Bringing patient, staff and
organisation needs together

Annual leave is the single biggest reason for NHS staff unavailability. Research commissioned by Allocate Software, has found that if unavailability were improved by just 1% across all acute trusts, it would equate to £339 million worth of hours made available for frontline work.

Source: Allocate Software – The Impact of Working Day on Unavailability in Nursing Staffing (May 2017)

For the first time, a comprehensive review of annual leave across NHS trusts has been undertaken as part of this report. This shows that the average percentage of staff on annual leave throughout the year is 13% and that annual leave peaks at Christmas, Easter and in August – the school holidays – resulting in fewer staff nurses per patient during these periods and increased use of agency and bank staff.

The data also show spikes in agency spend correlate with spikes in annual leave. Yet annual leave is often considered something managers cannot control. Smoothing out staff annual leave over a year could reduce the agency bill to hospitals.


Source: Allocate Software

The graph below shows annual leave mapped against planned and unplanned admissions (finished consultant episodes) and shows the pattern of annual leave against school holidays and the consistent and constant pressure for NHS services.


Source: Allocate Software

While peaks of annual leave have improved somewhat over time, e-rostering can support managers to match leave more closely to demand. Where possible, annual leave should be evenly distributed throughout the year to ensure that there are appropriate levels of substantive staffing.

To support this it is important for trusts to establish an annual leave threshold range, with an upper and lower limit, which is sensible, meaningful and has been tested with the workforce. If managers are able to keep leave taken within the threshold they should not see peaks and troughs in agency and bank staff use.

If leave is below the lower limit it means staff are building up annual leave which they will need to take later in the year. This will create pressure on the service and risk to patient safety, particularly during the peak winter pressure periods of January to March. Therefore, managers need to be sensible about managing leave and manage expectations of the workforce. Also during periods of high annual leave being requested, such as summer, managers need to be conscious that temporary staff are also likely to be on leave, meaning it may not be possible to fill shifts, but patients will still attend hospital.

It is important to recognise it may not always be possible for managers to make these changes, as they have to take into consideration the personal lives of team members when managing and approving leave. Therefore, managers face the challenge of finding the right balance between effective leave management and flexibility. E-rostering improves transparency and equity of annual leave allocation, supporting managers.

The complexity that comes with annual leave management does not mean managers should avoid attempting to smooth out annual leave to improve staff-to-patient ratios. However, managers may want to consider management of the more controllable leave types more effectively, such as working day, sickness, parental leave and study leave.

Angela Thompson, Director of Nursing and Deputy Regional Chief Nurse for NHS Improvement London, explained: “It is about using the workforce data to review patterns in sickness and parental leave, and with support from HR, ward managers need to ensure they are adhering to the Trusts leave and sickness policies, while being cognisant of their staffs personal circumstances.”

Getting your head around headroom

Improved understanding of the concept of ‘headroom’, also known as time-out, unproductive, non-productive, non-effective and uplift, would support trusts to manage leave better.

Headroom, which defines anticipated and planned leave, should be between 21% and 24.5% of whole time equivalent, dependant on whether parenting leave is included in headroom, or is retained as centrally accessed resource. Typically, annual leave entitlement will consume at least 14.8%, but this will depend on the age profile of the workforce and length of service, and sickness around 4%. Therefore, headroom that is too low will not cover these two items, let alone study, carers and parenting, and other leave.

Looking at past sickness, study, administrative, and parenting leave percentages to know the actual headroom at which wards operate will help managers understand what level they should set headroom for their ward, unit or hospital.

The table below shows how one trust with headroom of 23% set its key performance indicators (KPIs) for leave management and they formed part of the trust’s approval process for rosters. The trust reviews shift data against the KPIs to help improve leave management and better manage headroom.

Leave management:

The total percentage of non-working days (leave) should equate to the percentage of headroom that is built into each ward/unit establishment.

