understanding of resource availability. Pruinelli et al. 175 noted that prior to the development of the NMMDS, data related to nurses’ qualifications and licensure were found in disparate systems (e.g. human resources or unit manager’s records) and were difficult to access quickly. 175 The integration of data may also enable managers to identify potential problems, barriers and causes of unsatisfactory resource management. 171,172,206

Integration of the data to determine staffing demand

Determining staffing demand is complex; the integration of data from WPTs may empower NHS managers to make sense of the staffing need and have a positive impact on decision-making. Patient acuity and dependency were consistently key factors in the consideration of staffing deployment. 158,163,183,207,208 WPTs can integrate patient need with staffing resources:

I think what the electronic systems have started to do, and I think there is certainly scope for improvement in this, is accurately kind of . . . weigh up the acuity and dependency of the patients that we are seeing versus the number of staff rostered on duties.

Site 3, participant 3

One interview participant indicated how she used patient observation data, whereby acuity scores are entered on handheld devices, and aligned these with staffing availability:

I would make sure . . . that there’s no-one that’s sick who hasn’t been picked up, or make sure we’ve got enough people on there.

Site 3, participant 2

This participant explained how the system also indicated how many observations had not been done and this could suggest the need for support, as people are ‘struggling to get everything done’.

Hyun et al. 180 cautioned that it is not sufficient to use patient acuity data alone. For a data warehouse approach, Frith et al. 183 recommended multiple sources of information to determine prospective staffing decisions; these included patient factors, such as acuity, but also staffing factors such as skill mix, nursing characteristics and unit and hospital factors. 183 In a concept analysis, Cathro163 highlighted fifteen factors that have an impact on decision-making for nurse allocation; patient acuity and patient need factors were the most important, but other influential factors were regulatory and safety issues, continuity of care, anticipated flow and nursing factors.

Integration of the data for a triangulated approach

When nurse managers have effective support and facilitation of skills and knowledge development, they appear to be empowered to use tools effectively to understand the complexity of the problems they face. 145 The triangulated approach, consisting of the use of evidence-based tools, comparison with peers and the application of professional judgement, can be facilitated when data are integrated for effective decision-making. The evidence strongly suggested that the integration of data, including triangulating judgement with WPTs, enabled decision-making about safe staffing. 124–126,129,131,133,134,145–147,151,154,159–162,166–175,177,179–183,187,190–192,194,199,204,206,208–210

There were also examples of integrated data decision support tools, including CLASSICA MIS (version 1; Ruland RM, RAVN IH; Oslo, Norway)171,172 and the Dynamic Network Analysis Decision Support tool (DyNADS) (version 1; Effken et al. , AZ, USA). 145 According to Effken et al. ,145 tools can stimulate higher-level thinking, and, as managers are able to see the bigger picture, they do not seize on quick solutions that may not be effective in the long term. Tools do this as they:

. . . deal with multifactorial problems in ways that synthesize the disparate issues nurse managers encounter into higher level concepts and allow them to test hypotheses about the kinds of solutions they might implement in a safe, virtual environment.

Effken et al. 145

Importantly, in using their professional judgement for decision-making, managers need to critically interpret WPTs on patient data in order to align staffing allocation with patient needs. 131,158,163,176,178 van Oostveen et al. 178 indicated that a computerised decision support system:

. . . can never completely replace human insight. The nurse-to-patient assignment generated by the ILP [integer linear programme] model is a proposal that has to be assessed and, if necessary, adjusted by one or more nurses to obtain the final nurse-to-patient assignment.

van Oostveen et al. 178

Outcomes

Ensuring the efficient deployment of staffing skills, in the right place, is complicated. When nurse managers were visibly supported to develop knowledge and skills, and there is effective leadership, this had a positive impact on their engagement with and understanding of the data from WPTs;126,127,134,143,174,199 this facilitated effective decision-making for safe staffing deployment:

We rely on these systems to improve our ability to assure the right resources are assigned to the right situations that result in the best possible quality, safety, and financial outcomes.

Douglas199

The evidence indicated the importance of matching nurse characteristics, such as education and skills, with patient need. 134,158,160,174,184,189 However, multiple factors, such as finance and organisational requirements, also have an impact on efficiency in safe staffing. When nurse managers have effective support and leadership, this has a positive impact on their understanding and use of integrated data from WPTs to ensure appropriate and efficient deployment. For example, many sources of evidence indicated that managers use scoring tools to match up staffing effectively; these scores are linked to workload. 124,125,127,129,130,133,144,147,151,175–180

Furthermore, WPTs can have a positive impact on finance when data are integrated on resource availability in alignment with the budget. This results in efficient resource management leading to cost reduction, as a greater understanding of staff availability can reduce overtime and agency costs. 126,127,134,157,166,168,170–172,175,176,183,190,191,194,199,206 To conclude, although WPTs can facilitate sense-making of complex data, NHS managers require effective leadership and support to develop the skills and knowledge required to empower them to effectively interpret data from WPTs for safe staffing deployment.

When there may be misalignment

Poor understanding of WPTs can impede their use. 124–126,151,155,208 Evaluation studies found that poor understanding impeded knowledge on how and when to report data, and this had an impact on decision-making;134 poor understanding also conferred a lack of confidence in the data. 127 This appeared to be linked to a lack of resources and a dedicated support team126,193 and a lack of time to develop skills. 185 Poor access to training also appears to impede the use of WPTs. 125,126,185,193,208 However, van Oostveen et al. 124 and Anderson et al. 125 found that, even when training was provided, it lacked success in some areas. Kivinen and Lammintakanen182 found that some managers criticised the educational content for being too technically oriented. In relation to the nurse manager’s role in general, Eduardo et al. 148 found a mismatch in training received on decision-making and what is expected from hospital managers’ performance. Some interview participants indicated that WPTs were not used to their full capacity, often because of a lack of training:

They do put in acuity data and all that stuff, and their numbers, and all that numbers in, that does go in electronically and that goes onto e-rostering and to safe care. But that’s not what I would use, or anyone else would use, to allocate staff. Whether that is the intention of it I don’t know. It certainly isn’t being used for that, and it isn’t anywhere that close to being used for that at this present time.

It’s just not fit for purpose?

It’s taken a lot of embedding, and with the . . . so I don’t think . . . I don’t think that . . . maybe that is what it’s meant to be used for, but I’ve certainly never seen that being . . . we’ve not been taught how to do that, we’ve not been given reports from that or anything along those lines.

Poor leadership appears to also have a negative impact on nurse managers’ use of WPTs. 126,127,182 Poor leadership manifested when leaders lacked commitment to the implementation of WPTs; often they did not value them and failed to communicate effectively or to commit to providing training programmes. 126 Hockley and Boyle166 highlighted a lack of engagement by senior management in the adoption of IT solutions in England, despite established safety and efficiency benefits. Certainly, the evidence strongly suggested that weak leadership caused delays in the implementation of WPTs. 126,127,173,174 Robinson et al. 126 found that in organisations with a low level of implementation, leaders were sceptical and had not bought into the staffing methodology. Stable leadership also appears to be important, as Anderson et al. 125 indicated how a shift in leadership impeded the use of a tool. One interview participant indicated that restructuring within the organisation could impede nurses’ use of data effectively from tools:

Well they’re devolving it to an operational level, which is fine, and I think it needs to be at that level, but they’ve not allowed a period of transition of knowledge and competencies.

Site 1, participant 3

Leadership attributes

The stakeholder feedback strongly suggested that NHS managers should be good leaders; this was also suggested in some of the literature, for example, Connelly et al. 149 However, sometimes the published evidence did not distinguish between the terms ‘managers’ and ‘leaders’, or define the range of associated roles. 130,144,146,183,199,204,205 All of the stakeholder feedback strongly conveyed that NHS managers should be approachable and good listeners, but emphasised that having authority was vital. For PPI stakeholders, it was important that NHS managers had worked their way up through the system, so that they knew different roles and understood them. All of the stakeholder group feedback highlighted that NHS managers should also be compassionate and caring. Being supportive also appears to be an important attribute. 128 NHS manager stakeholders strongly emphasised the importance of having knowledge, wisdom and credibility and of understanding the system. The evidence also suggested that integrity and personal credibility is essential:148,188,201

Nurse managers who are able to provide guidance for addressing issues and problems in a credible manner may create a feeling of trust and safety in staff RNs and a subsequent perception of support from their nurse manage.

Zori et al. 188

Being resourceful and decisive appears to be important for nurse managers when making decisions on staffing deployment;150 the NHS manager stakeholder feedback also indicated that such attributes were essential. Furthermore, the published evidence, the NHS manager stakeholder feedback and the advisory group feedback suggested that flexibility is an important leadership attribute for safe staffing,143,150,188 as it is an important quality for NHS managers in order to deal with the unexpected:

. . . when charge nurses are flexible, they can quickly make adjustments to imbalances in available staffing and changes in patient’s acuity or census to prevent a breach in patient safety, and detect which patients or nurses need the most help.

Wilson et al. 150

There was agreement between all stakeholders and from the published evidence that nurse managers have an important role, outlined in the responsibility and accountability that they have for safe staffing work. 126,127,129,143,162,163 With the introduction of a staffing methodology, Taylor et al. 127 emphasised the responsibility of the nurse executive and Robinson et al. 126 found that areas with high levels of implementation demonstrated leadership accountability. Managers have to take responsibility for rostering,129,143,162 with Cathro163 stating that charge nurses are accountable for decisions on nurse allocation. Both Robinson et al. 126 and McIntyre129 emphasised that accountability and responsibility are important at all levels of nurse management. Furthermore, all stakeholder groups highlighted that NHS managers should take responsibility. Taking ownership and engaging with safe staffing appear to be important attributes of leadership:126,149

The charge nurse role is critical because of current staffing issues, the use of agency and/or foreign nurses, and the need for a proficient nurse to be accountable for issues arising in the unit. In short, charge nurses must take ownership for all unit activities.

Connelly et al. 149

Leadership skills

Leadership skills appear to be important for effective rostering and allocation to ensure patient safety. 163,167,186,189,199,202 Important leadership qualities for effective staffing deployment include interpersonal skills,128,149,190,201 with the ability to be a tactful communicator being a fundamental leadership skill for the deployment of staff:

Effective charge nurses used their communication skills to elicit information to obtain nuanced factors that may influence staffing decisions. They regularly moved beyond simply using the prescheduled nurse–patient ratio to meet the intra shift staffing demands.

Wilson et al. 150

The literature indicated that nurse managers have to communicate effectively to negotiate with nursing staff, other professions or departments to ensure that they have the right numbers of nursing staff on duty each day. 130,149 One nurse manager explained that negotiation skills are important for staffing deployment when staff need to be moved:

. . . it’s about focusing on what . . . the reason we’re here isn’t it? It’s about actually all the patients under our care, you know, are our responsibility, and that shouldn’t matter which area they are, you know really. It’s that we have to provide the care, don’t we? And any team at any time could go under the same problems, so each sort of area will have its problems at different times around staffing, and one team may have supported another area at a time when they were short. So, it’s probably appealing to the better nature of people, you know, to provide the service that we need to.

Site 3, participant 1

The ability to think critically is an important leadership skill. 131,148–150,188 This was emphasised in the NHS manager stakeholder feedback. Leaders need to have organisational and co-ordination skills;126,149,190 they need to be skilled in planning131,149 and collaboration,131,133 and they need to be able to set expectations, monitor and evaluate. 126,129,176,199,204 Furthermore, leaders need to be competent in IT as they set an example for their employees:

‘I had to do it myself, too. My own example was a good incentive. You sit beside the employee on the computer all day if needed.’

Lammintakanen et al. 197

The nature of leadership work

Being a visible presence was emphasised as an important part of leadership for NHS managers in all stakeholder groups, with many examples in the interviews of NHS managers being present on each ward daily to assess staffing requirements and to support other nurses or nurse managers:

Then I have a slow walk round the wards; check on all the ward sisters, if they’re in. Just a general feel for the wards really.

Site 2, participant 1

Using communication skills at all levels is an important part of the NHS manager’s leadership role. The NHS manager stakeholder feedback reiterated the importance of negotiation and of clear communication channels with different people in different levels within the system. Crist-Grundman and Mulrooney169 also indicated that leaders must engage at all levels. For the NHS manager stakeholder groups and the interview participants, communication and negotiation were linked to knowing the system and understanding the bigger picture. Having a broad perspective on staffing throughout the organisation, and not within only one clinical area, was also evident in the literature:150,201

. . . the matrons will come on and we will move staff because we’ve obviously taken that helicopter view of all nine wards and recognise that some wards need some additional support.

Site 3, participant 3

With regard to the attribute of flexibility, Burns205 proposed that leaders should focus on action rather than design and avoid excess control to:

. . . create environments that allow complex adaptive systems to explore, self-organize, and create successfully.

Burns205

However, Burns205 found that, although leaders strongly agreed that health care is chaotic and complex, they did not agree with principles and statements that encourage minimal specifications, uncertainty or becoming comfortable with tension and paradox and managing with freedom and intuition. 205 Health-care managers preferred strong direction and control. Although stakeholder feedback with NHS manager groups reflected the importance of innovation and of generating new of ways of working, there was also a strong focus on the need for knowing the parameters, barriers and risks, suggesting the need for clear boundaries. Having a ‘safety net’ with the NHS manager acting as a gatekeeper was a consistent theme in the NHS manager stakeholder groups’ feedback. Seeking staff feedback and acting on it to improve the service was a theme from all of the stakeholder work. Gaining feedback from patients ‘to know their needs’ was emphasised within the PPI stakeholder groups. NHS manager stakeholder groups emphasised the managers’ role in advocating for the patient when making staffing decisions.

