An analysis of the short term impact of the Safe Nurse Staffing Levels (Wales) Bill on temporary staffing costs, specifically Agency and Bank nursing.

Introduction.. 2

Research into nurse staffing levels in Wales report 3

Costs going forward. 5

Health Board Plans 2015-16 to 2017-18. 5

How Health Boards are looking to reduce temporary staffing costs. 6

Vacancies, turnover and recruitment 9

How do bank and agency costs vary? 11

Conclusions. 12



1.        It is widely accepted that to ensure the optimum balance between nursing productivity and flexibility, there is a need to invest in nursing bank resource whilst minimising the use of agency staff, especially from high cost nursing agencies. However, there is very limited information in the public domain in terms of current and future bank and agency costs in Wales. 

2.        As previously stated in the Explanatory Memorandum accompanying the Safe Staffing Levels (Wales) Bill, all Health Boards are planning to use a triangulated approach to safe nurse staffing levels in adult acute hospital settings: the Chief Nursing Officers principles; the acuity tool; and professional judgement.  There is continuing evidence that whilst progress is being made, further investment is needed to ensure that these standards are met.  Further research also confirms that some Health Boards are experiencing increases in temporary staffing costs, and are implementing plans to reverse these increases and in particular focussing on reducing usage of high cost agencies.

3.        The original Explanatory Memorandum reported Royal College of Nursing (RCN) research showing that the overall cost of agency, bank and overtime staff for the three years 2010-11 to 2012-13, had been £132.5 million, or in the region of £44 million a year (this goes beyond the use of such staff in adult acute hospital wards, which is the focus of the Bill.

4.        Since the Explanatory Memorandum was produced the following information has been produced:

  The Welsh Government commissioned Cardiff University to undertake a study Research into nurse staffing levels in Wales, which was published in May 2015.  This study does provide an estimate of 2013-14 temporary staffing costs on surgical and medical wards, but does not look at historic trends or make future projections.

  All Health Boards have provided draft three year plans to the Welsh Government, setting out financial plans over the three year period from 2015-16 to 2017-18, although not all of these plans have been officially approved. 

    We understand that, as part of the Integrated Medium Term Planning process, Local Health Boards will have provided the Welsh Government with financial proformas showing staffing plans, including the use of temporary staff and locums going forward over the 3 year planning period.

    We understand that, the Chief Nursing Officer is leading nursing workforce projections for the Welsh Government.

Research into nurse staffing levels in Wales report

5.        The brief of the Research into nurse staffing levels in Wales project included the analysis of findings related to developing a better understanding of the availability and accessibility of nurse staffing data in medical and surgical hospital wards in Wales.

6.        The report did find that:

“there is a worrying variety in terms of attempts at comparability and consistency of systems, processes and software packages used to capture and hold staffing information at the organizational level.  The only way to access nurse staffing data at ward level is via ad-hoc requests made directly to individual Health Boards.”[1]

7.        The study, which started in October 2014 and was published at the end of May 2015, included data from 181 medical and surgical ward areas from six Health Boards.  This was time consuming for the project team and:

“no staffing data appear to be triangulated with patient safety outcomes or other related quality outcome metrics such as patient length of stay.”[2]

8.        The report highlights that the data request template used combined agency and bank staffing, so it was not possible to represent the differences between the two types of staff.  The report also found there to be a:

“marked variation in temporary staffing usage on wards that are similarly staffed and face similar demands such as unfilled vacancies, patient acuity and turnover.”[3] 

9.        The report recommends that further studies should be initiated to better understand this.

10.     Based on returns from around 90% of wards in Wales, the study estimated that annual temporary nurse staffing costs totalled £13.5 million for bank and £5.5 million for agency staff in 2013-14.

Costs going forward

11.     There is clearly evidence that temporary staffing costs have been increasing in recent years.  There is no simple explanation for these increases which are due to a complex interaction of factors.  Whilst historic efforts towards safe staffing levels have undoubtedly led to an increasing demand for nurses, the supply of nurses has also been impacted upon by flow of staff both into and out of the NHS.