For example, for a ward/unit with 23% headroom:

Annual leave 15%

Study/training days 2%

Sickness 3%

Parenting (maternity, paternity, carers) 3%

TOTAL – less than or equal to 23%

Source: NHS Improvement Good Practice Guidance: Rostering – June 2016

Tips on effective leave management:

  • Use a leave calendar to set a maximum and minimum for how many members of staff can be on leave at the same time
  • Establish clear definitions on different types of leave (unavailability) and reinforce them via the roster policy.

Done properly leave management can reduce the time ward managers spend managing annual leave, freeing them up to focus on their clinical responsibilities.

Trusts also need to ensure they not only agree and set a headroom allowance, which should be funded, but also agree a staff recruitment headroom allowance. Typically trusts recruit substantive staff for 15–19% of the headroom, which means each clinical area has an allowance in the budget to use temporary staff to backfill short-term sickness or unplanned parenting leave – up to the total headroom allowance. For example, if a ward recruits substantive staff to 17% of their headroom they have 4.5% in the budget to flexibly deploy bank staff as and when needed. Whereas, if the ward manager recruits substantive staff to the total 21.5% they would have no budget left for flexible deployment of bank staff, if required due to higher than anticipated leave requirements.


Improving bank filled shifts and staff availability at Derbyshire Community Health Services NHS Foundation Trust


Deryshire Community Health Services NHS Foundation Trust first implemented e-rostering in August 2013, with rollout across wards largely completed by April 2014.

Since then, the trust has made use of the workforce data generated to make improvements in staff availability, unused contracted hours, bank staff filled shifts, agency staff spending (£2.21 million spending on agency and bank nursing staff in 2015-16), staff work-life balance, and quality of patient care.

In response to these challenges, Derbyshire Community Health Services NHS Foundation Trust developed a responsive workforce model, which brought together e-rostering, the staff bank, and a new responsive workforce team of about 15 clinical staff (registered and non-registered nurses, therapists and allied health professions).

Improving bank staff use

Before e-rostering the bank staff, challenges Derbyshire Community Health Services NHS Foundation Trust faced included:

  • Approximately 21% of available bank shifts were unfilled
  • Bank bookings required an intermediary telephone service, which was only available during weekday office hours
  • On average, staff was spending 20 minutes checking if a specific bank request had been filled. With 9,800 requests per year across the trust, this equated to more than 3,250 hours (approximately two whole time equivalents).

The introduction of direct booking for shifts meant that bank staff had 24/7 access to request shifts through their computers and smartphones. The e-rostering system also enabled the recruitment team to have a real-time view of all vacancies.

It has resulted in:

  • Bank staff fill rates for temporary staffing requests are routinely above 85% for nursing
  • A 60% reduction in weekly calls to the bank to make bookings for temporary staff, from around 1000 to 400 calls per week
  • Reduction in human resource staff required – five whole time equivalents managed through natural wastage.

Reducing agency spending and unavailability

The trust was able to reduce agency spend, to £41,000 per month between April and November 2015 from an average of £129,000 per month in 2014-15. Though winter pressures and expansion in late 2015 saw spend rise again, before falling back as new units were brought into the same management process for agency spend.

The trust had faced issues with substantive staff unavailability. It decided to change its approach to rostering by gaining tighter operational grip of the workforce deployment, through increased visibility and control via the e-rostering system. It tested this approach from April 2015 to June 2016 across the whole trust.

This brought about a 3.3% reduction in unavailability over a single year, which produced £2.1 million in efficiency savings. Such outcomes have facilitated improved care trust-wide through greater amounts of time to spend on frontline activity.


Source: NHS Digital and Allocate Software

Improving staff usage

Other trusts have also seen improvements in reduced unavailability through e-rostering include East London and Plymouth.

In 2015-16, Derbyshire Community Health Services NHS Foundation Trust spent around £60.2m on substantive nursing staff. A 1% reduction in unavailability released around £600,000 of capacity into better patient care. The gains since April 2014 equate to around £3m of released capacity.

If a conservative 1% improvement in reducing unavailable time were applied to all English acute trusts, it would equate to £339 million worth of hours made available for frontline work. This does not consider the additional spend required to backfill the time through agency staff and other knock on effects through lost productivity.