This is an excerpt from the report “Can NHS hospitals do more with less?”, published by “The Nuffield Trust” in 2012 (12/01/2012), focused on the Chapter 7 of the report “Optimising the use of staff and improving staff productivity”. Though a 9 years old report, most of the principles ate still valid.
Nuffield Trust
Dr Jennifer Dixon CBE; Jeremy Hurst; Professor Judith Smith; Sally Williams
12/01/2012
credit for the image on top: dreamstime
Optimising the use of staff and improving staff productivity (Ch.7)
Key points
- Staff productivity is critical for the efficiency of hospitals, because staff costs account for 65–70 per cent of total costs. There is aggregate evidence from the UK to suggest that contractions in staff numbers have been associated with rising crude productivity in hospital services, whereas expansions have been associated with falling productivity.
- Whereas cutting staff will save costs and is likely to raise crude productivity, there is strong (observational) evidence both from England and the US to suggest that patient outcomes can be harmed by reducing nurse/patient ratios and diluting nurse skill-mix.
- There is similar evidence linking outcomes and doctor numbers. In the US it seems that a richer nurse skill-mix would improve efficiency in many hospitals. The present research in England revealed uncertainty over the right skill mix in hospitals.
- One longitudinal study in the US suggests that following a price squeeze, hospitals that improved their performance, having started with high-quality care and high costs, did so partly by reducing nurse numbers and diluting skill-mix. Hospitals that improved their performance, having started with low quality and low costs, did so partly by increasing nurse numbers.
- There may be opportunities to cut out waste in staff costs by benchmarking departmental staff productivity (while not neglecting quality or outcomes).
- There is evidence to suggest that often, staff productivity in hospitals can be raised by reductions in the use of agency staff and overtime, and that improving occupational health services for NHS staff may be cost-effective.
- Similarly, the present research identified that ‘turnaround’ was generally associated with reductions in posts, offloading underperforming staff, reductions in sickness absence and replacing agency staff with internal bank staff.
- Management practice varied between those trusts which made almost all staff reductions through natural wastage — partly for morale reasons and partly to save redundancy costs — and those that made significant numbers of redundancies with a view to reducing only unwanted jobs.
Index
- Introduction
- Optimising the use of staff
- Improving productivity
- Sickness absence
Introduction
Optimising the use of staff and controlling waste in staffing are critical issues both for hospital efficiency and for overall cost control, because staff costs represent 65–70 per cent of total hospital expenditure (NHS Employers, 2009)[1]. Recently, the National Audit Office (2010)[2] has identified a failure to control staff costs effectively in some hospitals. This chapter addresses the question of finding the right levels and mix of staff. It goes on to consider improving staff productivity by controlling waste in staffing. A final section considers the question of sickness absence.
Optimising the use of staff
Contractions in the NHS workforce have been associated with rising crude productivity (that is, lacking adjustment for any quality changes), and expansions have been associated with falling crude productivity in recent years. For example, total nurse numbers fell by about 10 per cent in the UK during the period of Hospital and Community Health Services (HCHS) crude efficiency gains in England in the 1990s, reported in Chapter 2 (Office of Health Economics, 2008[3]). There were also steep falls in the employment of domestic and ancillary workers, although many of those displaced may have been re-employed indirectly in hospitals via NHS contracts with private cleaning, catering and laundry companies. Conversely, nurse numbers rose by 12.5 per cent between 2001 and 2005 during the recent period of falling hospital productivity (this time, quality-adjusted) in the HCHS (Office of Health Economics, 2008[4]).
Of course, staff reductions can have unwanted consequences and can go too far. In the case of nurses, as mentioned previously, there is literature which suggests, on the basis of observational studies, that there are positive associations between various indicators of patient quality and aspects of the nurse working environment, a richer nursing skillmix and higher nurse staffing levels.