Knowing staff

Understanding the knowledge, skills and competence of the staff appeared to be a requirement for effective nurse rostering and patient–nurse allocation. 129,131,146,150–152,207 This included an understanding of non-registered staff competencies. 131 This information may be gained from various sources, including paper and electronic staff records, but can often be based solely on the manager’s personal knowledge of the individual nurse,151 as illustrated by one interview participant:

I’ve just got all my experience and knowledge of my own staff, and knowledge of the patients at the time, really.

Participant 1, site 1

Having knowledge on nurses’ education and skills is essential for safe staffing allocation;128,133,144,151,152,157,158,160,164,174,175,180,181,184,189,191,198 however, the development of this knowledge may be dependent on features of an organisation such as its size and complexity. However, some sources of evidence suggested that nurse managers also need to know about nurses’ capacity, so that skills and competencies can be aligned effectively with patient need;143,150,204 particularly as nurses may be viewed as an expert in one setting and a novice in another. 133 Berkow et al. 152 found that nurses linked expertise to experience of the patient group or setting:

On units where patients are at high risk of deteriorating rapidly, participants overwhelmingly agreed that experience is the most valuable staffing enhancement because experienced nurses were believed to provide faster differential diagnosis and intervention.

Berkow et al. 152

Allen164 found that one study participant summed up competence in the nurse as ‘Can he/she handle that acuity?’. Knowing staff may require many sources of information. One participant in Drake’s study on rostering indicated that managers needed to know the staff circumstances, such as whether they needed money or time off. 143 van Oostveen et al. 178 felt that the mental and physical health and other duties of nurses could be incorporated into a computerised decision tool for making patient assignments, as they found that:

. . . nurses are significantly influenced by each other’s emotional commitment, private circumstances, and mental load.

van Oostveen et al. 178

The NHS manager stakeholder groups and interview participants also strongly indicated the importance of knowing the team, through managing personalities, building team loyalty and playing to the teams’ strengths:

You just know who can deal with what because you can have a very experienced nurse that might not cope at all well with any kind of pressure, or stress, but you might have a more newly qualified nurse who’s very, calm, safe, and practical, do you know what I mean?

Site 1, participant 1

Knowing the team was linked to being a visible, supportive presence:

I know every single person on the wards, right the way from the housekeeper, to the domestics, to the ward manager, you know, and it’s about going round, making sure that everybody is okay, everybody is safe. If people have complaints and concerns you’re listening to them, you’re doing your best to action, and communicating. And that’s what I say to the matrons, you need to be visible, you need to be out there, you need to be in with the team, you need to be talking to the staff.

Site 3, participant 3

Importantly, nurse managers need to understand the capacity of temporary staff. 149,151,160 In relation to risk management in ICUs, Rischbieth160 pointed out that managers must decide whether or not to engage staff without knowing their capacity:

If their dilemma results in a decision to just take anyone or just do without, this may immediately expose that ICU to an increased risk of potential problems.

Rischbieth160

Local knowledge was linked to being resourceful, as it means that managers know where the resources are, how to use them to get things done and how to identify who can help. 149,150 The importance of being resourceful was echoed in the NHS manager stakeholder group feedback. Some sources suggested that experience is important for understanding and knowing staff:134,151

Experienced nurse managers possess insights into the relationship between the numbers and characteristics of nursing staff and patient outcomes.

Anderson et al. 134

Effken211 suggested that, although nursing intuition has been perceived as an integral part of clinical expertise, it is information based. Kontio et al. 151 found that managers emphasised that their tacit knowledge, gained from experience, influenced their decision-making when information was not available from information systems; this may be linked to professional judgement. According to one manager:

I make decisions daily on allocating staff to different posts. I don’t get any supporting information from any information systems; rather, I do this based on my own experience.

Kontio et al. 151

Visibility of real-time resource

The published evidence and stakeholder feedback emphasised that WPTs may be a resource for knowing staff capacity and capabilities. The literature highlighted the need for visibility and transparency of data. 161,162,179 Staffing software should be able to quickly find additional staff for changing staffing needs. 170 Many sources demonstrated that WPTs can provide data on real-time resources. 151,168,172,181,194 Nagy and Nathan181 found that a staff scheduling system made data visible, for example, on hours worked and staff availability. Ruland and Ravn172 found that a nursing MIS provided information on costs, activity and staffing, and Wadsworth and Kurilla194 found that an automated attendance system enabled managers to react to staffing levels in real time.

Technology was also important to identify information on education and skills. 151,170,181,184,186 Although management staff have an extensive personal knowledge of staff characteristics, this information should be in electronic format for use in decision-making about staffing:151,183

‘I have the staff competencies in my memory, but these should also be in an information system where all of the training sessions that an employee has taken are recorded. There could also be an employee self-evaluation of his/her competencies.’

Kontio et al. 151

One interview participant has found that an e-roster helps her to understand nurse competencies effectively:

So, each person had a competency attached to them on the system. So, if you hover over the name on e-roster you can see whether they can be in charge, you can also see what abilities they have, so whether they can give non-invasive ventilation.

Site 2, participant 1

However, for one interview participant, personal knowledge of staff does need to be complemented by the technology to facilitate complex decision-making:

So that the correct grades of staff are on, and how many they’ve got on, and there is a way of looking at it and they’ll tell you, you know, it will tell you if it’s, if it’s staffed safely, or if the finance is over, or if there’s too many on annual leave and stuff.

Do you find that useful?

Yes. No but it’s really, again, about knowing your staff anyway knowing the competency of the staff you’ve got working and who they are.

You think it’s a bit more than just having the right grade and the right numbers?

Oh, yeah, yeah, you give, you’ve got to know your staff and know the capabilities and it’s getting to know them and their working really, you know? Like, just a band 5 they could be, it depends what they, what sort of training and everything, and how, you know competent they are, and if they could be in charge of the ward.

In conjunction with professional judgement, WPTs can therefore facilitate the visibility of staffing resources for decision-making. 151,175,183 Pruinelli et al. 175 found this with the nursing minimal management data set; McIntyre advocated that the live roster should be used in handover and bed meetings;129 Kontio et al. 151 highlighted the importance of data for managers’ decision-making, for example, in finding out staff availability when there is an urgent need to cover a shift. Crist-Grundman and Mulrooney169 reported that technology enables the availability of data at any time and anywhere, which facilitates analysis for decision-makers. One interview participant offered an example of how technology can give real-time information:

As they’ve embedded e-roster within the practice of the day, and constantly referring back to it and utilising it within the day.

Site 3, participant 4

Furthermore, technology and tools can provide graphical displays to make data visible:126,129,134,151,154,168,174,175,181,183,206,212

A dashboard provides graphic representations of data in real time that can assist the nurse leader to make staffing decisions, monitor productivity, and track the effects of staffing on patient outcomes.

Frith et al. 183

Interview participants explained that when technology is used to input patient observations and critical illness scores, managers can remotely access data to gain an overview of patient acuity levels. Interview participants also indicated that WPTs can offer transparency and visibility of resources:

I’ve got an overview of exactly what’s needed and what’s available across the patch.

Site 3, participant 2

It’s much more visible. It’s much more measurable, you know so, for instance, we can say to people right, you all need to get your rosters in 6 weeks in advance and this is the cut off, you know, so everyone can work to that. So, it helps with temporary staff planning as well. So, if you do plan 6 weeks in advance as required by the trust it means that you can get your requests out. It’s much, much tighter with payroll, you know, you rarely have any queries, whereas years ago you used to have lots and lots of queries.

Site 3, participant 5

Therefore, WPTs may enhance decision-making. Anderson et al. 134 indicated that intuitive judgements that contribute to staffing decisions are subjective:

The experienced manager can intuitively predict the outcomes of staffing levels or skill mix; however, this decision-making process has disadvantages. It takes time to develop expertise, and decisions based on intuition are difficult to justify.

Anderson et al. 134

Anderson et al. 134 indicated that a dashboard will provide the manager with an ‘objective alternative to that innate sense developed through experience’. However, one interview participant pointed out that data from technology can only complement the manager’s skills and knowledge:

The same way that I think standing in a ward, feeling the ward, is the most important thing you can do about staffing. So, I don’t think . . . all the technology in the world is good, maybe you could call it icing on the cake, but I certainly don’t think we should ever take . . . have that instead of, it’s got to enhance it.

Site 3, participant 2

Outcomes

The visibility of data on resource availability can increase NHS managers’ impact on safe staffing when managers demonstrate key attributes of leadership and know their staff.

Allen158 found that the starting point for matching nurses to patients was determining staffing numbers; the author highlighted the importance of allocating staff who had the competence to keep the patient flow going, particularly in clinical areas in which assisting in the movement of patients through the unit was vital. NHS manager stakeholder feedback indicated that unlocking the potential of people was an important leadership role, which empowered staff and was linked to good-quality care. Bonner et al. 156 described the process of the assessment of skill mix and numbers for each shift as ‘an onerous one’. However, having real-time data can make a difference to matching staffing to demand,168 particularly in safety-critical environments. 166

Managers need to critically interpret WPTs on patient data in order to align staffing allocation with patient needs. 131,158,163,176,178 When data are visible, this enables NHS managers to respond flexibly to patient need. When WPTs facilitate the visibility of resources, this can be matched to performance metrics and finance. 131,134,168,206 Anderson et al. 134 suggested that dashboards improve resource visibility and metrics for performance improvement, and this can be linked to monetary reward. Finn206 found that the use of an e-roster improved the understanding of staff availability, bank and agency use, and this reduced staffing costs. Barton found that technology resulted in more effective resource management, as resources were made visible:

The managers often didn’t know what resources were available hospital wide. For example, a unit might order contract labor from an agency when float nurses were idle.

Barton168

Likewise, McIntyre’s129 guidance indicated that an effective roster provides an opportunity to respond to real-time staffing challenges by providing an overview of staffing across organisational units.

When there may be misalignment

Some of the evidence suggested that ineffective leadership attributes may impede safe staffing work. According to Anderson et al. ,125 a shift in leadership impeded the use of a tool, suggesting the importance of a consistent approach. Several sources found that poor leadership was a barrier to the implementation of WPTs. 126,127,173,174 Furthermore, the evidence suggested that some managers fail to use WPTs to full capacity to make needs and resources visible. Finn206 found that an e-roster was challenging for some managers, with some preferring manual methods over the automated function. Some staff members reported that e-rosters are made too far in advance, and others felt that they had lost the personal touch in allocation/requests. This was also noted by Drake:

Some Ward Managers viewed it favourably, while others felt that it removed the ‘human’ aspect of rostering and often produced unworkable patterns.

Drake162

There was also a suggestion by interview participants that WPTs were not used to their maximum capacity, for example, for understanding staff skills:

Would it be useful to have that on that e-rostering system so you could track that? I know some systems will tell you somebody’s experience, whether they can, you know . . .?

Yeah, I think probably that . . . I think probably there is a bit of that on there already; I just don’t think people use it.

Furthermore, despite the visibility of needs and resources, in some instances, WPTs may not facilitate NHS managers’ work on safe staffing. A study on a workforce measurement system found that nurses (including nurse managers) felt that they were filling in bits of paper but not seeing any benefit from it; this suggested that, on a day-to-day basis, the staffing changes resulting from the system were not welcomed:

If the system indicates that there is a surplus of nurses for the current workload, nurses will be sent out to float on other units; however, if it identifies a need for more staff, rarely will this result in extra staff support being provided to the unit.

Hall et al. 144

A culture of evaluation

The implementation of a critical system does not end when the technology goes ‘live,’ rather it is an evolving journey, where the expectation of use is clearly communicated and practiced; where configurations and practices are re-evaluated as goals are achieved and new challenges are faced.

Crist-Grundman and Mulrooney169

The evidence supported the importance of evaluating the use of WPTs. 124,126,129,131,148,172,174,190,195,199 Robinson et al. 126 categorised organisations based on their implementation of a staffing methodology and found that, in organisations in which there was a high level of implementation, staffing data were fed back to the units. Formal staff feedback appeared to be important. 195,199 There were many examples of gaining staff feedback when WPTs were trialled. 124,127,144,146,161,174,176,179,192,197 Staff satisfaction surveys with scheduling practices were recommended. 189 One organisation surveyed staff on self-scheduling technology and found that 75% of staff viewed it positively, with 87% finding it easy to use and 68% commenting that they had received fewer calls asking them to come in for work. 191

All of the stakeholder groups emphasised the importance of gaining feedback from staff on planning and deployment and having open communication channels to facilitate this, with NHS managers having an important role in listening to staff. The advisory group emphasised the importance of an evaluation culture for a positive learning climate and organisational learning. The NHS manager stakeholder groups suggested that reflection was vital to learn for the future through determining what has worked, and innovating and making changes when necessary. Acknowledging when good work has been done, giving praise and celebrating success were also important. PPI feedback also reiterated the importance of listening to, and gaining feedback from, patients.

A standard approach to articulate needs and resources

An organisational culture that values engagement and learning may be strengthened when this is informed by standardised, credible data on needs and resources. Standardised data facilitate organisational learning through evaluation and comparison. The data from WPTs make needs and resources visible for evaluation in the light of planning and deployment outcomes; for example, Anderson et al. 134 indicated that their staffing tool, which used a dashboard format, aimed to:

. . . make current, relevant data available to nurse managers at the unit level and integrate evidence on staffing with patient and organizational outcomes.