Health Board Plans 2015-16 to 2017-18

12.     The updated three year financial plans that have been submitted to the Welsh Government this year are likely to provide the best idea in terms of a forward look.  These plans should include commentary on how Health Boards are planning to address pressures and growing costs in these areas.  We understand that local health boards were also required to provide financial proformas which should show projected temporary staffing costs for each financial year 2015-16 to 2017-18.

13.     On 22 May 2015 Kirsty Williams asked the Minister for Health and Social Services in written correspondence whether it would be possible for her officials to have sight of the relevant parts of these three year financial plans. Unfortunately, on 1 June, the Minister advised that the plans “would not be available in time for your purpose.”  However, in further correspondence, dated 20 June, the Minister noted that a limited number of plans (Cardiff and Vale UHB, Cwm Taf UHB and Powys) had been publically considered in draft, and as such these draft versions of plans were publicly accessible.  However, the detailed financial proformas that should accompany the plans for the relevant health boards do not appear to be in the public domain.  Even where plans have not been approved as yet, these proformas would constitute the best estimates of LHB temporary staffing intentions going forward.  With access to a summary of this information it would be possible to undertake more robust modelling of plans for future temporary staffing costs.  Unfortunately, as yet, the Minister’s office has not been able to provide the information requested.

14.     Cardiff and Vale UHB’s last two financial plans have set out temporary staffing costs.  The overall variable pay bill includes: agency, nursing bank, nursing overtime, non-nursing overtime, locum medical and dental, waiting time initiative – medical and on-call.  The total variable pay bill was almost £27 million or 5.46% of fixed pay bill in 2012-13, this fell to £23 million in 2013-14, or 4.68% of the total bill, in the first 6 months the proportion of total fixed pay bill was back up to 6.2%.

15.     Of these costs, it is possible to show agency and nursing bank costs separately and compare how these have changed.  We have combined information from the last year’s plan, a draft for this year from Cardiff and the Vale UHB and made a simplistic calculation of annual temporary staffing costs based on the first 6 months of 2014-15 (this takes no account of how demand may fluctuate over a typical year).

Table 1:  Temporary agency and bank staffing costs - Cardiff and Vale UHB

Table showing pay bill of agency and nursing bank (seperate and combined) for 2012-13, 2013-14, 2014-15 for Cardiff and Vale UHB.
 The combined costs rise from £14,036,423 in 2012-13, to £14,086,728 in 2014-15.

Source: Cardiff and Vale UHB plans and National Assembly for Wales Research Service calculations


16.     Cwm Taf UHB estimate that their usage of temporary staff is higher, accounting for around 8% of the pay bill.  As with the Cardiff and Vale UHB, the proportion is highest in the medical workforce compared to nursing and midwifery.  Cwm Taf did not provide a breakdown of agency costs for 2013-14 and 2014-15.

17.     Officials supporting Kirsty Williams AM in originally developing the Safe Nurse Staffing Levels (Wales) Bill have not been able to access plans of the other Health Boards with large numbers of acute beds, Betsi Cadwaladr, Abertawe Bro Morgannwg and Aneurin Bevan.  Nor has any of the information included in financial proformas relating to temporary staffing projections for health boards, with significant acute services, been provided at an individual or overall level for Wales.

How Health Boards are looking to reduce temporary staffing costs

18.     Cardiff and Vale UHB along with other Health Boards are implementing strategies to reduce the cost of temporary staffing and in particular agency staffing costs.  It can be seen that there has been some success in reducing agency costs, but nursing bank has risen considerably over the first 6 months of 2014-15.