Meanwhile, lower nurse staffing levels are associated with higher nurse dissatisfaction with their jobs, higher burnout and higher turnover. This literature has been reviewed by the US Association for Health Care Research and Quality (Kane and others, 2007[5]). The Association’s review concludes that higher registered nurse staffing is associated with less hospital-related mortality, inpatient cardiac arrest, hospital-acquired pneumonia and other adverse events. Limited evidence suggests that a richer nurse skill-mix is also associated with lower mortality. More overtime hours are associated with an increase in mortality and other adverse events. However, the review was unable to conclude that these associations are necessarily causal. It is possible that the associations are indirect: both nurse staffing characteristics and patient outcomes might be attributable, for example, to variations in the management and quality culture across hospitals. An English study on variations in nurse staffing levels (Rafferty, 2006[6]) contains very similar findings to those gathered in the US.
In a rare attempt to explore the implications of such findings for efficiency, Needleman and others (2006)[7] estimated some of the costs and benefits of improving nurse staffing in the US (making the assumption that the associations are causal). Needleman and colleagues base their calculations on an earlier study of the outcomes of variations in nurse staffing in 799 US acute hospitals. They estimate that increasing the proportion of registered nurses to total nurses in hospitals below the 75th percentile hospital to the proportion in the 75th percentile hospital, would cost $811 million but would more than pay for itself in terms of avoided days of stay and avoided adverse events such as hospital-acquired infections. There would be a net reduction in costs of $1,821 million or $242 million, depending on whether fixed costs were recovered or not. In addition, there would be a reduction of about 5,000 deaths. Increasing nurse staffing hours per patient day (without changing skill-mix) to the 75th percentile level would not pay for itself, but would avoid 1,801 deaths. The cost per avoided death would compare favourably with the value of a statistical life used by federal agencies in the US. Assuming that the associations between nurse staffing and outcomes are causal, these findings would suggest that there is a clear business case for many US hospitals to improve nursing skill-mix, but not necessarily for increasing nursing levels. This contradicts what often appears to be the conventional wisdom — that diluting skill-mix would improve efficiency.
“Increasing staff/patient ratios has a positive effect on outcomes … but adds to costs
Valdmanis and others (2008)[8] estimated efficiency variations across 1,377 US hospitals in 2004, using frontier analysis and including quality measures in their study. They concluded, among other things, that high-quality hospitals tended to have too many labour inputs (slack in the number of staff) and that low-quality hospitals tended to have too few labour inputs. Clearly, there is a trade-off: increasing staff/patient ratios has a positive effect on outcomes, subject to diminishing returns, but adds to costs. However, the current literature does not permit the identification of optimal levels of staff inputs (Kane and others, 2007[9]; Newbold, 2008[10]).
A rare and interesting longitudinal US study (Jiang and others, 2006a[11]) examined changes in nurse staffing, among other things, in relation to changes in the quality and cost performance of 934 US hospitals between 1997 and 2001, following the Balanced Budget Act of 1997 (see Box 7.1). The findings of this study suggest that the adjustments that hospitals made to improve performance following the Act’s squeeze depended as much on their initial conditions as on the external pressures that they faced. Cost-containment strategies were helpful to those hospitals which started out with high costs. Revenue-enhancing strategies were helpful to those hospitals with high mortality. Optimising nurse numbers seemed to be critical for achieving the highest performance. The findings also suggest that performance can be changed by organisational and managerial initiative, independently of external incentives.
________________________________________________________________
Box 1: Factors associated with longitudinal improvements in efficiency in US hospitals
An American longitudinal study (Jiang and others, 2006a[12]) examined the determinants of both sustained high performance and improved performance among 934 US acute hospitals between 1997 and 2001 following the Balanced Budget Act of 1997, which restricted federal payments to hospitals under Medicare.
The authors were able to categorise the hospitals into four different performance groups in each of the two years: high-quality/low-cost (highest performance); high-quality/high-cost; low-quality/lowcost; and low-quality/high-cost (lowest performance) — where quality was measured by risk-adjusted mortality for six medical conditions and four surgical procedures, and cost was measured by adjusted cost per discharge.
About half of the hospitals changed category over the period, and about 11 per cent moved from other quadrants to the highest performing group. The hospitals that stayed in the highest performance group over the period tended to be characterised by for-profit status and system membership. They were characterised also by relatively low nurse staffing ratios and higher discharges per bed — presumably an indicator of shorter length of stay.