Anderson et al. 134

When the organisational culture values information, technology facilitates its use for organisational decision-making. 182 Diers et al. 154 suggested that the technology may equip managers:

. . . with powerful tools for retrospective analysis of process management.

Diers et al. 154

E-rostering facilitated measurement of, and understanding about, staffing deployment. 206 Data from WPTs enable retrospective review to evaluate staffing levels124,170,175 and to identify the need for further resources. 168 When there is visibility and standardisation of data, this enables monitoring for cost, efficiency and quality purposes. 129,175 Importantly, this facilitates financial review. 194,210 This means that the impact of bank/agency spend is visible,129,206 as is the difference between actual staffing and predicted staffing:206

It’s all tracked . . . tracked on the system, so you can see at any point exactly where everybody is. So financially you can follow the trail.

Site 2, participant 1

Some sources emphasised a need for a standard approach for ease of measurement for comparison and for adherence to legislation, mandates or guidance. 124,126–128,131,147,175,183,195,199,206 Taylor et al. 127 emphasised the need for a standard approach, through which staffing methods are mandated. Fenton and Casey147 emphasised the importance of validating data and ensuring consistency for the Safer Nursing Care tool. Using a standard approach for a patient dependency tool was found to be a reliable method for delivering evidence-based workforce plans. 206 Furthermore, van Oostveen et al. 124 pointed out that a standard approach to a staffing system facilitated comparison:

. . . the RAFAELA system offers a fully Information and Communication Technology supported and uniform system for all clinical nursing wards, which facilitates a clear communication about nursing care intensity on all management levels throughout the hospital, and even on regional and national levels.

van Oostveen et al. 124

One interview participant reported that data from WPTs enabled comparisons in English providers:

. . . and we upload that as part of the Unify process. And that is then seen on the model hospital, and that allows us to benchmark with other organisations.

Site 2, participant 1

The evidence suggested that benchmarking data can confer organisational benefits. In a study by Robinson et al. ,126 a participant from an organisation categorised as medium for achievement in implementing WPTs said that the staffing methodology was a really good tool for benchmarking and that it was useful in the achievement of good-quality care. Benchmarking data can ensure data credibility for the calculation of costs per patient-day in organisations in which there are similar financial systems. 183 Finn206 found that, according to data from the e-roster, staffing unavailability or headroom was comparable to that reported in other organisations, at 28.8%. Fagerström and Rauhala210 found differences between organisations in costs and nursing workloads when applying the RAFAELA tool and suggested that the findings offer opportunities for organisational learning.

Outcomes

When an organisation’s culture is supportive and values evaluation, WPTs can produce standardised data on needs and resources to enable learning about safe staffing. Lammintakanen et al. 197 commented that:

Electronic information systems were seen as a shared effort of learning and development of practices among nurses and nurse managers.

Lammintakanen et al. 197

One interview participant indicated that she would monitor how other sites had implemented technology for e-rostering to determine staff capabilities:

But I think there are other areas that are building on reports that they’re going to be getting, and as they produce those reports and we see how useful they are I think we will be able to say that looks great, let’s do it that way. So, when we see more benefits coming out of it I don’t think it will be a big leap to go and say right, I want to do that now . . .

Site 3, participant 3

Data from evaluations can identify where quality improvements need to be made133,145,174 in line with agreed national standards. 147 Anderson et al. 134 indicated that data in the form of scorecards or dashboards are increasingly being used to improve performance. Juntilla et al. 192 reported that an evaluation of a data warehouse approach identified primary data errors, van Oostveen et al. 124 suggested that further education is required following the evaluation of the implementation of a tool, and Fenton and Casey147 claimed that evaluation of the results of the Safer Nursing Care tool could lead to changes in staffing establishment numbers. Furthermore, learning about safe staffing following the evaluation of WPTs can result in the formulation of action plans to improve their use and to deploy staffing more effectively. 129,195,200

Learning about safe staffing can also be facilitated through the comparison of data. The evidence suggested that data can be used for benchmarking between organisations. 124,126,129,131,147,171,179,180,182,183,210 Data can also be used for organisational benchmarking with similar organisations. 124,126,127,131,147,168,175,180,183,203,206 This includes using comparative data across specialties and different localities to promote learning. 131,210

Adaptation and refinement of WPTs may also ensue following learning about safe staffing. The evidence indicated how changes were made following trials and pilots: 124,126,174

The Dashboard is a work in progress. Many of the indicators that had originally been established were refined and will be revised based on suggestions from project participants and tested across the [organisation name].

Fram and Morgan174

Continuous evaluation is recommended, and this can contribute to the formalisation of organisational processes, which are important for nurse managers:129

. . . to determine what data they will generate from the system and what will be done with that data. Information is power, but only if there’s a process in place to interpret the data and then make policy and procedure decisions based on the evidence the system provides.

Crist-Grundman and Mulrooney169

When there may be misalignment

In less supportive work cultures, nurse managers were less empowered to challenge situations:

. . . a fear of punitive action prompted nurse manager silence.

Shirey et al. 187

If an organisation appears to be less supportive and does not have an evaluation culture, learning about safe staffing may be impeded. In the first instance, this may impede the implementation of WPTs to enable learning; Botter173 noted that the lack of a supportive culture impeded the use of a patient classification tool. Silvestro and Silvestro185 found that a lack of sharing of good practice, with no feedback loops between ward performance and ward objectives, hampered good practice with regard to rostering. When evaluation does occur, it is important to use findings to promote learning. Kane179 indicated how stakeholder feedback was not acted on in the trial period for a community tool, as the team leaders felt that stakeholders (caseload holders) were not fully engaged in the process. Kane179 reflected that this was a missed opportunity:

. . . to advance the knowledge of caseload analysis at all levels of the service.

Kane179

Some evidence suggested that evaluations are not effective. Murtola et al. 186 found that evaluations of information systems were mainly based on staff feedback of usability, but the authors felt that this was not focused on effectiveness or the impact on decision-making. van Oostveen et al. 124 found that, despite gaining positive staff feedback on the applicability and usability of the RAFAELA system, staff did not implement this system effectively. Furthermore, cross-organisation comparison is not always welcomed. One interview participant was concerned that comparison with other organisations may have the potential to reduce staffing:

. . . because we’re actually . . . we’ve got a very richness to our skill mix and we’ve got a very good fill rate, so there’s a fear that when we benchmark with other organisations, similar organisations, we might be seen as being much better than others in terms of staffing. And we’re slightly concerned about what NHSI [NHS Improvement] will say if a range, or a suitable range . . . for this type of ward, and if we’re over that what does that mean on our current staffing. So, I think they would be our concerns, you know.

Site 2, participant 1

Context and mechanism

Several sources within the synthesis of the literature indicated the importance of collaboration and partnership to design or select a staffing tool or technology. 126,127,131,134,146,147,154,169–171,174,175,183,190–192,194,195,197,199,200,203,208,209 Collaboration and partnership appear to be generally important for the NHS managers’ role around safe staffing. Dent130 emphasised that a shared vision is important for improved nurse staffing. Effective communication between nurses and other relevant groups appears to be important for partnership approaches to nurse staffing. In nursing, a team approach to care can promote communication. 121 Interview data and NHS manager stakeholder feedback emphasised that effective communication is important to manage the ever-changing environment for staffing:

So, we’re constantly talking, constantly kind of chatting about it, because it changes all the time. With, you know, in excess of four hundred staff it’s constantly moving, staff are going off sick, staff are returning to work, you know, it is always changing. Every shift, it changes.

Site 3, participant 3

The NHS manager feedback and the PPI stakeholder feedback revealed that managers need to be effective networkers. Szumlas204 suggested that nurse managers can create a culture of partnership and accountability to ensure staffing excellence through ‘consistency, dialogue, and idea and data sharing’. Effective collaboration and partnership in the design, selection and development of WPTs can promote shared accountability and ensure that systems are ‘simple and efficient to adopt and use’. 169

The synthesis of evidence strongly suggested that collaboration with multiple groups is essential for the achievement of an acceptable system, with collaboration being important for technology design. 209 Key stakeholders need to work together to design and develop WPTs. These groups include nurses and nurse managers from all levels of the organisation,126,146,154,168,175,192,199,203 expert panels,126,127,146,175 IT specialists,147,192,200 finance partners,174,195 human resource personnel,168,195 administrative staff195 and researchers. 146,171 Furthermore, linking with vendors or industry is important to develop systems that fulfil organisational requirements. 170,175,191,192,197,199,208 Douglas199 indicated that external technology consultants may be useful, but emphasised that it is important for nurse leaders to engage with the process and not to leave it up to the consultant.

The literature strongly recommends a prominent role for nurses, both nurse managers and staff nurses, in the developmental process. 174,190,194 Douglas199 advocated the formation of a multidisciplinary staffing committee with strong nurse representation to drive selection and implementation strategies, but noted that it is important that nursing staff (and not only the IT department) take ownership of the tools. 199 Robinson et al. 126 also found that a staffing methodology was more likely to be implemented if staff were included in the process. Collaboration with clinical nurse experts appears to be important in order to ensure the inclusion of data that influence nurse staffing. 146,192 Junttila et al. 192 used a nursing expert group, consisting of nurse directors, head nurses, experts in clinical nursing and hospital informatics, to determine management information needs in designing an information system for a hospital data warehouse approach. This integrated different sources of information of relevance for staffing decisions. Valentine et al. 190 reported that collaboration on the development of the OpenShift self-roster tool (Red Hat, Inc., Raleigh, NC, USA), for staff to self-select shifts, involved multiple partners across departments, from top leadership to front-line staff. Wadsworth and Kurilla194 included nurses in the selection decision for a staffing tool in a Magnet-designated hospital with a collaborative, inclusive philosophy:

. . . so before the final decision was made, we knew it was important to let the nurses weigh in. We asked them to also score the vendors and share their opinions. Their judgments aligned well with the opinions of the evaluation team.

Wadsworth and Kurilla194

Different formats for collaboration have been used to select, develop or design WPTs. Fasoli et al. 146 used an interdisciplinary team, comprising clinical experts, leaders and researchers, to develop a staffing methodology. Barton168 reported that a cross-disciplinary team of finance and human resources representatives, nursing unit managers and the staffing director collaborated in the development of a new staffing programme technology and suggested that this team approach was a major factor in its success. Diers et al. 154 used a pilot group of 14 nurse managers with operational experts, finance experts and computing experts to design a system of user-defined decision support tools. There were many examples in the literature of different processes for collaboration in the design, development and selection of WPTs. Wallace and Pierson195 established a corporate working group: the Best Practice Scheduling Task Force. Pruinelli et al. 175 pointed out that the NMMDS was developed via expert review panels, focus groups and national Delphi surveys to identify the core variables needed by managers to make decisions. The Scottish toolkit was formulated following a national conference to consider the workload and workforce tools that are appropriate for Scotland. 203 The evidence from interview participants also indicated the importance of collaboration. One interview participant believed that partnerships between NHS managers and IT teams help to ensure that the e-roster is fit for purpose:

And they meet on a fortnightly basis with one of the lead nurses, who goes through it with them, with the e-roster team. And with the work with the e-roster team, and getting the rules set correctly in the background on e-roster, the ward managers are actually finding it much more helpful now.

Site 3, participant 4

Defining need for staffing decisions

The evidence indicated that a key purpose of collaboration is to define need and establish goals for WPTs. 125,127,134,146,171,178,195,199 Other requirements included identification of the critical data elements and sources of data. 125,134 Douglas199 advocated gaining a clear understanding of the organisation’s current staffing and scheduling practices before choosing a tool:

This will formalize the picture of what is working and what needs to be fixed. It will also set in motion the involvement and support that will be needed later. From your assessment of current state, identify the goals you want to achieve.

Douglas199

The NHS manager needs to have information on multiple elements to make safe staffing decisions; these relate to the wider systems in which WPTs need to operate. According to Kivinen and Lammintakanen,182 the use of IT to aid operational decision-making, such as around staffing, is challenging, as health-care environments are complex because:

. . . rapid changes in the operating context and characteristics such professional culture, complex organization structure and management system.

Kivinen and Lammintakanen182

Burns205 suggested that health-care organisations are made up of complex adaptive systems whose evolvement cannot be predicted. Therefore, it is unsurprising that nurse staffing is also complex, given the number of inter-related systems and elements that need to be considered. For example, Dent130 formulated nine principles that relate to staffing decisions, with each of these being linked to different organisational systems, such as finance and staff recruitment. Pruinelli et al. 175 also explained how the NMMDS combines information of different systems, such as human resources and staffing systems, with electronic health records and information on the clinical area. This information is used to support the decision-making process and measure the impact on patient outcomes.

