Whatever the conclusion of the arguments amongst statisticians about how efficiency and productivity should be measured, in the real world there is increased interest and urgency as, like commissioning, increased efficiency is seen as the answer to the financial problems we face. However, efficiency is a by-product of how systems are designed and not an end in itself. The last thing that is needed is for it to be added to the long list of must do actions in already heavily-laden managerial agendas, without it being fully integrated with all the other tasks that have to be done. In the enthusiasm for productivity and efficiency improvement, its worth reflecting on some of the lessons we have already learnt about how to make a difference.
First, benchmarking against other organisations is not enough. Whilst benchmarks and metrics of the kind published by the Department and Dr Foster are helpful aids to diagnose where change may be required they are not the full story. The experience with the use of benchmarking data over the years is not very encouraging. The discovery that one is an outlier sometimes leads to major action and a radical step change in performance, but frequently doesn't. Benchmarking has to be accompanied by a detailed understanding of how superior performance is achieved. Local characteristics and methods of data collection can sometimes explain differences. Evidence suggests that even when the elements of the successful system are copied, higher performance does not always result. It may be difficult to identify the components that truly create high performance by visits and reading.
Second, don't run an efficiency programme. I don't mean that we shouldn't be rationalising back office functions, improving procurement and making support services work more efficiently. But this is not enough. The real money is in the delivery of clinical services, but getting clinicians on board is likely to be less successful if the focus is on productivity and efficiency, rather than what adds value to patients and reduces hassle for staff. They may be two sides of the same coin but language is important and the message that this shift in focus sends is significant.
So, what is the answer? We should focus on removing non-value adding steps. In healthcare and many other activities, the proportion of work that really adds value to the patient or customer is very low. This means that however efficient we make the operation, the processing of the pathology test or the consultation with the specialist we will have very little impact on over all efficiency.
Optimising parts of the system does not necessarily improve efficiency or productivity: rapid treatment in A&E, followed by long wait in the ward or high speed test processing, followed by a long delay for the results to be sent out adds little to efficiency. It may even allow for new errors or inefficiency to creep in.
2.5 per cent is hard, 10 per cent could be easier. This paradoxical statement seems to be supported by the evidence of 20 years of salami-slice cost improvements. A review by York University found that many of these improvements were not achieved. This is not surprising, constant trimming at the edges does not fundamentally change cost structures or the way people work. If we could find ways to allow trusts to shed fixed costs over longer periods, to dispose of assets and restructure their cost base, then some of the redesign and strategic change ideas that could yield very substantial improvements in costs and efficiency could be implemented. We know how to do this. During the 1980s and 1990s large psychiatric hospitals were shut using 5-6 year financial frameworks that meant that income and costs came out together and there was financial headroom to make changes. If we are going to have a top-slice then this is the sort of purpose it needs to be put to.
Economies of scale seem to be remarkably elusive. Many merger plans, centralisations and other changes seem to be based on the ideas that there will be economies of scale. Very little evaluation of these claims seem to take place but I have spoken to several chief executives who have taken over merged organisations who seem clear that the merger has increased costs and not produced many of the promised benefits. Centralisations may produce economies of scale but not without creating some diseconomies as well. Large complex systems can require additional co-ordination. Bringing together different cultures is expensive and risky especially as it seems that poor culture infects good culture.
Allocative efficiency is important too. Creating efficient and streamlined processes is important - it will reduce errors, stress on staff and improve patient experience, but it is pointless unless we are sure we are doing the right things. Shifting outpatients from hospitals to community settings might be more efficient, but if the patients did not need to be seen it is not much of an achievement. The development of programme budgets and much better cost effectiveness analysis is required.
Beware the search for the cheapest care. One of the most salutary lessons to be gained from a study tour to the USA is the risk of chasing the lowest cost setting. 'If we move all our patients who don't need to be in hospital into rehabilitation or skilled nursing facilities we will save money.' Perhaps, but probably only if you shut the ward, close the building and knock it down and sell the land. Otherwise, you will end up paying for patients in both places. The average cost will fall but total costs will rise.
Knowing what to do is one thing. Having the time and space to do it is entirely another. We need more space and time for managers and front line staff to be able to redesign their systems to eliminate waste and non value adding steps, to reduce the opportunities for errors, to reduce the number of steps and hand-offs and to make care more systematic. There is no magic bullet answer, just time and some careful application of analysis and improvement methods.
Nigel Edwards
Director of Policy
NHS Confederation
Health Service Journal - 15th June, 2006