The hospitals which moved from the high-quality/high-cost group to the highest performance group were characterised by operating in markets with high levels of competition. They were characterised also by reductions in nurse staffing, dilution of nurse skill-mix and reductions in high technology procedures — presumably all indicators of cost-containment strategies.
The hospitals which moved from the low-quality/low-cost group to the highest performance group were not associated with any external characteristics, but were characterised internally by increases in patient volumes, higher discharges per bed and increases in nurse staffing. Here, perhaps some of the extra nurses were needed to raise volume and some were used to raise quality.
The hospitals which moved from the lowest performance group to the highest performance group were not associated with any external characteristics and did not change nurse staffing significantly, but there were signs that they increased day surgery and high technology procedures — both likely to be revenueenhancing.
These findings suggest that optimising nurse numbers is critical for achieving the highest performance. The findings also suggest that performance can be changed by organisational and managerial initiative, independently of external incentives.
The authors suggest that further work is required to identify the key organisational and management strategies which have led to success.
________________________________________________________________
Positive associations can be found between levels of doctors’ staffing and patient outcomes. For example, Pronovost and others (2002)[13] conducted a literature review of studies of doctors’ staffing variations across intensive care units. They concluded that high-intensity medical staffing was associated with lower intensive care unit and hospital mortality, and with reduced intensive care unit and hospital length of stay. In England, Jarman and others (1999)[14] reported that variations in standardised mortality across hospitals are significantly and negatively associated with hospital doctors per bed, and with GPs per head of population in the areas from which hospital patients are drawn, after controlling for other variables. However, such findings — like those with nurses — are not sufficient to identify the most efficient levels of doctors’ staffing, given that additional medical staffing comes at a high cost and is certainly subject to diminishing returns. The precise trade-offs remain unclear.
Given that most, if not all, hospital care depends on teamwork, this leads to wider questions about hospital staff skill-mix generally. Of course, there is literature suggesting that nurses can be substituted successfully for doctors in some circumstances, but most of this literature appears to relate to primary care. A review of nurses in advanced practice roles across all types of setting has suggested that whereas there is evidence that nurses can provide care that is technically equivalent to doctors in various settings, and sometimes greater patient satisfaction, little or no evidence is available on the cost-effectiveness of these arrangements (Buchan and Calman, 2005[15]). A review of studies of the use of nurse practitioners in hospital emergency departments suggested that nurse practitioners were neither better nor worse than house officers in treating minor injuries (Dealy, 2001[16]).
“Positive associations can be found between levels of doctors’ staffing and patient outcomes
Improving productivity
Staff productivity can be improved by reducing spare labour capacity and, often, the use of agency staff and overtime. An obvious way to identify spare capacity is to use a benchmarking approach.
A relevant American publication entitled Superior Productivity in Health Care Organisations: How to get it, how to keep it (Fogel, 2004[17]; see Box 7.2) focuses almost exclusively on identifying and reducing spare capacity among staff using benchmarking. It also includes a management strategy to bring about such change (the latter echoes some of the findings in Chapter 4).
________________________________________________________________
Box 2: Improving labour productivity in US hospitals
An American author, Paul Fogel, has suggested ways in which labour productivity in hospitals might be improved (Fogel, 2004[18]). His book starts from the observation that in many US hospitals, revenues are not keeping pace with expenses. This makes it essential to obtain tighter control over the largest single cost — labour.
Fogel suggests that there are some strategies that do not work, including over-complex measurement and reporting systems and lack of management authority and accountability at the right level for hiring and firing decisions. He suggests that diluting skill-mix is often ineffective, partly due to loss of labour flexibility. He identifies the use of agency staff and overtime as major sources of excess labour costs.
His approach to improving hospital productivity can be broken down into three main elements. First, it is necessary for each hospital’s senior management to develop a written productivity policy incorporating sound labour standards. The standards should be derived from historical benchmarking of labour productivity over the past three years or so, in each hospital department.
Second, he suggests negotiating agreement with departmental managers over productivity goals based on historical benchmarks, delegating authority to those managers for meeting these goals, and holding them to account with the help of intensive productivity monitoring.