A collaborative approach can result in an enhanced understanding of the systems and elements that have an impact on staffing decisions. In developing a new scheduling scheme, one task force group devised a questionnaire for clinical managers and clerics involved in scheduling, in order to understand the ‘reality of scheduling on the front line’. 195 Fasoli et al. 146 reported on a consensus approach for the Department of Veteran Affairs to determine design principles for a nationally standardised nurse staffing methodology; following an evidence review and brainstorming session with an expert panel, indicators were agreed on to reflect nursing work, patient-centred variables, nursing characteristics and unit and hospital attributes. Also in relation to the Veteran Affairs staffing methodology, Taylor et al. 127 highlighted that this collaborative process helped to develop nurse-sensitive indicators, in the light of unit workload and patient group, to generate nursing full-time equivalent requirements. Kivinen and Lammintakanen182 also advocated collaborative discussion for nurse management systems, in order to identify the information requirements of different groups of managers and to establish common concepts and meaning; this was to promote more effective use of the information. van Oostveen et al. 178 found that there were different information requirements for different clinical areas, although patient acuity was consistently the most important. Furthermore, interview participants highlighted that collaborative processes were needed to ensure that nursing needs were firmly embedded within WPTs:

So, the e-roster team are from an IT background and I think that was the problem really to start with.

So, it was just really not understanding the needs for that . . . for being a nurse?

Partly. And I think for the ward managers, who did the work with them at the time, not recognising the importance of all the information they needed to provide. And I think it’s only when you start using the system that you realise that it is essential to get that information absolutely accurate at the beginning.

Partnership for trial, adaptation and innovation

Probably the strongest aspect of the project was its collaborative nature, as staff nurses and management staff worked together on a matter of high importance to both, providing valuable recommendations for shared nursing and interprofessional planning, Dashboard development and project management.

Fram and Morgan174

Collaboration appears to be important for the trial and adaptation of WPTs. 147,163,174,190,191 This can mean that WPTs can be effectively adapted for specific groups of patients or clinical areas,163 or that elements can be refined or added in order to improve validity. 174 Valentine et al. 190 discussed how a self-rostering technology was customised to organisational needs. In addition, for a specific self-rostering system [BidShift (Cerner, North Kansas City, MO, USA)], Brown191 charted how collaboration with the vendor resulted in a trial and eventual acceptance of differential shift payment incentives for hard-to-fill shifts. Furthermore, effective communication appears to be essential to the process of engaging and collaborating with staff to work in partnership to develop WPTs:

Developing a solid communications plan at the beginning of the project helped solidify the necessary relationships and environment for moving the project forward.

Valentine et al. 190

Engaging nursing staff to collaborate was an important part of implementing a self-rostering system; it was reported that regular communication, with the opportunity given to staff to ask questions about the new system, resulted in staff feeling that they had ownership in the process. 191 Fenton and Casey147 gave an example of a health organisation in which:

The trust gained the cooperation of staff by involving nursing staff from the outset, and encouraging open discussion of the advantages and disadvantages of using the tool.

Fenton and Casey147

The importance of trialling WPTs is evident in the literature,124,127,144,146,161,174,176,179,192,197 with stakeholder engagement and feedback being recommended by many sources. 126,127,144,174,179 Ruland171 used multidisciplinary team focus groups to design a staffing tool and emphasised that end-users need to have the time to identify and communicate their requirements clearly to system designers to promote continuous learning. Collaborative partnerships led to identification of areas for further development of the design of WPTs. Adaptation and tailoring to need were recommended or evident in many sources of evidence124,126,127,144,147,163,171,176,179,186,200,202,203,208 Murtola et al. 186 suggested that systems should be flexible so that they can be modified on the basis of user need. A collaborative process for a practical guide for patient assignments resulted in local area and population modifications in order to support the flexibility of the tool. 163 Robinson et al. 126 found that managers made recommendations for improving a staffing methodology tool, such as adding family leave. Fram and Morgan174 indicated the importance of acting on feedback for a dashboard project that displayed integrated staffing data for retrospective and projected views for staffing decisions:

The Dashboard is a work in progress. Many of the indicators that had originally been established were refined and will be revised based on suggestions from project participants and tested.

Fram and Morgan174

Effective collaboration may also result in innovation and in-house design. There are some examples in the evidence of the development of in-house staffing WPTs. 154,161,174,176 Fram and Morgan174 found that a collaborative approach to the dashboard project resulted in a practical tool that facilitated learning, as participants gained an appreciation of the depth of knowledge required for decisions on staffing numbers. Wyatt and Healey161 developed a system to escalate and monitor clinical capacity (STEAM) for children’s services; staff reported satisfaction with the tool, as they felt more in control, supported and confident in contacting senior staff. Diers et al. 154 reported how the user-defined data analysis and reporting tool was collaboratively designed to fit the way nurse managers think about the work; nurse managers were quick to understand the system and the importance of their role for valid data input. Pappas et al. 176 discussed how a collaboration of clinical nurses led to the design of a patient risk assessment tool. This emphasised the importance of closer nursing surveillance, as patients at higher risk of experiencing clinical events could be assigned nurses to mitigate the risk. The impact was an improvement in nursing-sensitive indicators and cost-savings, particularly in end-of-shift overtime. Pappas et al. 176 reported that in using the tool, nurses felt empowered, as patients’ clinical conditions were factored into staffing decisions and the nurse was placed in the pivotal role for decision-making. In addition, one interview participant spoke of how their organisation had designed its own staffing software; the software was nationally acclaimed and indicated staffing requirements for NHS managers:

We developed it in-house, so it was quite cheap. All the wards have iPads [Apple Inc., Cupertino, CA, USA], and on the iPad they have an app [application] that allows them to put in actual staffing compared to planned staffing.

Site 2, participant 1

Collaboration for communication and decision-making

In enhancing NHS managers’ understanding of staffing needs and resources, WPTs are proximal to decision-making processes. An increased understanding of staffing resources and requirements is facilitated through communication. Some evidence from the literature suggested that embedding WPTs in local practices and systems can promote collaboration through enhancing communication about nursing care. 151,175,186,199,209 The widespread use of handheld computer devices, e-mail, text messaging and other communication devices to alert NHS managers to staffing issues was evident from stakeholder feedback and interview participants. There were also examples in the literature of the integration of communication technology in order to enhance staffing decisions. A nursing management data set system was integrated with communication and alerting devices to ensure effective communication. 175 Kontio et al. 151 indicated that instant messages and text messages ensure timely data-sharing and support effective communication within organisations.

van Oostveen et al. 132 also noted that effective staffing systems may enhance communication for nurses:

Nurses desired a valid nursing care intensity system as an interdisciplinary and objective communication tool that makes nursing care visible and creates possibilities for better positioning of nurses in hospitals and further professionalization in terms of enhanced authority and autonomy.

van Oostveen et al. 132

Good decision-making is fundamental for effective staffing, and collaboration also appeared to be important for decision-making. 159 Data from WPTs inform decision-making. 171–174,180,182,184,186,210,212 Kivinen and Lammintakanen182 indicated that technology is important for managing information to use in decision-making by nurse managers in their daily work. Effken et al. 212 indicated that computerised decision support tools may facilitate nurse managers’ decision-making:

. . . by encouraging systemic, systematic, efficient decisions based on, not only a careful assessment of the nature and causes of the problem, but also an evaluation of alternative solutions to achieve the desired outcome.

Effken et al. 212

When there has been partnership and collaboration in the design of WPTs, the data are usable and near to the decision. 171,173 Following collaboration in the development of a patient classification system, Botter173 concluded that its primary purpose was to generate information for decisions about the provision of available nursing resources. Nurse managers used this patient classification tool for decision-making in changing or unusual situations. In addition, following partnership in development, Ruland171 felt that the CLASSICA decision support system offered a better understanding of financial processes that were proximal to the decision-making process. The advisory group indicated that data from WPTs inform decisions; managers triangulate different data sources to make professional judgements on skills, attributes and skill mix to get the right nurse for the patient. The ability to integrate professional judgement with technology was also reflected in the in-house designed app discussed by one interview participant:

But the key thing about it is it allows them to add in whether they, in their professional judgement, so that’s the nurse in charge, not admin, not matron, but in the nurse in charge, to say whether they feel safe or not.

Site 2, participant 2

One interview participant commented on the process of integrating decision-making with data from a tool:

. . . our view is always that it requires a common-sense approach when you’re looking at the Safe Nursing Care tool anyway. So, it gives you raw numbers but you need to ask some questions behind it . . . to then be able to make decisions based on staffing. So, I don’t think the systems necessarily help us to make decisions about staffing, but they certainly allow us to access the information we need about the staffing . . . to make those decisions.

Site 3, participant 4

Collaboration for a user-friendly system

Technology can be used to help NHS managers with decision-making with the aim of improving quality and safety outcomes, but it must fit into nurse managers’ workflow and be easy to use. 145 For decision-making, data should be presented in a format that can be understood. 183 The evidence suggests that, when a collaborative process has been instigated, systems can be selected and adapted to be user-friendly. 170,172,183,190,192,197,202,203 It was recommended that a collaborative process for tool selection should focus on user-friendly technology, with automated and balancing functions and intuitive software for use. 170 Following a collaborative process to develop a nursing MIS (CLASSICA),171 Ruland and Ravn172 found that nurse managers reported a high level of satisfaction with this information system, including the perceived ease of use. Junttila et al. 192 also used a collaborative process to implement and evaluate a nursing MIS and were able to make recommendations to increase its usefulness for the nurse decision-making process using a data warehouse model. Flynn et al. 203 charted how the collaborative process for WPTs in Scotland culminated in a suite of tools for selection, to be used alongside professional judgement. Evaluation of these tools showed fitness for purpose, but also areas for refinement, with Kellagher et al. 202 reporting user satisfaction and skills development on staffing and budgeting, following the pilot studies. One interview participant indicated that an e-roster had been made more user-friendly by users working in partnership with the IT team:

It’s easier to use now since it’s been updated. It was much harder before, it was a lot more time-consuming to do small tasks on it. So it’s, more, a lot more user-friendly now, the new version we’ve got.

Site 1, participant 1

Outcomes

NHS managers who find WPTs useful exhibit a change in their thinking, and this has an impact on how the WPTs are used, placing them close to the decision-making point. The literature contains examples of managers’ attitudes towards WPTs changing as they come to understand their utility for staffing decisions. Kane179 found that managers using an eCAT tool started to challenge the caseload data. The tool enabled this to be done in an objective and systematic way, in order to facilitate equity of caseload allocation within the team with. Diers et al. 154 found that managers grasped the ideas from decision support systems and wanted to use the technology. In a pilot on the implementation of the Scottish toolkit, Kellagher et al. 202 found that managers understood finance and human resource issues, developed budgeting skills and felt empowered to challenge staffing resources and change their ways of working. Furthermore, it was found that managers thought that electronic information systems strengthened their role and enabled them to fulfil their tasks more easily. 197 Finn206 observed that with an e-roster:

Ward managers have informally reported staff spending an increased amount of time with patients due to better allocation of leave, greater transparency of rosters, challenging historic working practices, accurate recording of under/over-contracted hours and reviewing of break and shift times.

Finn206

Furthermore, Effken et al. 145 suggested that decision support tools allow nurses to test their hypotheses about staffing solutions in a safe virtual environment. Effken et al. 145 proposed that decision support tools may facilitate the evaluation of decisions and considerations of alternatives to facilitate more long-term solutions rather than just satisficing:

By stimulating higher level thinking, managers might be less likely to engage in ‘fire stomping’ or jump to quick solutions without considering the larger picture.

Effken et al. 145

When there may be misalignment

Some of the evidence suggested that the utility of WPTs is hampered when nurses have not been involved in the collaborative process effectively. Murtola et al. 186 discussed this issue:

But still a discrepancy exists between the information systems needed by nurse managers and the existing ones. One possible explanation for this is that nurse managers might not have been able to communicate their needs.

Murtola et al. 186

Other reasons for a lack of effective collaboration are also evident in the data. Fram and Morgan174 found ongoing challenges in scheduling regular dashboard project development meetings in order to engage nurses, as nurses were unable to leave the wards because of staffing shortages. A lack of understanding within partnerships may be another issue; one interview participant expressed the belief that ineffective collaboration resulted in poor nurse allocation:

That was all part of the problem to start with really, about getting that information right about the department, so what are the skills that your staff should have in that department, and identifying the skills that the staff then do have. Initially, I think perhaps a lack of understanding on the e-roster team’s perspective . . . and working with the ward managers, it wasn’t . . . it wasn’t as accurate as it needed to be. So, it would be rostering staff on together that didn’t cover the skills that needed to be covered. But it was really about the rules in the background, it wasn’t about the actual system that was a problem.

Site 3, participant 4

Weak partnerships may hamper staff engagement with WPTs. In a pre-implementation study of the RAFAELA tool in the Netherlands, van Oostveen et al. 132 found that, despite the reliability and validity of the tool, it was not well accepted by nurses. van Oostveen et al. 132 felt that they should have focused more on a social context, such as strategies to enhance team functioning. Poor interprofessional communication may also be a problem; Lammintakanen et al. 197 found that a lack of collaboration between medicine and nursing impeded the use of WPTs. Kivinen and Lammintakanen182 also found that a lack of collaboration resulted in a failure to identify information needs at different levels of management; this had an impact on data utility for staffing decisions:

This emerged from descriptions that the information needs of managers working in different positions and units were not sufficiently identified and discussed and middle and front-line managers especially argued that MIS [Management Information Systems] do not enhance their daily work and decision making.

