Feature / Shifting position

06 December 2016 Steve Brown

The key to more sustainable use of temporary staff is data. As a minimum, providers need to see easily what is driving their current usage of temporary staff and how that matches with the actual demand. But the main goal should be real-time visibility of how they are using their most important resource – their workforce, including substantive and temporary workers.

Centrally imposed caps on individual staff rates and overall budgets have had some success in reducing spend and restricting growth (Healthcare Finance, November 2016). But at a recent masterclass facilitated by the HFMA, workforce software firm Allocate Software looked at the problem from a different angle – highlighting the key opportunities for trusts to reduce their need for temporary staff in the first place.

Lord Carter’s report on NHS productivity was clear that much of the NHS was only paying lip service to e-rostering as a means of optimising the deployment of staff. ‘Even where trusts have invested in such technology, we found trusts were not getting meaningful use of it,’ it said. 

Finance staff and clinical managers at the masterclass agreed delivering improvement required good data, and e-rostering systems were key to this. But the reality in some trusts is that e-rostering has been more about supporting the payroll department and investment in e-rostering has stopped with the purchase of software.

Hugh Ashley, Allocate’s general manager UK and Ireland, said this was changing as organisations realised the potential benefit to staff and the savings for their employers.

Mark Oldham, director of finance and strategic planning at Mid Cheshire Hospitals NHS Foundation Trust, agreed that e-rostering was invaluable, but it was pointless to automate something that didn’t work. ‘You need to get the policy and culture right, translate that into the right processes and then automate,’ he said.

The single overarching benefit of proper e-rostering is the visibility this can give the organisation on how it is using staff, where it has over- and under-supply, and where the opportunities for improvement arise. This converts a system from just an operational tool making the production of staff rosters easier, to one that is strategic. Actual improvements of course only materialise if the data produced is reported and acted upon. 

CHPPD move

 
Lord Carter called for care hours per patient day (CHPPD) to become the ‘principal measure of nursing and care support deployment’. Managers agreed this was more about compliance currently, although trusts are starting to get to grips with it. It is generally reported to boards as required, alongside other staff metrics. But non-executives did not always understand what they were looking at.  

Daphne Thomas, deputy director of finance (operations) at Milton Keynes University Hospital NHS Trust, said: ‘The board could see variation between wards but it wasn’t always easy for those without a detailed knowledge of ward acuity to interpret the variation.’

Colin Ovington, chief nurse at Sandwell and West Birmingham Hospitals NHS Trust, said the measure was useful if triangulated with other data. ‘We look at early-warning triggers of potential problems on a ward and to see if the CHPPD metric is telling the same story.’ 

Su Rollason, director of finance and strategy at University Hospitals Coventry and Warwickshire NHS Trust, said the metric was really just a starting point. The trust was already quite strategic with its use of staff data, but the comparison of care delivered tended to be against rostered hours and needed to become more sophisticated. ‘The next stage is about contact hours – about that part of the equation,’ she said. 

To get the maximum value out of the metric,  said Mr Ashley, reporting needed to be four-dimensional. Trusts should report their actual CHPPD delivered alongside the planned hours (included in the roster) and the actual required hours (informed by the acuity and dependency data collected on a shift-by-shift basis). This then needs to be looked at alongside skill mix (at least broken down by registered nursing staff and care assistants).

Allocate head of customer success Leigh Malyon said such analysis was becoming ‘imperative, not just useful’. By adding in analysis of how actual care hours were delivered (see graph) – by substantive, bank or agency staff – the information could become really powerful in understanding what is driving staffing costs and to flag up quality concerns. 