Third, Fogel suggests a set of incentives and consequences for the managers. Management compensation should be strongly linked to cost saving, and persistently incompetent managers should be encouraged or required to depart. Also, managers should be encouraged to aim higher than the relatively cautious benchmarks outlined above.
There may well be a need for monitoring of quality of services to avoid productivity being obtained at the cost of quality, but Fogel is quite unspecific about how this should be done.
The book includes a chapter on the politics of productivity. There is a section on medical staff, which suggests that collaboration between doctors and managers is vital for improving departamental productivity. There is a section on unions, which suggests that improving productivity can offer a win–win opportunity because higher productivity can avert lay-offs. There is a section on executives, who may oppose the delegation of hiring and firing authority to managers. Finally, there is a section on managers, who may resist their new responsibilities.
________________________________________________________________
In England, staff productivity issues were identified by the NHS Institute for Innovation and Improvement (2006)[19] among its nine areas with potential for efficiency and productivity improvements in hospitals. Two of the areas highlighted in 2006 concerned improving staff productivity and managing staff and recruitment costs, respectively.
The first of these areas identified variations in the apparent crude productivity of consultant medical staff, presumably building on the work of Bloor and Maynard (2007)[20], which revealed wide variations in activity per medical consultant across hospitals in England.
________________________________________________________________
Box 3: Improving staff productivity
Inevitably, staff costs are the first area that trusts look to for efficiency savings. All six of the trusts the present study visited had reduced staff numbers significantly in recent years. As well as reducing costs, it was about ensuring that the organisation had staff with the right skill set to drive productivity and to offload ‘deadwood’ (that is, underperforming staff). David Loughton, Chief Executive of The Royal Wolverhampton Hospitals NHS Trust, says:
There is no question about relieving deadwood, especially when trying to take money out of the system. You cannot motivate staff if there are people in the way who don’t perform and you don’t do anything about it.
For most of the sites, staff reductions were achieved through natural wastage, freezing vacancies, early retirement and some voluntary redundancies, with very few mandatory redundancies.
Staff turnover at most NHS trusts in England was typically between 10 and 20 per cent in 2008 (NHS Information Centre, 2010[21]), and executives were in agreement that compulsory redundancies were rarely necessary.
In contrast, Sandwell and West Birmingham Hospitals NHS Trust adopted a deliberate policy not to impose a blanket vacancy freeze, out of concern that losing a random set of staff would have a detrimental impact on services. Instead they maintained a limited recruitment programme alongside a programme of redundancies. “You can address it through natural wastage, but the wrong jobs go. We found that overall performance improved during this period,” remarks Chief Executive, John Adler. Natural wastage still played a part and the number of redundancies was much lower than expected. Staffing levels have since drifted up and the Trust envisages losing 300 posts during 2010/11.
Skill-mix review has been a feature of all six trusts, although there was considerable uncertainty over whether to increase or decrease the ratio of qualified to unqualified staff, particularly around nursing. This echoes limitations in the literature around the optimal level and mix of nursing staff in hospitals.
The Royal Wolverhampton Hospitals NHS Trust had succeeded in removing £2 million of costs through a skill-mix review, which resulted in half of wards gaining staff and more than 100 nurse vacancies being filled. The end result was a ratio of 70 qualified nursing staff to 30 unqualified for most areas of the trust.
Another way in which the six trusts had reduced their head count was by limiting, or stopping altogether, the use of agency staff, and giving preference to in-house banks instead. An interesting observation was that internal bank staff, unlike agency staff, would be familiar with the hospital’s quality culture. In addition, many of the trusts had worked to reduce their sickness absence rates. NHS staff are absent for 10.7 days each year on average, and more than 45,000 NHS workers call in sick every day, according to NHS Employers (2009)[22].
Robert White, Finance Director at Sandwell and West Birmingham Hospitals NHS Trust, advocates better medium-term planning of the clinical workforce in particular:
I’d like to see departments think creatively about whether they will need to replace people and to start thinking about it now. That thinking needs to be ready to pull off the shelf, rather than leaving it until someone leaves.