Kivinen and Lammintakanen182

Context and mechanism

Following reports on high-profile care failings,2,3 there are policy requirements in England and the rest of the UK to provide real-time quality assurance data on staffing levels for central scrutiny, such as the UNIFY safe staffing data collection in England. NHS organisations also have a legal duty of candour under the Health and Social Care Act regulations;75 current staffing levels should be displayed at the entrance to wards and departments, where the information can be viewed by patients and the public. 76 Hockley and Boyle166 indicate how WPTs have the potential to provide clarity on the reporting of these data, and highlight views from senior nurses about increasing public confidence, which may accrue from the availability of data. 166

Perspective of safe staffing

According to Flynn and McKeown,165 staffing and skill mix are central issues for health service governance, service users and the recruitment, retention and well-being of nurses across different sectors; however, they note that staffing levels and skill mix have to be defined in terms of sociocultural, political, economic, professional and service user contexts, with conflicts between the perceptions of safe staffing in different groups. From an organisational perspective, Flynn and McKeown165 suggest that directors of finance may have a different view of safe staffing from nurses. Effken expands on the theme of different perspectives, pointing out that professionals are the primary workers for care delivery and may take a role in organisational management:

This means that the primary workers have considerable latitude in how the organization’s work is conducted and also that the primary workers are more likely to have humanitarian values that may conflict with the organization’s commercial or financial goal.

Effken209

The evidence from the literature linked staffing to quality and safety. 128,132,133,158,161,162,164,165,167,176,180,198 Subirana et al. 128 linked safe staffing to patient outcomes for quality and safety, but pointed out that quality in elements of nursing care is multidimensional and difficult to distinguish from the overall health-care system. Flynn and McKeown165 commented that quality and safety in staffing are key areas for scrutiny from the public:

. . . there is a growing influence of service user movements in health and social care, which are perhaps equally interested in the impact of nurse staffing levels and skill mix on quality of care.

Flynn and McKeown165

The interview participants also noted that patients and the public are interested in staffing issues and link staff numbers with the quality of care:

So, now people are . . . are more aware that the . . . when you look round the ward and you’re looking at the number of staff, that that somehow equates to whether or not your mum and dad are being looked after properly. But what there is at the moment is that there is an incredible absence of understanding about what that actually means.

Site 1, participant 4

Understanding the public’s perception around safe staffing levels is complex. Flynn and McKeown165 suggest that service users have changing expectations and the nursing profession needs to determine appropriate levels of nursing care in the light of patient outcomes. One PPI stakeholder termed their perspectives on staffing shortages in stark terms:

If you are supposed to have four people on an ICU unit and you only have two, people are going to die.

PPI stakeholder

Furthermore, in the interview data, participants indicated how patient perspectives on staffing may differ from establishment-level interpretation:

Well it’s quite interesting patient perceptions really because that, we do an audit, the quality and safety audit every month, and, we’re only struggling with staffing in the last, this couple of months really, we’re quite lucky otherwise that we have been fully staffed, but the patients will always comment in the feedback, ‘I feel they need more staff, there’s not enough of them’, so they obviously feel that we need more.

Site 1, participant 1

Low or unsafe staffing levels have been linked with adverse patient and organisational outcomes. 144 The public may therefore lack confidence in organisations that have gained a reputation for poor patient outcomes that are linked to unsafe staffing levels. Certainly, Hockley and Boyle166 highlighted that recent reports on care failings related to the Mid Staffordshire enquiry centred on staffing issues, although the political focus was still at the micro level (nurse–patient interface) rather than the macro level (organisational processes). 2,166 Flynn and McKeown165 suggested that patient groups may join resistive alliances with nurses to challenge organisational practices on staffing. The public may also perceive that poor staffing is a consequence of underfunding and/or poor management:

When I have gone round visiting patients I have never heard a complaint about the nurse, I have heard complaints about management and short staff. From what I have come across there is always a lot of sympathy for the nurse, however at the same time, the patients anxiety levels must be raised if they know there is a shortage of staff.

PPI stakeholder

Barton168 suggested that administrators balance cost, quality, satisfaction and staffing. Balance is also referred to in the literature related to other public services; Wilson and Weiss153 commented on the sensitive balance between funding for staffing and public satisfaction in the police. The PPI stakeholder feedback reflected an understanding of the need to save money, but also strongly asserted that the NHS needs money.

Safe staffing, accountability and risk

Accountability and responsibility were linked to safe staffing;126,127,129,130,143,162,163 this can be seen to be intrinsic to organisational and professional objectives for the achievement of good-quality care. The staffing roster links to the mission of the organisation:

. . . every patient should be assured that there be skilled, adequate staff to provide safe, competent care.

deSilva et al. 170

Wallace and Pierson195 underlined that nurse managers have ultimate accountability for staff scheduling and are aware of the links between staffing deficits and patient outcomes. Furthermore, nurses are professionally accountable for raising concerns when patients are at risk, with professional bodies driving for a culture of openness and transparency to facilitate staff to speak out on safety concerns. 166 The advisory group commented that the organisational culture for managing risk may be at odds with professional values. As noted, nurse dissatisfaction with care standards may have an impact on organisational recruitment and retention, which further compounds poor staffing levels128,134,185,193 and potentially has an adverse impact on the organisational reputation. 185

Staffing skill mix may have more resonance with nurses than staffing numbers in relation to safety and risk management. Flynn and McKeown165 suggested that a diluted skill mix may present a threat to the professional concept of the nurse, with nurses being concerned with protecting the position of the nurse from encroachment by health-care assistants or other health-care practitioners. Skill mix remains a contentious issue for its impact on patient outcomes. 133 Nevertheless, for the public, the visibility of staffing numbers may be a source of comfort:

At the [University hospital in England] it was so marvellous to see more staff. You could tell the nurses weren’t under pressure.

PPI stakeholder

However, in the case of RNs, staffing levels may still be seen as unsafe if support workers or agency nurses are used to make up the numbers. The evidence suggested that an increasing use of agencies may have an impact on patient safety outcomes,133,160 and can have an impact on efficiency and continuity of care issues. 129 Conversely, low staffing levels may increase the use of an agency in nursing, despite the adverse financial impact of agency costs. 129,157,170,181,184,185,190,191 Wilson and Weiss153 also noted this phenomenon in the police literature. Rischbieth160 reported that nurse managers in intensive care felt pressurised to use agency nurses without knowing their capabilities for such a specialised area, and proposed that this may pose a risk. 160 Davidhizar et al. 189 found that nurses and nurse organisations often oppose ‘floating’ or working in an understaffed area, as it may put them in litigious or vulnerable situations, as nurses may not perceive that they have the relevant expertise required for optimal patient care. Consequently, high staff numbers may be reassuring for some patients, but suboptimal skill mix or capability may have an impact on nurses’ confidence levels and potentially on patient outcomes.

Many authors have described how staffing is integral to risk management. 150,160,176,198 Clark et al. 193 saw rostering as a strategic issue and suggested that unfilled shifts are a risk, with the potential for negative consequences. Safe staffing is part of both organisational and professional risk management. 193 According to Rischbieth,160 organisational attendance to risk and associated issues of patient safety is paramount, in the light of increased patient acuity and global nurse shortages. 160 Corporate indicators to gauge risk were apparent in the data; for example, Fram and Morgan174 discussed how a traffic light system was used in a dashboard to determine if key staffing data were acceptable at corporate levels, with red indicating when the levels were above or below the acceptable targets. Rapala198 stated that:

Certain interventions must be put into place to manage that risk, and staffing effectiveness is a critical part of clinical risk management.

Rapala198

The evidence from the literature shows that nurse managers focus on potential risks that could have an impact on patient safety. 150,176 Allen158 discussed how nurse managers make decisions on nurse–patient allocation while understanding the impact on patient safety. Shirey et al. 187 found that nurse managers considered how their decisions would have an impact on patients’ safety and their satisfaction with quality standards. Wilson et al. 150 described how nurse managers anticipate and manage staffing imbalances, such as high patient-to-nurse ratios, in relation to patient acuity:

This chronic wariness is defined as preoccupation with failure, in that at any moment a latent problem may erupt.

Wilson et al. 150

The evidence thus suggests that nurse managers are preoccupied with the potential risks from inadequate nurse staffing and their impact on patient care. One interview participant indicated that staffing deficits had an impact on the quality of care delivery:

. . . they [nurses] don’t have the time, they just have to get through, get the basics done for the shift and don’t always have that extra time to sit with them and do those extra things, and you do see the standard, the standard of paperwork, that drops when you go through periods like this and the acuity’s high, so it does have an impact.

Site 1, participant 1

Patient and public involvement stakeholder group members identified the possible detrimental impacts of staffing shortages on patients:

If patients see that the ward is over stretched, it would have to increase the stress levels of both patients and relatives I would think.

PPI stakeholder

I think that one thing that we have not been discussing is ‘the guilt of the patient.’ If patients were lying in bed knowing the ward is short staffed, would you feel guiltily asking for something? That could play on the minds of some people who don’t want to be a nuisance, especially those of the ‘older’ generation! We would just probably lie there, wet the bed or anything rather than ask for help.

PPI stakeholder

Disclosure of data on staffing and the manager’s role

Staffing data have not always been found to be robust, owing to the dynamic clinical environment; some authors noted a difference between the roster and the reality of staffing levels. 143,166 In a study in Malaysia, Drake143 found potential differences between rules on shifts and staffing and reality:

. . . published rosters were approved as safe and fair, but the actual safety, fairness and cost-effectiveness of the worked roster were unknown.

Drake143

Organisations do have requirements for public disclosure for scrutiny to determine financial processes and facilitate comparison with other organisations. One interview participant commented on this disclosure of data (all trusts in England are required to upload every month, through UNIFY, data on actual staffing and planned staffing):

And it’s about ensuring that trusts measure in the same way, and that they upload their data to be visible to NHSI and to the public.

Site 2, participant 2

It has been suggested that the use of WPTs may provide more accurate data. 131,166 According to Hockley and Boyle,166 that reliable systems to record workforce timekeeping will potentially have an impact on safety and efficiency and monitoring routine staffing data should be tied to visions for good-quality patient care. One interview participant held that a range of data relating to risk management, quality improvement and patient satisfaction and staffing can be accessed via technology:

When we receive complaints, all that is held at a central point, either through [patient feedback system] or through our formal complaints department, . . . we encourage our nursing staff to submit incident forms when the staffing is suboptimal . . . within the department, to the point that they feel it’s impacting on the ability to provide care to the standard that we’d want to. So that information is then accessible to us by being drawn together, and requesting that information through the systems that it’s held.

Site 3, participant 4

When organisations are required to be candid about staffing levels, it is at the ward or department interface that patients and the public will encounter indications of staffing levels:

So, we have our data at the door, the information is out . . . is available for visitors, parents, on a daily basis of what our staffing levels should be and what we’ve actually achieved for the day.

Site 3, participant 4

However, public perceptions may go beyond numbers to satisfaction with experience. Both the interview participants and the PPI stakeholders noted that ward activity has an impact:

All I would say, if your staff is running round . . . the perception is that it’s unsafe . . .

Site 2, participant 2

Managers forget that [hospital] patients are there 24 hours a day, they don’t need to be told they are short staffed, they can see they are short staffed because they are running around like headless chickens. Patients are underestimated in that sense, they have got nothing else to do but look at around at the staffing and they can see if they are short staffed.

PPI stakeholder

Nurses, and nurse managers in particular, are likely to be the source of communication with patients and the public on staffing levels and, as such, represent the organisation. PPI stakeholder feedback indicated that NHS managers should have good customer service skills, but acknowledged that conveying information about staffing shortages is sensitive:

Staff need to be prepared if a patient or their relatives asks about staffing. They need to be aware of how they communicate so they don’t make patients fearful. Provide reassurance, ‘yes we are short staffed but we are here if you need us so please ask if you need us’. Better to tell on person-by-person basis, it’s more personal.

PPI stakeholder

Shirey et al. 187 found that nurse managers comply with organisational aims:

Nurse managers quickly learned to go with the desired agenda.

Shirey et al. 187

Despite having a duty of candour, imposed by both professional and organisational requirements, nurse managers are likely to be mindful of patient experience and expectations for good-quality care. Shirey et al. 187 also noted that ward managers tend to customise messages to staff from senior management to frame issues in such a way that they will be well received. Although the framing of messages on staffing to patients is thus a possibility when staffing deficits occur, it may be argued that a lack of transparency of communication does not meet professional obligations. However, the duty of candour may possibly have limits in order to preserve the organisational reputation and ensure patient confidence. Certainly, the evidence noted that nurses can ‘game’ to achieve higher staffing numbers. 125,143 Anderson et al. 125 found that nurses using a patient classification system did not score nursing intensity as low during periods with low activity for fear of management sanctions. One interview participant described how nurse managers ‘cheated brilliantly’ as they used the data from WPTs to determine where they could find staff without declaring:

So, they’ll ring each other up and they’ll say ‘How are you doing?’, and they’ll say ‘Well I’ll send you a registered nurse and you send me an NA [HCSW], because I’ve got this . . . something special in the cubicle and you’ve got this other thing. But I’m going to go to admissions for a bit, but we won’t tell the bed managers we just . . . we just won’t . . . won’t affect your admissions if you can take them for the next couple of hours’. ‘Yeah, yeah, we can do that’, and you think . . .

Site 1, participant 4

However, Shirey et al. 187 found that nurse managers recognised that their actions had consequences, particularly in negative working cultures, where they identified unspoken lines of how far they could challenge staffing decisions. 187

The NHS manager stakeholder groups, the advisory group and the interview feedback suggested that NHS managers work hard at ensuring that patients and their family members have confidence in the system, through the use of effective communication and acting on patient feedback:

We have a system called ‘Tell us today’, we have the X [patient feedback system] department, and we have a formal complaints process. And, as ward managers and matrons, we try and meet with families just to check that everything is okay, by going round the ward. And if parents are raising that concern then we will . . . we will look at the care that’s being provided.

