Year: 2007
Type: article
Abstract: We thank Dexter et al. for thought-provoking comments on our paper, but we found their language almost impenetrable. Unravelling their train of thought, we think they are saying: they are unhappy that we criticised their earlier efficiency formula [1]; there are numerous types of economic efficiency; their earlier formula concerned ‘allocative efficiency’ and our new formula concerns ‘technical efficiency’; our formula can be easily manipulated (for example, by using different definitions of cancellation); and they want more robust ‘proof’ (in a mathematical or statistical sense) for our new formula. We would like to respond to each of these points as follows. Firstly, Dexter et al. are unduly sensitive in interpreting our comments to be critical of their earlier approach as somehow ‘incorrect’. On the contrary, we acknowledged that their ‘extensive work in the field’ was far ‘more sophisticated’ than any similar work in the UK. The appendix of our paper argued that the two formulae are similar when costs are taken into account. So, there is little in their earlier work we would dispute. We simply argue that some of their analysis needs adapting for the NHS [2]. Next, they are correct to state there are understood to be numerous forms of economic efficiency [3]. Technical efficiency is perhaps the most intuitive and concerns the ability to produce output from a given quantity of input: crudely, the more output for the same input, the more ‘technically efficient’ the system [3–5]. There are several definitions of allocative efficiency. This concerns the concept of optimising ‘inputs’ to produce services: it is a market condition where resources are allocated to maximise net benefit to the community. It is also defined as maximising profits by matching marginal costs (that is, the cost per extra unit of ‘product’) to marginal revenue (the income per extra unit of product) [3, 4]. The notion of ‘microallocative efficiency’, on the other hand, concerns efficiencies at the level of a single hospital and overlaps with ‘technical efficiency’, so blurring the boundaries between these different ‘efficiencies’ (and indicating that one might plausibly influence the other) [6]. Other forms include ‘productive efficiency’ and ‘x-efficiency’[3–8]. Nonetheless, Dexter et al. are probably correct that, of all these forms, our formula primarily concerns ‘technical efficiency’. We are not persuaded, however, that their earlier formula [1] strictly meets all criteria for ‘allocative efficiency’ (for example, they did not use this term in their article). Bickering over whether a formula deals with one or other class of efficiency is not a useful end in itself. What matters is that any analysis helps solve problems. Our formula helps clinical staff to manage surgical lists in the NHS. There is much in Dexter et al.'s letter that leads us to suspect they do not understand how NHS hospitals function. The NHS is a monopoly and hospitals are not in competition (incidentally, we concede that monopolies are to a degree inherently allocatively inefficient [7]). NHS hospitals cannot exert any control over the workload presented to them. No income is gained by performing surgery; instead, hospitals receive arbitrary funding to meet costs. Good data are sparse, making it difficult to guide quantitative decision-making [8]. It is a matter of UK public policy that political decisions and targets determine day-to-day (that is, tactical as well as strategic) running of all hospitals [9]. So in the NHS context, many sentiments expressed by Dexter et al. (for example, ‘…the caseload of each specialty…determine[s] if more workload for each specialty can be achieved…’) are frankly meaningless. We agree that all formulae can be manipulated and ours is no exception. Hospitals could easily use make-believe data, or disregard cancellations and over-runs. All this misses the point, not least because any independent analyst could use our very same formula with the real data and so establish the true state of affairs. Finally, we agree it is highly desirable that formulae should be rigorously ‘proved’ (or more correctly withstand attempts at disproof [10]) using pure mathematics, statistics, or experimental applications to large data sets. It can take centuries to ‘prove’ even some formulae that always seemed highly likely to be correct [11]. We anticipate that more rigorous proof of our new formula will be offered, and will involve intricate statistics. In the meantime, we offer the following simple proof by reductio ad absurdum: our formula assigns as most efficient those surgical lists that utilise all their time, do not over-run and have no cancellations. Our formula does this in a unique manner. Therefore, any alternative formula (of whatever nature) would, by definition, give credit instead to lists that over- or under-run and/or suffer cancellations. Clearly these alternatives are absurd, which indicates that our formula is – in the most general sense – correct.
Source: Anaesthesia
Authors: J. J. Pandit, S. Westbury, M. Pandit
Institution: John Hunter Hospital
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: bronze
OpenAlex ID: W4256619390