19.     Of the draft Health Board plans officials have been able to locate for 2015-16, a key policy in terms of reducing the use of agency staff has been the investment in internal staff.  For example, Cardiff and Vale UHB are undertaking:

“A high level assessment and consideration is also being given to over-recruiting Band 5 nurses so that the UHB is ahead of any natural turnover occurring, any future need for extra capacity and almost eliminate the use of bank and agency.”[4]

20.     Cwm Taf UHB are:

“continuing to realign the nursing workforce to meet the agreed establishments to meet the safer nursing recommendations. This involves the rebalancing of the nursing workforce across our wards and hospitals and an associated reduction in bank and agency usage.”[5]

21.     The Cwm Taf UHB plan also states that:

“… In order to improve the consistency of patient care in the acute hospital wards a different, more innovative workforce strategy is being employed. Rather than relying totally on bank and /or agency staff to cover gaps in the rotas, a decision has been taken to appoint externally to a pool of 8 WTE qualified nurses that could be deployed to different wards on a longer term basis than the usual bank cover in order to cover more long term absences, e.g. long term sickness and maternity leave.”[6]

22.     The original Explanatory Memorandum stressed the importance and value of the planned implementation of e-rostering systems to improve the efficiency of workforce management.  The Cardiff University study also noted that e-rostering and sickness management was an important development.

23.     This is reflected in the Health Board plans that we have been able to view, Cwm Taf UHB has developed a workforce tracker which will be implemented as part of the policy to improve workforce management and reduce the need for temporary staff.

“This tracker enables the Health Board to rebalance the ward back to the establishment by turnover, retirements and also monitor the impact this should have on nurse bank spend. The Health Board can then make an informed decision regarding moving qualified staff if the workforce plan doesn’t show the movement needed. The tracker and that for Facilities are used to inform decisions by the Vacancy Control Panel.”[7]

24.     Coupled with review to rebalance nursing and health care support staff where there was previously either over or under staffing:

“It is expected there will be a reduction in bank/agency usage and this will be monitored through the workforce tracker and finance.”[8]

25.     Cwm Taf’s second phase of e-rostering was to be implemented in June 2015, meaning that adult acute hospital wards will be online by then. Cardiff and Vale’s implementation of e-rostering is well underway.

26.     The 2014 report by London School of Economics, NHS Safe Staffing: Not just a number highlighted the considerable benefits that e-rostering aligned with workforce management can have on temporary staff costs. 

27.     Following the implementation of an e-Rostering system, Basildon and Thurrock University Hospitals NHS Foundation Trust:

“is benefitting from recurrent annual savings of £100,000 through reduced input time, errors and corrections. The more efficient use of permanent staff, has brought about a reduction in the use of temporary staff that was saving £670,000 per month, which included a 37% (£5.8 million) reduction in total temporary nursing spend… On the basis of the savings claimed at Basildon and Thurrock, this would suggest that across England the potential gains to the NHS in a transition from basic e-rostering to a fully electronic rostering and timekeeping system could be up to £41 million annually. The robust data provided by such a system would also allow the use of resources to be carefully managed within a proactive system of activity analysis and workforce planning, rather than with a reactive system for managing shortages.”[9]

28.     Though they have few acute beds, Powys UHB state in their three year plan that:

“It is anticipated that there will be an increase in staff in post during the first two quarters of the year as vacancies are filled when compared with the 31st January 2015. As a result, there may be a reduced requirement for the use of temporary staffing, although for the purposes of this plan, we are assuming that current usage will continue. The budgeted projection of 1333.88 FTE projected for the duration of this plan will never be fully realised through staff in post as this includes headroom for the use of temporary staff. The effective use of temporary staffing will be continued and currently represents 7% of the total workforce.”[10]

29.     Powys UHB also include multi-professional education commissioning numbers as an annex to their plan.   This shows, for example, the academic intake of nurses and other staff and when they will complete training.  This links into the recommendations in the Cardiff University Research into nurse staffing levels in Wales report.