Site 3, participant 4

Furthermore, the interview participants identified the importance of paying attention to patient concerns around staffing and how this is often related to the expectations of patients or their families not being matched by the service provided, which culminates in potential complaints:

So, we would have then a very transparent conversation with the family about what is achievable and how we can . . . how we can achieve it, but actually then work with them and monitor that . . . that with them over the next few days to make sure we’re achieving it to a standard that they feel is okay.

Site 3, participant 4

In some respects the patients often focus on staffing as an issue, but it’s not necessarily staffing, and it’s about engaging patients to see ‘well this is how it should be with the set amount of staff we’ve got’.

Site 1, participant 3

Outcomes

The patient experience is a key priority as part of a drive for improved quality and safety,145 with NHS Education for Scotland131 stating that management decisions on staffing can be judged in terms of patient outcomes and patient satisfaction with nursing care. Allen158 indicated that a key priority for decision-making on the nurse–patient assignment was to ensure the best possible experience for patients and families. The data from the literature indicated that ensuring patient satisfaction was important for nurse managers:145,158

Managers spent much time daily checking in with patients and assuring their positive experience in the hospital because improving patient satisfaction was paramount.

Effken et al. 145

All of the stakeholders, and also interview participants, highlighted the major importance of gauging patients’ and families’ satisfaction in relation to staffing. For them the patient is the top priority. They highlighted that the patient should be the focus. Interview participants identified the importance of ensuring that the patient experience was good, and this was often linked to safe staffing:

. . . in our establishment reviews, we would take into account the . . . the friends and family element of what somebody’s saying about a ward. So, on paper we might say this looks okay, but a divisional nurse would say ‘But our patient experience or family experience is not so good’, and this is the justification of why you might need an additional role, or . . . additional numbers.

Site 2, participant 2

When there may be misalignment to the context–mechanism–outcome

Some evidence from the literature suggested that inadequate staffing levels are correlated with poor patient satisfaction. 134,145 The interview data aligned with this evidence:

We do get complaints about it, and we do get comments made. I mean you get those who will say ‘Oh, you’re all so really busy and I know you’re trying your best’, and then . . . you’ll get your families who really feel that care is being . . . compromised as a result of the nursing staff being busy.

Site 3, participant 2

However, interview data also indicated that some complaints, in which poor staffing was cited as an issue related to the quality of care, were often not substantiated when WPT data were reviewed:

In many now, what we do with complaints, if they, they do mention staffing, there’s about two that went out last week actually, it’s just, that’s why I can remember it quite clear, where they contributed the staffing, when I looked at the staffing they were actually above their template staffing. So they may have perceived it to be an issue, but it was necessarily wasn’t an issue.

Site 1, participant 3

Some stakeholder feedback suggested that displays of staffing numbers outside wards have little impact on the confidence of patients and the public:

If I was to arrive on a ward, having waited for 12 months to have knee replacement or hip replacement, and I see they are one staff member down, it would make no difference whatsoever.

PPI stakeholder

These display boards offer no context, what does that mean, how does that affect the quality of care? I wouldn’t know what the ramifications were. It’s a meaningless number.

PPI stakeholder

However, the majority of the PPI stakeholders strongly felt that displays indicating real-time staffing deficits were not a positive factor and that they would have a negative impact on public confidence:

If I saw a sign like that [indicating staff below establishment] that automatically means that there may well be a lowering of the quality of the care. It can’t mean anything else. Its common sense isn’t it.

PPI stakeholder

I think it raises more anxiety, I understand this idea for staffing levels to be visible, but I think the very idea of it being visible will raise people anxiety levels.

PPI stakeholder

I don’t see the benefit in telling the patients that you are short staffed, I imagine that would be counterproductive, because not only have you got the physical health to consider you have the patients mental health as well. If they really aren’t very well, they don’t need to hear that – I think it’s a bad idea.

PPI stakeholder

1. Qualify what is meant by ‘unsafe’ and ‘understaffed’

Managers initially sought to clarify what is meant by ‘unsafe’ and ‘understaffed’. NHS managers’ assessment of safety revolved around elucidating the nature and cause of the pressure(s), to determine if staffing levels had fallen and/or to determine changes in patient needs:

The first thing I’d be thinking about is, is wanting to know specifically more about why it’s unsafe, what are the risks, what are the specific risks.

Site 3, participant 5

In most cases, the managers reported that this necessitated visiting the ward and assessing the environment. Participants emphasised their sense of responsibility and accountability for safe staffing:

I would have to assess it myself before I could action anything, because obviously I am accountable and I would have to account for my actions.

Site 3, participant 3

Accountability and responsibility related to several CMOs; in particular, CMO 2 was linked to organisational strategic requirements for the promotion of the patient safety agenda. NHS managers focused on the potential risk ensuing from inadequate staffing levels and how this could affect patient care. The data supported CMO 2, because when organisational strategies align to promote safe staffing, they convey certain expectations on the NHS manager to ensure that this occurs. NHS managers understood that organisational strategies underpin good-quality care outcomes and felt obligated to uphold them:

We obviously respond to any new policy and strategy coming, and obviously, we have to adhere to those policies. So, it’s my role to really understand how we might use them and implement them within the trust.

Site 2, participant 2

NHS managers understood that a safety agenda is embedded within strategies, and this influenced their priorities for safe staffing. Establishment data from WPTs provide the leverage for negotiation or change in the light of organisational strategic goals:

A tool gives some consistency of approach and it’s only if you’ve got something objective and you can articulate clearly what the risk is that the best decision can be made.

Site 3, participant 5

Accountability and responsibility are also reflected in CMO 1, in which proximal and distal aspects of the staffing system need to integrate, so immediate staffing decisions have a positive impact on organisational reputation and future safe staffing. Safe staffing is integral to risk management and WPTs make needs and resources visible and highlighted staffing deficits, as articulated in CMO 4. NHS managers were strongly aware of the risks from staffing deficits in relation to patient outcomes and were mindful of their professional obligation to address safety concerns. Furthermore, this sense of accountability and responsibility links to CMO 7, as NHS managers understood their duty of care for transparent communication with patients and the public. In support of CMO 7, there was a strong focus on reinforcing patient confidence in the organisation’s ability to deliver safe, good-quality care. One participant stated that they would:

. . . reassure the patient’s relative or the patient that the nurse in charge of the ward has made an assessment and deems it to be safe.

Site 2, participant 2

Furthermore, NHS managers emphasised that the application of professional judgement is essential to interpret data from WPTs on patient needs and resources related to safety. Some NHS managers implied that they had ‘a gut feeling’ about whether or not wards were safe; although this may be linked to intuition, it is more likely to have its basis in the clinical expertise and local knowledge, articulated in CMO 4, that is integral to professional judgement. The WeChat#WeNurses audience emphasised that professional judgement is essential for safe staffing and suggested that it can be developed within supportive environments; this corroborates CMO 3, in which visible support for managers to develop skills can facilitate sense-making of data from WPTs for safe staffing decisions.

2. Assess nurses’ capacity and capability

The next priority of NHS managers is to ascertain the skills and training of their staff on duty; the aim here is to relate the real-time skill mix to the establishment figures articulated by the WPTs. This involves:

. . . looking at what skills the staff have got, what banding are they . . . making sure that each bay of patients has got a nurse in, so that your nurse-to-patient ratios aren’t massive.

Site 3, participant 5

Managers considered the competence of the RNs, for example who can take charge, who can administer intravenous medication, and equally of those in vital support roles, such as HCSWs, discharge co-ordinators and administrative staff. Determining the skill mix required to meet patient needs is vital, and this emerged as an important theme in WeChat#WeNurses. Furthermore, NHS managers need to know the capacity of their staff to cope under certain pressures; this often subjective understanding relates to their professional judgement and ‘knowing their staff’, as articulated within CMO 4. NHS managers recognised the skills of their team and were able to manage personalities and balance skill sets to ensure safe staffing deployment. In support of CMO 4, NHS manager participants reported that they used their local knowledge and the data from WPTs on real-time staffing resources to make decisions:

We can identify that we want a nurse in charge to be on duty, so the e-roster will flag up if there isn’t a nurse that we deem capable to be in charge. So, to some degree, you can then put it into the pattern that, say, you wanted three nurses who have the capacity to be in charge, that will give you some assurance that you’ve got the right skill mix on duty.

Site 3, participant 6

When participants used data from WPTs with their local knowledge, they did so in consideration of staffing outcomes. This corroborates CMO 1, in which proximal and distal decision-making need to align to have a positive impact on staffing outcomes. Through knowing their staff, NHS managers consider staff well-being and their concerns for safe care:

Because it’s very easy to go it’s OK, there’s six nurses, but it’s all about the quality of those nurses and actually are they trained to look after the patients that you’ve got. There are a lot of complex patients out there who need a wide variety of skills, we can’t just treat nurses as numbers, and they’ve got to be a nurse that’s trained to look after that patient.

Site 3, participant 5

NHS managers explained that they observe staff to see if they are coping and use their leadership skills to offer support and guidance. They emphasised the importance of working with nurses to evaluate the situation and to develop perspective and problem-solving abilities. They recognised that being supportive, compassionate and caring instils trust and respect, which helps to foster collaborative relationships for flexible staff deployment. This supports the importance of leadership attributes for greater influence on safe staffing, as articulated in CMO 4:

There is a particularly poorly patient that’s making her feel so on edge. And does she need extra support?

Site 2, participant 2

NHS managers acknowledged the importance of the investment of time and resources to support nurses and managers in their role; these elements were surfaced in CMO 3 and CMO 5, in which a supportive culture is important to enable learning about safe staffing:

If I didn’t feel, after what she’d told me, it was unsafe, because sometimes it’s about ‘do you think you’re unsafe or are you extremely busy?’ because there is a difference. So, it’s about using that as an opportunity just for them to reflect.

Site 2, participant 6

If they needed any further training, if we hadn’t highlighted it in their PDR, [professional development review] we would see how we could support them through that so they didn’t feel that they weren’t supported.

Site 3, participant 1

In support of CMO 1, the NHS managers articulated that their real-time proximal decisions, although predominantly driven by patient safety, reinforced distal safe staffing systems. Therefore, participants realised that their assessment of staffing capacity and capability had a strong influence not only on immediate patient safety but also on future effects on the organisational reputation and public confidence referred to in CMO 7.

3. Assess patient acuity, dependency and needs

The interview participants indicated that WPTs that offered accurate data on patient acuity, dependency and need enabled them to form a real-time account of staffing requirements in the light of patient need; this sense-making of complex data supports CMO 3. Participants suggested that WPTs can provide information for effective staffing decisions, such as whether to visit a patient in the community:

We also have what we call Chameleon, which is attached to the hospital so if one of our patients is in hospital and hasn’t come home we can see they had their bloods done this morning then we know they are still in hospital and we can see what their blood results are. It’s a really good system actually.

Site 3, participant 1

So, you would use the acuity and dependency scoring tool, so you’d say well, on this 28-bedded ward I’ve got so many level 1As and 1Bs, which would automatically go yes, that’s going to be tough.

Site 2, participant 2

However, the NHS managers articulated how they used the objective data on patient need and applied their professional judgement to inter-relate this to staffing decisions:

We’ve got an in-house system called Bed Man, so, that gives you patient demographic information, but also what’s going on with those patients at a glance. . . . I can use my sort of clinical judgement to see what’s going on with that patient.

. . . we have quite a pragmatic approach that we’ll look at each individual case and look at the staffing required from that.

Site 3, participant 6

The interview participants recognised that ineffectual staffing decisions based on data from WPTs that make no allowance for professional judgement can have repercussions for organisational morale, nurse satisfaction and retention. This underpins the importance of managers’ understanding of how proximal and distal decisions have an impact on the staffing system, as illustrated in CMO 1, and also supports CMO 5, in which a collaborative approach to the design of WPTs can facilitate usability:

Having been a ward sister and now a matron, there is nothing more patronising than a manager coming along and saying well I have gone onto my tool and you don’t need, you can manage on that number, but what that tool probably won’t pick up is the variance in acuity.

Site 1, participant 1

4. Draw on immediate internal resources from across the organisation

NHS managers reflected that, when staffing deficits are made visible by WPTs, they would initially harness the resources they have immediately available to them. Data from WPTs can make resources visible, as articulated in CMO 3, and this information can be used alongside professional judgement. Depending on the nature and cause of the pressure, NHS managers explained that:

. . . there are things that staff can do to make it better.

Site 1, participant 1

NHS managers explained that they use leadership strategies, which could involve implementing tactical or creative solutions, strategically switching and amending shifts:

Sometimes, it’s about maybe you need to cohort the confused patients into one bay, and then you can have one health-care assistant in that bay and alternate with another health-care assistant overnight.

Site 1, participant 1

NHS manager participants indicated that they used core leadership attributes, such as critical thinking, resourcefulness and decisiveness, when making decisions on staffing using WPTs; this affirms CMO 4. This process also interlinks with CMO 1, showing that a staffing system that makes efforts to respond to demand can positively reinforce distal aspects, such as the organisational reputation and integrity. NHS managers would think about seeking support from wards where staff have similar specialties and training to move staff, in order to ensure an effective skill mix, rather than just focusing on numbers. Good negotiation skills are critical in this process and part of the range of leadership skills needed, as highlighted in CMO 4.