The need to reduce staff costs and raise staff productivity may be precipitated by a crisis. The NHS found itself with a financial crisis in 2004/05 when more than onequarter of NHS organisations had fallen into significant financial deficit. In response, the Department of Health introduced a ‘turnaround’ programme, targeted on the organisations with the largest deficits. Articles in the Health Service Journal (2007[23]; Dent and Creamer, 2007[24]; Mooney and others, 2007[25]; Vaughan and others, 2007[26]; Vize, 2007[27]) report that one ‘turnaround’ trust had reduced posts by about 10 per cent over two years — mainly by natural wastage. A popular tactic was for senior management to take control of the filling of vacancies for a period, sometimes requiring sign-off by the chief executive. Another popular tactic was to reduce the use of agency and locum staff, often by strengthening internal bank arrangements or appointing more permanent staff selectively. NHS Employers recommends making the most of natural wastage.
Sickness absence
The subject of staff sickness absence in the NHS was visited by the Boorman Review in 2008/09 (Department of Health, 2009b[28]; 2009c[29]). The NHS has higher rates of staff sickness than other parts of the public sector. A review of the relevant literature (Hassan and others, 2009[30]) suggested that although good evaluations are lacking, there is evidence that in a number of industries several types of workplace interventions for a variety of conditions are effective in improving the health and wellbeing of staff. In addition, in the case of health care and social services, there is moderate evidence that ergonomic and other prevention programmes for musculoskeletal diseases are worth undertaking on economic grounds. Similarly, there is moderate to limited evidence that occupational disease prevention interventions can have positive financial implications. The Boorman Review made a number of recommendations, which were swiftly accepted by the then Secretary of State for Health, for more investment in occupational health services for NHS staff centred on prevention and early intervention.
Based on previous experience at Royal Mail and British Telecom, it was assumed that current rates of sickness absence in the NHS could be reduced by one-third, adding the equivalent of 14,900 whole-time equivalent staff and saving the NHS £555 million a year. Improvements in patient satisfaction and outcomes, and reductions in the use of agency staff and staff turnover, could be anticipated (Department of Health, 2009b[31]).
References
[1] NHS Employers (2009) Leading the NHS Workforce through to Recovery. Briefing 66. London: NHS Employers.
[2] National Audit Office (2010) Management of NHS Hospital Productivity. Report by the Comptroller and Auditor General, HC 491, Session 2010–11, 17 December. London: The Stationery Office.
[3] Office of Health Economics (2008) Sixty Years of NHS Expenditure and Workforce. London: Office of Health Economics. www.ohe.org/lib/liDownload/613/Sixty%20years%20of%20NHS%20expenditure %20&%20workforce.pdf?CFID=4986281&CFTOKEN=91811432. Accessed 30 March 2010.
[4] Office of Health Economics (2008) Sixty Years of NHS Expenditure and Workforce. London: Office of Health Economics. www.ohe.org/lib/liDownload/613/Sixty%20years%20of%20NHS%20expenditure% 20&%20workforce.pdf?CFID=4986281&CFTOKEN=91811432 . Accessed 30 March 2010.
[5] Kane R, Shamliyan T, Mueller C, Duval S and Wilt T (2007) Nurse Staffing and Quality of Patient Care. Rockville, MD: US Agency for Healthcare Research and Quality.
[6] Rafferty A, Clarke S, Coles J, Ball J, James P, McKee M and Aiken L (2006) ‘Outcomes of variation in hospital nurse staffing in English hospitals: crosssectional analysis of survey data and discharge records’, International Journal of Nursing Studies 44(2): 175–82.
[7] Needleman J, Buerhaus P, Stewart M, Zelevinsky K and Mattke S (2006) ‘Nurse staffing in hospitals: is there a business case for quality?’, Health Affairs 25(1), 204–11.
[8] Valdmanis V, Rosko M and Mutter R (2008) ‘Hospital quality, efficiency, and input slack differentials’, Health Services Research 43(5), 1830–48.
[9] Kane R, Shamliyan T, Mueller C, Duval S and Wilt T (2007) Nurse Staffing and Quality of Patient Care. Rockville, MD: US Agency for Healthcare Research and Quality.
[10] Newbold D (2008) ‘The production economics of nursing: a discussion paper’, International Journal of
Nursing Studies 45(1): 120–8.