Vacancies, turnover and recruitment

30.     In correspondence, the Minister provided a snapshot of the number of nursing vacancies handled by the NHS Wales Shared Services Partnership.  This showed that, overall, there were 2,610 nursing vacancies and 744 healthcare support worker vacancies in Wales in January 2015.  Average staff turnover nationally was 8.3% amongst nurses and 8.6% amongst healthcare support workers.  A rough estimate therefore could be that there would be around 1,000 nursing vacancies in acute, elderly and general wards and in the region of 400-500 healthcare support worker vacancies.

31.     The average time from advertising a post to filling a post was 69 working days in February 2015.  It takes 37 days to advertise and interview for the post and 32 days from accepting the post to starting, including various security and qualification checks.

32.     With all Health Boards there is a need to ensure that there is a supply of qualified nurses to meet the need. 

There are factors that are needed to ensure that this happens.  While information on average time to fill nursing posts indicates that nursing posts are being filled relatively quickly, there is anecdotal evidence provided in three year plans that some Health Boards are finding it more difficult to fill all their nursing vacancies than had been the case in the past.  Although this is not within the remit of the Bill, there are a number of actions that are underway and need to be implemented to ensure that there is a sufficient supply of nurses to meet the demand in the future.  Some of these are set out below:

    The Welsh Government is investing in extra training and education places for health professionals, including nurses.  The number of training places for nurses in 2015-16 will increase by more than a fifth (22%).

    The Welsh Government is also providing further investment for professional development for existing staff.  There is potential to enhance the career prospects of experienced health support workers wishing to progress to being qualified nurses( wording and link to funding / report).

  By raising standards within adult acute and medical wards the Safe Nursing Staffing Levels (Wales) Bill will reduce the disincentive to leave the NHS in Wales and will increase the incentive for nurses who have left the NHS in Wales to return.  This will also make Wales a more attractive destination for nurses from outside the NHS.

  The investment and career progression opportunities mentioned above will also further enhance the incentive for current staff to remain in the Welsh NHS.

  The Cardiff University study highlighted a need for more understanding of why staff leave the NHS to join agencies.  By offering flexibility along with the certainty of employment, training and other benefits of NHS employment, Health Boards can reduce the incentives for such staff to leave the Welsh NHS.

  The Cardiff University study also highlighted how better workforce planning could make better use of existing staff.  The use of e-rostering can lead to far more efficient deployment of existing nursing staff and could improve down sickness rates. Also, it was felt that there was both overstaffing and understaffing on wards which was not explained by the data collected.  More efficient workforce planning could inform and utilise staff more effectively.

  Better workforce planning: the Cardiff University study also highlighted a lack of data and planning within Health Boards considering the age of staff, leaving intentions or tracking of future graduates coming through the system.

How do bank and agency costs vary?

33.     Three year plans available in the public domain highlight the importance of getting the best value from their use of temporary staff.  In general, this will involve using bank staff where possible and avoiding using the highest cost agencies. 

34.     Bank staff, as a rule, are generally paid their usual salary if they also have a substantive post.  If they work for the bank only, they are paid from various points in the NHS pay scales.

35.     Hourly rates for agency nurses for 19 companies who are part of the All-Wales 2015-16 contract were provided by the Welsh Government.  The most expensive company charges almost 50% (48.4%) more for nurses in the same band as the cheapest company would, for working on the same shift.  This variance holds for all midweek or unsocial hours shifts. 

36.     On occasion, Health Boards may need to use employment agencies which are not part of the all-Wales contract.  The Welsh Government provided evidence that the range of costs will be even wider for these non-contract agency staff.

37.     There are therefore savings to be made not only in terms of reducing the use of agencies, but also making sure that when agency staff are utilised, Health Boards use companies offering the best value for money for the given circumstances.


38.     Although the Minister did not provide Kirsty Williams’ Bill team with access to the latest Health Board plans, the Minister did state that:

“it is expected that HBs will use the triangulated approach to set staffing levels in these areas and therefore a narrative about the three elements of the methodology will be required to confirm compliance.” 