Likewise, NHS managers’ flexibility to work clinically and ensure their physical presence on the ward increases their integrity and credibility as a leader; and offers a role-model for a flexible workforce. NHS managers’ ability to show support clinically for their team permitted greater influence in their safe staffing work and also supports the nature of leadership work in CMO 4:

The ward managers don’t routinely work clinically, but often they will flex in and do some work on the wards, you know, if need be, and that helps . . . well, two things, they can be a good role model, it’s clinical credibility, and also its maintaining patient safety.

Site 3, participant 6

Resourcing staff requires the alignment of organisational strategies, as articulated in CMO 2. When WPTs provide data for strategy achievement, they give leverage to NHS managers’ opinions, empowering managers’ negotiation for changes to staffing:

My plan is to try and get a registered mental health nurse onto the ward to support the staff because we do really struggle. Having that tool [Safer Nursing Care tool] will give us a better case, better argument to get that.

Site 1, participant 1

Central to this stage is the need for a flexible and supportive workforce. Building trust and collaboration across teams is an important agenda for managers, and reiterates CMO 5, which highlights the value of fostering a supportive culture:

Our vision is to try and make it a respiratory unit and over the last few years that is what we have been working on, so even the Christmas party, we would all go together and be together, and it’s evident from the internal staff rotations across to the other team wards for a few weeks, that we all feel quite comfortable with each other so it much easier to say ‘oh could you go opposite and help out for 2 hours’, it’s a little bit more flexible.

Site 4, participant 1

This theme also resonates with CMO 1, in which NHS managers’ efforts to build compassion and empathy within their team can have positive influences system integration to impact positively on staff satisfaction, retention and recruitment. Furthermore, the collaborative processes evident in the data appear to articulate the organisational learning on safe staffing in support of CMO 5.

NHS manager participants indicated that they valued WPTs for accessing staffing resources. The WeChat#WeNurses audience also reflected that WPTs made staffing data available quickly for remote access; they indicated that innovative collaborative practice had resulted in the development of WPTs that are user-friendly, and which connected staff looking for shifts and facilitated real-time shift swap. This theme is exemplified in CMO 6, and shows that tools need to be fit for purpose, designed with the users in mind, to influence managers’ attitudes and change how WPTs are used:

. . . yeah, it’s fantastic, it’s so useful and we are already planning for quite far in advance . . . it’s really visual and quite easy to use, it’s easy to manage staffing hours, staff can the obviously view the e-roster on employee online for themselves or they can do that on their phone its very user friendly . . . I would never go to work and not go on it and update it with something.

Site 4, participant 1

Conversely, the interview and WeChat#WeNurses audience indicated that WPTs are not always user-friendly, and do not always reflect patient acuity or connect demand to capacity effectively. The problem of a lack of up-to-date information was highlighted. The WeChat#WeNurses audience emphasised that nurse involvement in the creation of technology is therefore vital for effectiveness in supporting practice, which again supports CMO 6. NHS managers and the WeChat#WeNurses audience emphasised that the most important factor is the application of WPTs through professional judgement.

5. Seeing the bigger picture; escalation of concerns and balancing the risk

When immediate internal resources do not fulfil staffing requirements, NHS managers escalate their concerns to senior managers and engage with those who have a strategic and clinical overview of the organisation to see if there is latitude for movement of staff from diverse areas, or if they need to use external staff. Participants noted that WPTs can make staffing resources visible, in support of CMO 4. They suggested that WPTs that fit into their workflow and which are user-friendly and easy to navigate enhance communication to escalate staffing requirements and promote shared accountability. Seeing the bigger picture is critical during this process and is reliant on NHS managers’ understanding of the clinical environment, patient group needs and the organisational structures and procedures, which are incorporated into CMO 1 and are imperative for responsive decision-making for real-time and future need:

So, if a matron is running a group or orthopaedic wards then you sort of manage them together, so you make sure that every ward is safe. . . . failing that, you could then go wider and look at the whole trust and see, or campus, and see who we could move.

Site 2, participant 2

However, there was some indication that WPTs were fit for purpose at the senior management level, but not for others. This underpins the need for partnership in design within CMO 6, so that the WPTs fulfil different requirements at different levels within the organisation:

Personally, I think the staff on the ground could do without it [the WPTs]. I would say for more senior management position to get more of a strategic overview, I think it’s imperative that there is some kind of standardised tool. We could do with something to be able to demonstrate to the organisation the lay of the land really out here, actual teams, workloads, and ratios.

Site 3, participant 2

Often, requests for additional staff within this process required justification based on professional judgement:

If we wanted additional staff members above and beyond our establishment we would have to get a matron approval so the next level approval to say, for example, we have got four really confused patients all wondering, one has fallen over, we need someone to sit in that bay or have a supervised bay or one to one, ‘Can we have an extra staffing member?’

Site 3, participant 1

NHS managers explained that they sought evidence from WPTs for in-the-moment decision-making (expected patient flow and discharges) and for forecasting future demands (patient dependency/acuity classification data) and associated cost:

I mean the other thing that we do look at number of patients in the ED [emergency department]. So, what the patient flow is throughout the hospital, you know. So, that’s something that’s done at quite a high level. And then the other things that we’ve got is we use a system, a software system called Nerve Centre which allows us to predict discharges.

Site 2, participant 2

This process is elucidated in CMO 1, and shows that managers who understand the complexity and dynamicity of the system can use data from WPTs for responsive decision-making. WPTs are also used to specify for codes to ensure that agency staff have specific skills depending on patients’ requirements:

We use a lot of agency staff at the moment, we book people who have a 04 code which mean they have emergency experience even if it is not at our trust.

Site 2, participant 1

Furthermore, the data suggest that within this process, NHS managers consider both immediate and future demands and how they can respond effectively to attain the best outcomes (which supports CMO 1), and that they under-score when managers are responsive and show agility and flexibility to staffing demands. This can create positive reinforcement of distal aspects, such as the organisational reputation and recruitment. Furthermore, this relates to a culture of evaluation for organisational learning, as articulated in CMO 5:

We sort out the immediate situation, but then we have to look at long term, is this something that’s going to be a recurrent problem, if so, what we can do about it, how do we manage that situation? Who needs to be aware of that situation that’s senior to us, or who needs to be aware of that situation that’s involved with that patient, and how do we manage the more long term then?

Site 3, participant 4

The NHS managers highlighted the need to be transparent and honest about staffing levels with patients and families, while carefully addressing any concerns, to ensure confidence in their service. They indicated that they take care to be sensitive and sympathetic to the apprehensions of patients and families, especially when there are deficits in staffing levels; emphasising that the training, experience and skill mix of their establishment is sufficient to meet patient need. This staffing work in balancing candour and refining communication supports CMO 7:

I would say ‘we would normally plan to have seven on, but unfortunately today, because of unforeseen circumstances we’ve only got six. However, we’ve got a really experienced team on, we’ve looked at the patient allocation accordingly, we’ve got additional people round if we need them . . . if you’ve got any concerns please come and talk to me.’

Site 3, participant 6

NHS managers used data from WPTs to highlight to patients or the public that their resources are sufficient to meet demands:

It’s about assessing skill mix, assessing experience, assessing acuity. So, occasionally you could say ‘Well actually we have got less staff, but we are safe because of these . . .’ the things that I’ve just mentioned, you know?

Site 2, participant 2

Summary

The data from the think-aloud interviews in phase 3 were found to resonate with all CMOs, reinforcing the idea that they combine to explain the programme theory of what works to support NHS managers’ use of WPTs. This was corroborated by the WeChat#WeNurses discussion. This evidence further supports the integrity and accuracy of the explanations offered within the evidence synthesis.

1. To identify the different workforce planning and deployment tools and technologies used to deploy the nursing workforce, paying attention to the ways in which they are assumed, and are observed, to work

A range of WPTs were identified in the functional typology (see Chapter 3, The functional typology of nursing workforce planning and deployment tools and technologies). The evidence suggests that WPTs work by making data visible so that patient need can be matched to staffing resources. 124,125,127,129,130,133,134,144,147,151,175–180 The ability of WPTs to integrate relevant data for understanding complex issues in staffing (e.g. data on staffing skills and patient need) facilitates sense-making for effective decisions. 126,129,134,167,169,175,180,183,192,206 This can ensure that the right resources are assigned for efficiency quality and safety. 199 The findings indicated that WPTs can be used to forecast real-time and future demand for financial and resource planning. 125,151,168,171,172,174,182 The evidence also suggested that, when there is a standardised approach to the use of WPTs, the data produced can be used for evaluation and comparison against organisational and national standards required by legislation, mandates or guidance. 124,126–128,131,147,175,183,195,199,206 Thus, data from WPTs can be used to identify areas in which improvements need to be made for quality improvement and for the achievement of organisational goals. 133,145,147,174

The evidence suggested that there is a diverse range of WPTs. Patient dependency and acuity assessment tools and e-rostering were more prevalent in acute adult settings, with limited evidence for managers’ use of community caseload technology. 179 There was evidence that some organisations embraced technology and used it to its full capacity, for example e-roster, self-roster apps and electronic bed management systems. However, there was also evidence of reliance on written rosters or bed states, with some resistance to the adoption of technology noted in the literature. 132,143,178,182,197 Disparities in the use of WPTs were also evident within different clinical areas in the same organisations. It would appear that there is little consistency between health organisations in how they select or implement WPTs; this may become more standardised with the advent of recent guidance for benchmarking. 10 Legislation, mandates and strategy appear to strongly influence the implementation of WPTs. 132,147

WPTs have to be based on up-to-date data in order to be valuable for real-time decision-making. 155,166–170 When WPTs were embedded in organisations, inputting data in a timely fashion was the accepted practice;126,206 however, stakeholders suggested that, when staffing was reduced, updating data was not a priority. Many of the managers emphasised that they valued ‘the walk-around’ in hospital settings to obtain up-to-date information, falling back on traditional paper methods, rather than relying on data from technology. This highlights the dynamic and complicated nature of health care and how managers need to have confidence in the data underpinning staffing decisions. 127 Even when WPTs were acknowledged to identify needs and resources accurately, the evidence supports the importance of using professional judgement and human insight,178 which was described by a manager as ‘feeling the ward’. Managers need to critically interpret data from WPTs for safe staffing decisions131,158,163,176,178 and know their staff to determine whether or not staff can cope with particular situations and to identify the best staff to move when shortages occur, in order to ensure patient safety and good-quality care. These decisions use clinical and personal knowledge of nurses and organisational processes beyond the skills, attributes and numbers identified by WPTs. Furthermore, NHS managers act as a supportive presence to ensure the best outcomes for both patients and staff. They prioritise patient needs, but are mindful of balancing these with long-term requirements, with an understanding of how staff satisfaction relates to retention and recruitment:

We sort out the immediate situation, but we have to look at the long term.

Phase 3, site 3, participant 4

2. To explore the range of observed impacts of these technologies in different health-care settings and other public services, paying attention to contingent factors

The majority of evidence available related to adult acute settings. The effective implementation of WPTs can have a positive impact on organisational strategic aims for efficiency and good-quality care. A standardised approach facilitates monitoring for cost, efficiency and quality. 129,175 The increased understanding of staffing availability and the use of temporary staff articulated by WPTs can also result in cost-savings. 190,191,206 WPTs provide standardised data for comparison and benchmarking within the organisation and against other organisations. 124,126,127,129,131,147,168,171,175,180,182,183,206,210 Data from WPTs can therefore be used to set staffing parameters. 168 WPTs can collate data to provide evidence for the achievement of organisational aims in good-quality care in the light of national standards,168,192 to result in organisational learning. Furthermore, WPTs can produce visible staffing data for patients and public scrutiny. The impact of this lacks clarity; staffing levels may be meaningless for some patients, and for others, these may cause anxiety.

When WPTs offer credible real-time data, they are able to support managers in their decision-making role through the articulation of the availability of staff skills;128,133,144,151,152,157,158,160,164,174,175,180,181,184,189,191,198 identifying patient acuity and dependency;158,163,183,207,208 and matching patient scores to workload. 124–126,129,130,133,144,147,151,175–180 WPTs can also be used by managers to forecast future demand based on current patient data. 134,151,168,171,172,174,182,183 The project team found evidence that high levels of vacancies have an adverse impact on staff morale in nursing143,151,156,163,178,185,189,193 and in the police. 153 When WPTs enhance effective rostering, this can have a positive impact on recruitment and retention169,189–191 as poor rostering practice can result in staff dissatisfaction. It is a balance to ensure safe staffing and maintain staff well-being. Auto-rostering may offer parity in staffing allocation, and self-rostering offers staff more control. 143,190,191

The integration of data enables managers to have an understanding about the complexity of staffing needs and requirement. WPTs can present the bigger picture,145 offering a ‘helicopter view’. When WPTs are used effectively, they may empower managers to make changes. 129,132,169,179,192 The stakeholder work suggested that data can be used to argue for more resources, as managers were aware of their professional accountability for safety and communicated risk to secure safe staffing. Data from WPTs can enhance communication and provide managers with the evidence to negotiate for resources and justify staffing decisions. 131,132,144,169,179,192

The evidence emphasised the importance of the alignment of organisational strategies for safe staffing. 126,127,129,131,144,168–170,184,185,190,193 Perhaps unsurprisingly, there was strong evidence that when there was misalignment between financial and other organisational strategies related to budget restrictions, this could have an adverse impact on the effective use of WPTs;126,127,160 this was echoed in police staffing. 153 If WPTs identify the need for resource investment and this is not forthcoming, staff are unlikely to value their use. 126,144 Poor understanding of WPTs also impedes their use,125,126,132,151,155,208 and can be linked to a lack of alignment of strategies, which results in underinvestment in support or training for WPTs. 126,127 Furthermore, organisational systems need to be integrated so that real-time decisions are in synergy with long-term planning strategies, particularly around recruitment and retention, for the efficient deployment of WPTs, so that flexible staffing resources are available. One manager conveyed that the two most important things in her job were recruitment and rostering, as they were embedded with patient safety.