[11] Jiang H, Bernard F and Begun J (2006a) ‘Factors associated with high-quality/low-cost hospital performance’, Journal of Health Care Finance 32(3), 39–52.
[12] Jiang H, Bernard F and Begun J (2006a) ‘Factors associated with high-quality/low-cost hospital performance’, Journal of Health Care Finance 32(3), 39–52.
[13] Pronovost P, Angus D, Dorman T, Robinson K, Dremsizov T and Young T (2002) ‘Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review’, Journal of the American Medical Association 288(17), 2151–62.
[14] Jarman B, Gault S, Alves B, Hider A, Dolan S, Cook A, Hurwitz B and Lezzoni L (1999) ‘Explaining differences in English hospital death rates using routinely collected data’, British Medical Journal 318(7197), 1515–20.
[15] Buchan J and Calman L (2005) Skill-Mix and Policy Change in the Health Workforce: Nurses in advanced roles. OECD Health Working Paper No 17. Paris: Organisation for Economic Co-operation and Development.
[16] Dealy C (2001) ‘Emergency nurse practitioners: should the role be developed?’, British Journal of Nursing 10(22), 1458–68.
[17] Fogel P (2004) Superior Productivity in Health Care Organisations: How to get it, how to keep it. Baltimore, MD: Health Professions Press.
[18] Fogel P (2004) Superior Productivity in Health Care Organisations: How to get it, how to keep it. Baltimore, MD: Health Professions Press.
[19] NHS Institute for Innovation and Improvement (2006) Delivering Quality and Value: Focus on: productivity and efficiency. London: Department of Health.
[20] Bloor K, Barton G and Maynard A (2000) The Future of Hospital Services: Management updates. London: The Stationery Office.
[21] NHS Information Centre (2010) ‘NHS staff turnover statistics 2007–2008’. www.ic.nhs.uk/statistics-and-data-collections/workforce/nhsturnover/nhs-staff-turnover-statistics-2007-2008. Accessed 4 April 2011.
[22] NHS Employers (2009) Leading the NHS Workforce through to Recovery. Briefing 66. London: NHS Employers.
[23] Health Service Journal (2007) ‘Have trusts turned the corner or just papered over the cracks?’, 17 May.
www.hsj.co.uk/news/have-trusts-turned-the-corneror-just-papered-over-the-cracks/56843.article.
Accessed 5 April 2011.
[24] Dent, E and Creamer, R (2007) ‘Green shoots of recovery’, Health Service Journal, 28 August. www.hsj.co.uk/resource-centre/green-shoots-ofrecovery/93583.article . Accessed 5 April 2011.
[25] Mooney H, Allmark R and Evans O (2007) ‘Turning curves’, Health Service Journal, 1 January. www.hsj.co.uk/turning-curves/58827.article. Accessed 5 April 2011.
[26] Vaughan V, Mooney H and Nolan A (2007) ‘Turnaround tales from the edge’, Health Service Journal, 1 January. www.hsj.co.uk/resource-centre/turnaround-tales-from-the-edge/59128.article. Accessed 5 April 2011.
[27] Vize R (2007) ‘Bring community medics in from the cold’, Health Service Journal, 6 September. www.hsj.co.uk/bring-community-medics-in-fromthe-cold/93651.article . Accessed 5 April 2011.
[28] Department of Health (2009b) NHS Health and Well-Being Review: Interim report. London: Department of Health.
[29] Department of Health (2009c) NHS Health and Well-Being: Final report. London: Department of Health.
[30] Hassan E, Austin C, Celia C, Disley E, Hunt P, Marjanovic S, Shehabi A, Villalba-Van-Dijk L and Van Stolk C (2009) Health and Well-Being at Work in the United Kingdom. London: The Work Foundation. www.nhshealthandwellbeing.org/pdfs/Interim%20Report%20Appendices/Literature%20Review.pdf. Accessed 1 April 2011.
[31] Department of Health (2009b) NHS Health and Well-Being Review: Interim report. London: Department of Health.
Originally published at:
https://www.nuffieldtrust.org.uk/research/can-nhs-hospitals-do-more-with-less