39.     The Safe Nurse Staffing Levels (Wales) Bill (as introduced) likewise requires that Health Boards take all reasonable steps to deliver a safe level of nurse staffing, determined through a triangulated approach.  As such, the initial costs associated with setting safe staffing levels in Wales, as required by the Safe Nurse Staffing Levels (Wales) Bill should be equivalent to those in Health Boards’ forward looking three year plans.

40.     Health Boards are all implementing plans to reduce the use of temporary staffing, especially agency costs, and the Welsh Government is providing additional investment in nurse training and professional development. As such, even a cautious estimate of costs can assume that while the rate of increase shown in bank nursing costs in the first half of 2014-15 may continue into 2015-16, it will plateau in 2016-17 and start to decrease in 2017-18. 

41.     This fits in with the statements of direction in both the Cardiff and Vale and Cwm Taf UHB plans.  It should be acknowledged that it is difficult to draw assumptions from a limited set of figures, as the Cardiff University report highlighted that there is considerable variation between wards and Health Boards.  However, these are the only figures which are already in the public domain. We have assumed that the move away from high cost agency staffing will continue to see modest 5% reductions each year, to reflect the strong focus on reducing these costs in Health Board plans and build on the reductions shown in the Cardiff and Vale UHB’s performance in 2013-14 and 2014-15.  We have assumed that bank staffing costs continues to rise from 2013-14 through to 2016 and start falling towards the end of the 2016-17 financial year, as ongoing investment to increase the supply of nurses will take time.

42.     The Safe Nursing Staffing (Wales) Bill would reinforce the triangulated approach using the Chief Nursing Officers Guidelines, incorporating the acuity tool and professional judgement.  Implementation is assumed to take place from 2016-17, with guidance being agreed to potentially be implemented for the financial year 2017-18. 

43.     This would mean that if implemented for the start of the 2016-17 financial year- until further headline data from local health board financial proformas is made available by the Welsh Government- the best rough estimate is that the £19 million bank and agency costs in 2013-14 would have risen to almost £25 million in 2016-17 and around £65 million for the three years post implementation.  

44.     Given the limited information available, we have deliberately erred towards overestimating the potential increases in bank nursing costs. The Health Board plans we had access to notably have policies in place to work towards eliminating agency costs by 2017-18.

Table 2 Extrapolation of adult acute agency costs in 2013-14 (using limited data from three year financial plans)Table showing extrapolated Agency, Bank and Combined costs for Wales from 2013-14, through to 2018-19.
 The Combined costs rise from £19,000,00 in 2013-14 to £24,460,00 in 2015-16.
 However, the Combined costs then fall to £18,990,000 by 2018-19.

Source: Cardiff University report, publicly available Health Board plans and Research Service calculations

Note: Figures are rounded to the nearest £10,000 so may not add to totals


Graph showing temporary staffing costs across Wales rising from 2013-14 to 2015-16, and then falling from 2015-16 to 2018-19. 
 The graph also indicates a possible range of extrapolated costs for each financial year. The size of the range increases over time.

45.     These estimates are set in a scenario based on current Welsh Government funding intentions for the NHS.  This assumes that Health Board plans are implemented and the Welsh Government continues to provide at least as much funding as already agreed for these three year plans.  These costs form part of the estimated costs of nurse staffing highlighted in the original Explanatory Memorandum that accompanied the Safe Nurse Staffing Levels (Wales) Bill. 




[1] Welsh Government,  Research into nurse staffing levels in Wales, 2015, page 5

[2] ibid, page 6.

[3] Ibid, page 9.

[4] Cardiff and Vale University Health Board, Progressing our Future, page 80, 2015.

[5] Cwm Taf University Health Board, Three Year Integrated Plan, 2015, page 255.

[6] Ibid, page 271.

[7] Cwm Taf University Health Board, Three Year Integrated Plan, 2015, page 272.

[8] Ibid, page 271.

[9] Hockley, T. and Boyle, S. (2014) NHS Safe Staffing: Not Just a Number, page 15.

[10] Powys University Health Board,