Allocate offered a number of tips on how to minimise temporary staff costs:

  • Roster early - An organisation rostering four weeks ahead has 50% of the agency costs compared to one rostering two weeks ahead. This gives more time to fix issues and avoid unnecessary use of temporary staff. 
  • Get the headroom right - In calculating establishment levels for a ward, trusts take account of the amount of time staff will be unavailable for work, through holiday entitlement, sickness or study leave – typically an uplift of 22%. Getting the calculation right is one thing, but how headroom is managed is also important. Should trusts look to cover all this headroom with permanent staff? Trusts might set an establishment to meet predictable absences such as holidays and rely on temporary staff for unpredictable absences. Keeping establishment down below the level of real headroom was described as a ‘false economy’.
  • Get the right balance with flexibility - If established to cover headroom, a ward should be able to manage with average absence levels and average levels of holiday absence without temporary cover. Providers should then be clear what the policy is on taking leave – the number of nurses on leave at any time, say. This could be reinforced through a self-service system for booking leave. A rigid approach to leave needs to be balanced with flexibility, which can be important for staff retention. But flexibility arrangements should be periodically reviewed to check inequitable arrangements and unnecessary roster complexity.
  • Make it easier to book bank shifts - Trusts that offer a facility for staff to book their own shifts use 24% fewer agency staff and leave the bank office able to concentrate on the harder to book shifts.

 Required care hours v delivered hours graph









New approaches 

Some trusts have looked at different ways of organising nursing teams to help them respond to changes in demand. The masterclass discussed whether teams could be defined at specialty or directorate level – making it easier to move staff between wards. Mid Cheshire has already introduced fixed contract staff pools – specifically recruited to work in multiple locations as demand requires. 

Mr Oldham said current market conditions tended to work against these approaches. ‘The shortage of qualified nurses makes it difficult to use this resource flexibly,’ he said, with staff often attracted to more substantive ward roles or working through agencies. 

Sandwell and West Birmingham was looking at creating a pool for one-to-one care, a significant driver of agency spend. But Mr Ovington said broader changes were needed involving volunteers and family members to support such services (One to one, Healthcare Finance, November 2016). 

Ms Rollason said that where such arrangements were put in place, it was important to be clear the changes were being made to improve quality. This was acknowledged by the whole group. 

However, Allocate said that research showed that paying a premium to staff in these ‘float’ arrangements was also a success factor.

Some trusts already monitor staff demand in real time using patient acuity and dependency data recorded each shift to inform the movement of staff between wards. It was agreed that this was where all trusts should be headed.

The use of medical locums is arguably a bigger challenge currently than that of temporary nursing staff. Breaches of capped locum rates have not reduced and where there is an ‘overpayment’, it remains significant.

My Malyon said rostering medics was often complex, but there were big opportunities to improve the deployment of the existing medical workforce, which would have an impact on other staff. The starting point was the job plan. ‘You still see organisations trying to do this on paper, limiting the opportunity for analysis. The most opportunity for being more effective is at the planning stage and then delivering on that plan,’ he said. 

Job plans needed to meet expected service demand and then be completed as intended, he added, underlining Lord Carter’s call for greater analysis of job plans.

He gave examples of where job plans could be improved. For example, the company’s research suggested that on average programmed activities (PAs) were miscalculated by 0.62PA per consultant, typically due to errors around rota or out-of-hours working. On top of this, there is frequent miscalculation of unpredictable programmed activity for on-call duties. Mr Malyon said this should always be based on objective measurement of typical activity – yet this doesn’t always happen and is often overestimated.

Leave entitlement is widely misunderstood too, with mistakes both in calculating overall entitlement and how much leave is needed for weeks involving on-call duty. This can disadvantage or benefit consultants. But overall, the company estimates that leave is being over-allocated by 10%.

This may not always translate into a cash saving, said Mr Malyon, but might offer opportunities for enhancing services. 

He acknowledged it needed to be handled sensitively, recognising the goodwill provided by the medical workforce. But there was also a need to ensure equitable treatment of the whole medical workforce.

The key point was better visibility, from job plans to actual practice. And there were benefits for doctors too, with systems able to show doctors quickly when they are working, when they are on-call and who with, and helping them to manage any swaps process. 

These are difficult issues. But addressing the demand for temporary staff – both nursing and medical – is as important as tackling the direct costs of those temporary staff. 

 

Supporting documents
Features pg 38-41 December 2016