Limits to demand for health care

Authors’ reply

Two of these letters are fascinating in demonstrating how unwelcome the suggestion can be that demand for health care is, in important respects, finite. The rationing movement has been very influential during the past two decades of political hostility to public provision. Now that substantial investment in the NHS is occurring, the claim is becoming more common that "rationing" is simply a neutral gloss for making sensible choices. This is completely unconvincing. The association between rationing and denial of legitimate demand was deliberately evoked by those who coined this term: when did we hear lottery winners asked how they would ration their winnings? The earlier assertion that those promoting rationing "adopted this term because it provokes the greatest public controversy"⁽¹⁾ still rings true.

This tendency to evacuate the distinctive meaning of the term rationing is apparent in Maynard and Sheldon’s treatment of the provision of elective surgery. If people are only exposed to the risks of surgery when the benefits outweigh the likely harm, this is protection rather than rationing. To equate rationing criteria with treatment criteria is to ignore the clinical rationale of defining who benefits from health care, which is not an act of rationing, but of reviewing evidence of efficacy. Medical care is, fortunately, not applied health economics, and hopefully never will become so. Unfortunately the assumption that supply cannot meet demand is deeply embedded. An interesting instance of the difficulty that otherwise informed people have with questioning this view came from the editorial committee of the BMJ. An epidemiological paper that implied that rationing of primary total hip replacement was unnecessary, which was later accepted by the Lancet, ⁽²⁾ was rejected by the BMJ on policy rather than scientific grounds: "We remain unconvinced by the argument about the lack of need for rationing [of total hip replacement]… .You say … that an increased provision of 50% over a 5 year period would clear the backlog. But where is this increase to come from and so how is rationing to be avoided?" (rejection letter, 10 June 1998). The answers to these questions, which were not sought as the correspondence was firmly closed, is, first, only one additional operation every three weeks by each consultant orthopaedic surgeon, as the capacity of the NHS is prodigious, and, second, from funds that have since been allocated. At the time the BMJ ’s editorial committee could not conceive of the eventuality of a soluble problem and so was unwilling to publish a paper that might have pointed towards that solution. Rao’s letter shows the common problem in this area of limiting observations to generalities, and ignoring the sorts of empirical evidence, some of it expressed in our article, that shows that assumptions about the cost pressures of ageing and innovation are highly questionable. The issue of rationing for equity is dealt with elsewhere.⁽³⁾

We are relieved of the task of correcting Maynard and Sheldon’s misrepresentation of the origins of the dominant assumption that demand must exceed supply by two distinguished medical historians, one of whom wrote the official history of the NHS. The fallacy that public provision must fail to satisfy demand was always, and continues to be, more political than empirical, as Loudon and Webster point out.

The main purpose of our paper was to point out how the rationing debate is currently unencumbered by the conventions of empirical enquiry; rather it relies on reiterating in increasingly strident terms a set of unquestioned axioms. Questionable assumptions have provided a pretext for the multiple injustices that are the consequence of continued underfunding and denial of cost effective treatments in the NHS. One has to have passed through an intellectual hall of mirrors to be able to assert that "rationing will be good for our health."⁽⁴⁾ Attempting to provide better coverage of unequivocally beneficial remedies would presumably be even better for our health, if, as seems likely, the barriers to doing so are based more on prejudice than evidence.

Stephen Frankel
professor of epidemiology and public health
stephen.frankel{at}bris.ac.uk

Shah Ebrahim
professor of epidemiology of ageing

George Davey Smith
professor of clinical epidemiology

1. New B. The rationing agenda in the NHS. BMJ 1996;312:1593-601.
2. Frankel S, Pearson N, Greenwood R, et al. Population requirements for primary hip-replacement surgery: a cross-sectional study. Lancet 1999;353:1304-9.
3. Davey Smith G, Frankel S, Ebrahim S. Rationing for health equity: is it necessary? Health Economics 2000;9:575-9.
4. Neuberger J. Why rationing will be good for our health. Times 2000 Feb 28.