3. To investigate ways that can help NHS managers to identify, deploy and evaluate the nursing workforce resource to have the greatest impact on patient care

Both organisational support and effective leadership have key roles to play in NHS managers’ use of WPTs. 124–126,129,134,143,147,151,155,169,200,208 These are linked to the alignment of strategies and system integration. The findings suggested that organisational factors can work to support or impede NHS managers’ use of WPTs, so a collaborative culture and a standardised team approach are needed. 132,200 Professional judgement is inherently important to safe staffing decisions. It can be postulated that this may be developed in organisations that adhere to ‘Magnet values’, as these offer a supportive, collaborative organisational culture in which staff feel valued and professional development is facilitated. 118,119 Organisational investment in resources for WPTs has a positive effect in facilitating the provision of specialist support, education and training. 126,127,134,174,199

The findings indicated the importance of organisational collaboration and partnership to develop or select WPTs. 126,127,131,134,146,147,154,169–171,174,175,183,190–192,194,195,197,200,203,208,209 Collaboration can result in WPTs that fulfil organisational goals and that are user-friendly and fit for purpose. Data need to be presented in an understandable way, as there are different information needs at different managerial levels. 131,180,182,186,197 A robust evaluation culture supports organisational and individual learning around WPTs and safe staffing. 124,126,129,131,148,172,174,190,195,199 The interview participants were aware of WPT-related developments in other clinical areas and hoped to learn from them.

The evidence strongly suggested that NHS managers can be supported to use WPTs through training, expert support, peer support, mentoring and feedback. 126,127,174,179,195,199,200 An individualised approach to training appears to be effective when staff are not confident with IT. 179 Staffing decisions are complex, so not only do NHS managers need to have a clinical understanding of patient needs, they also need to have support to develop their knowledge of the clinical environment, patient group needs and the organisational structures and processes, as this is important for effective rostering and patient–nurse allocation. 129,131,144–152 NHS managers also need to draw on a range of leadership, communication and negotiating skills. The evidence suggested that managers need training and education on staffing decisions,150,174 but that this should be linked to the development of leadership skills163,167,186,189,199,202 and financial skills. 154,171 Furthermore, it appears that given the requirements for transparency of public data, NHS managers need guidance on how to convey staffing data to address patient and public concerns.

4. To generate actionable recommendations for management practice and organisational strategy

The cumulative findings provide orientation for approaches to supporting NHS managers’ use of WPTs through surfacing the knowledge and understanding of what helps to develop management practice and how safe staffing can be underpinned by organisational strategy.

For the effective operationalisation of WPTs within the complicated, dynamic, health-care environment, the findings suggest that the alignment of organisational strategies helps to support safe staffing and the implementation of WPTs. Pivotal to this alignment is a financial strategy that works as a foundation for real-time patient safety but also acknowledges the need for investment and long-term planning for staff well-being. Organisational investment to embed WPTs, through training and support, appears to be fundamental to an effective financial strategy, particularly as WPTs have a key role in the achievement of financial efficiency. 10,129,166 When the financial strategy is supportive of immediate and future safe staffing, this facilitates the alignment of other strategies, such as those on quality and recruitment.

The evidence indicated that the integration of organisational systems can reinforce a seamless approach to real-time and distal safe staffing decisions. The alignment of organisational strategies will contribute to a synergistic approach to system integration and the operationalisation of WPTs. When organisational systems are integrated, WPTs can be used to their maximum capacity, as staff will be equipped with the knowledge and skills to use them and be empowered to make effective decisions on safe staffing in the light of organisational processes and patient flow.

Collaboration and partnership are central to this process, so that staff from different strata of the organisation understand and facilitate the NHS managers’ role and contribute to organisational aims for good-quality care. Collaboration is also an important feature to ensure that WPTs are fit for purpose for NHS managers at different levels of the organisation. The alignment of organisational strategy and system integration can promote the collaboration needed for the co-design, implementation and evaluation of WPTs. This can ensure that they are appropriate for organisational processes and user-friendly. In this way, WPTs can help to achieve organisational outcomes.

Finally, the evidence suggested that organisations need to invest in education and support for NHS managers to develop the leadership skills and knowledge required to make and communicate safe staffing decisions. Although training, access to resources and expert support in the use of WPTs appears to be fundamental to this process, a more individualised approach, such as through mentoring and feedback from senior staff, appears to offer NHS managers the level of support needed for many of the challenging staffing situations they face.

5. To contribute to the wider understanding of the nature of the nursing workforce, nursing work and the quality of patient care

The programme theory, which is a cumulation of the seven CMO configurations, offers an evidence-based explanatory account of what works to support NHS managers in the use of WPTs. Although this programme theory is specific to the use of WPTs, the CMOs refer to concepts that can be transferred across to other service contexts to support workforce managers to deploy staff effectively and efficiently. It is important to acknowledge the evidence on patient safety, as the effective use of WPTs has a contribution to make to monitoring and evaluating staffing against safety objectives, all of which have emerged as a focus for prioritisation. 2,4 The evidence from the synthesis suggests the need for inter-related data, and this is echoed within the patient safety literature. 213 The Vincent et al. 214 framework for safety measurement and monitoring indicates five stages: (1) past harm, (2) reliability, (3) sensitivity to operations, (4) anticipation and preparedness and (5) integration and learning. Each of these have resonance with the findings; the synthesis indicates that WPTs are used to determine staffing levels and skill mix when harm has occurred. WPTs need to have real-time data for reliability, but can then facilitate system flexibility (or sensitivity to operations). Finally, the findings suggested that WPTs have been linked to organisational planning (for preparedness) and learning.

When the CMOs configure effectively, change occurs at the individual and/or organisational levels. Mid-range theories have been drawn on to offer an explicit understanding of the contingent processes underlying these changes. 109,215–241 These may point to perspectives that provide a new lens on the use of WPTs that could be pursued in future research. The theories that relate to the CMO configurations have been organised into five conceptual themes and summarised in Table 5, with further detail given in Appendix 13. Some of these mid-range theories are more dominant than others and have relevance across CMOs.

Strengths

A major strength of this synthesis was its methodological approach. A realist synthesis was eminently suited to the complexity of nurse workforce planning and deployment as it acknowledges contingencies within structures and processes that can impact on a range of outcomes. 92 A more traditional review would have a chief focus on outcomes, rather than exploring important contextual influences that may impede or trigger the mechanisms that influence outcomes. In adhering to realist principles, a theory-driven approach was followed to develop the programme theory, constituting seven CMO configurations. This explanatory account offers an insight into what works to support NHS managers to use WPTs and offers some consideration of the factors that may impede their use. The development of a functional typology of WPTs further enhances the understanding of the complexity of safe staffing. The use of the think-aloud technique in phase 3 articulated the decision-making processes undertaken by NHS managers; this refined and corroborated the emerging programme theory.

A realist synthesis embraces an eclectic approach to the range and quality of evidence for review, placing emphasis on evidence that is relevant to the emerging programme theory;94 unlike the traditional systematic review, quality is judged on this relevance. Therefore, the focus was on credible ‘nuggets’ of information that shed light on the programme theory, rather than criteria of validity and reliability within evidence hierarchies. By paying close attention to different perspectives from the published literature and stakeholder work, the review was iterative and accumulative; these processes were captured transparently within the description of the methods in Chapter 2.

A further strength of the realist approach was the inclusion of extensive stakeholder work embedded within the synthesis. A broad range of stakeholders participated, from NHS managers, experts in NHS staffing and PPI representatives with more extensive engagement through to people in a Twitter chat in phase 3 that was hosted by the chairperson from the advisory group. Their involvement and perspectives improved the trustworthiness of the findings by enhancing its credibility. They have also helped to mobilise knowledge from the synthesis as part of the review process. A more traditional systematic review may have impeded the understanding of nuanced factors, for example those relating to professional judgement.

Limitations

As this was a synthesis, the project team relied on scientific papers and reports, which often lacked detail on how managers applied their professional judgement in conjunction with the data from WPTs to make staffing decisions; consequently, the stakeholder engagement was amended to explore this further through a limited number of think-aloud interviews. The evidence about how managers develop skills in the communication of staffing levels to patients, the public and carers was very limited. Furthermore, the synthesis materials were often unclear about nurse managers’ roles within the organisational structure, their professional backgrounds or their capabilities. More of a range of relevant literature from the police and other public services was anticipated; however, extensive searching of relevant databases and using personal contacts did not surface significant evidence. There is also acknowledgement that in searching for ‘nuggets’ of explanatory evidence, the synthesis may have failed to track down all of the relevant literature; however, an extensive review of the literature within phase 1 sought to minimise omissions. It is also acknowledged that participants in the stakeholder engagement activities may have presented idealised versions of reality. However, the purpose of this engagement was to add depth to the reported evidence rather than to generate new findings per se.

The iterative nature of a realist review can present challenges in articulating and justifying the nuanced decisions made on which threads to pursue when developing the programme theory. It is acknowledged that the study, in focusing on NHS managers’ use of WPTs, examined only one part of the system of nurse workforce planning and deployment. However, Chapter 2 attempted to provide a transparent account of the approach. The decision-making process is the result of the uniqueness of the team, but can be viewed with confidence given the experience of the team members in realist methodology (CRB, JRM, LW, DF and BH). The study project team met monthly throughout the 18-month study period and offered a team-based approach to decisions; this was further guided by the advisory group; for example, the group suggested that balancing risks should be examined in safe staffing. The PPI group members articulated the importance of public scrutiny on staffing numbers for patients and carers.

Nurses

• Free-text keywords: nurs* or RN

• Medical subject heading (MeSH) headings: Nurses, Nursing, ‘Specialties, Nursing’, Nurses, Nursing Staff, Nursing Manpower.

Nurse managers

• Free-text keywords: ‘Nurse manager*’, ‘Nurse administrator*’, ‘ward sister*’, ‘ward manager*’, ‘charge nurse*’, ‘nurse supervisor*’, ‘nurse leader*’, ‘community manager*’, ‘community nurse manager*’, ‘clinical area manager*’, ‘nurse executive*’, ‘nurse director*’, ‘nurse in charge’, ‘head nurse’, matron*, ‘shift supervisor*’, ‘Patient care manager*’, ‘Middle manager*’

• MeSH headings: Nurse Administrators, ‘nursing, supervisory’, ‘nursing management’, ‘nurse managers’.

Staffing

• Free-text keywords: assignment, allocation, caseload, workload, ‘nurse:patient ratio*’, ‘nurse-patient ratio*, scheduling, reallocation, ‘time allocation*’, staffing, manpower, workforce, ‘skill mix’, ratio, roster, deployment

• MeSH headings: Personnel Staffing and Scheduling, Workload, staffing, shiftwork, shift workers, nurse-patient ratio, skill mix, Health Manpower, workforce, nursing manpower.

Evaluation of tools (iterative search 1)

• Free-text keywords: evaluation, effectiveness

• MeSH headings: program evaluation.

Complexity (iterative search 3)

• Free-text keywords: Complex*.

Nurse managers’ decision-making (iterative search 4)

• Free-text keywords: ‘decision aid*’, ‘decision making’, ‘decision support*’, ‘choice behaviour*’, ‘choice behavior*’, ‘decision behaviour*’, ‘decision behavior*’

• MeSH headings: ‘decision support techniques’, choice behaviour, ‘decision making’, ‘decision support systems, clinical’, Decision Support Systems, Management.

Allocation or assignment of nursing staff (iterative search 5)

• Free-text keywords: assignment*, allocation, caseload, workload, ‘nurse:patient ratio*, ‘nurse-patient ratio*’, scheduling, reallocation, ‘time allocation*’.

Staffing tools or information technology used in staffing (iterative search 6)

• Free-text keywords: tool*, technolog*, ‘decision making’, triangulation, ‘professional judgement’, acuity

• MeSH headings: Personnel Staffing and Scheduling Information Systems.

Capacity and capability of staff (iterative search 7)

• Free-text keywords: (‘personal capacity’, ‘work capacity’, capability, competence, experience, ability

Risk taking (iterative search 8)

• Free-text keywords: ‘risk taking’ OR ‘risk assessment’ OR ‘risk taker’ OR ‘risk propensity’ OR ‘risk management’)

• MeSH headings: ‘Risk-Taking’, ‘Risk Management’, ‘Risk Assessment’, Attitude to Risk.

Support for nurse managers (iterative search 9)

• Free-text keywords: leadership, ‘professional development’, ‘managerial competence’, self-efficacy, ‘self efficacy’, ‘leadership development’, ‘staff development’, ‘professional development’, confidence

• MeSH headings: ‘Staff Development, Leadership.

Code 1: data for evaluation – comparison of clinical areas’ actual staffing with predicted staffing