46. There are no precise estimates of cost increases due to changes in health care technology (equipment, drugs, procedures and immunizations), but if income elasticity of demand for health care services is between 0.4 and 1.0, as available estimates appear to suggest, the supply side factors are responsible for 50-70 per cent of expenditure growth. From the view-point of expenditure control, its relevance is even greater if we take into account that the demand factors (basically income and age structure) are not controllable.

Within supply factors, two can be pointed out as critical: health technology and increased staff and equipment inventory. There is also abundant incidental evidence that technology is the most important factor: increased cost of bed/day, with no increase in the average number of hospital stays or patients per day; the evolution of costs of Health Maintenance Organizations running parallel to personal health care expenditure; and the consistency (around 5-6 per cent) in the growth of expenditure of national health care systems, whatever their organization, management and insurance model.

47. The relevant question is why expenditure in health care technology grows at such a brisk rate. Four reasons can be pointed out:

(i) Interest groups that benefit from the acquisition and use of new technologies: manufacturers of equipment, pharmaceutical companies and physicians.

(ii) The surplus - in some countries - of physicians and beds has led to competition not via a reduction in subsidized costs, but via an increase in subsidies with marginal effects, if any, over health status.

(iii) Lack of studies and knowledge on willingness to pay which causes any innovation with a positive marginal benefit, however small, to have great possibilities for implementation regardless of cost.

(iv) Reimbursing «ex post» all incurred costs (payment by service).

48. All these factors point to the tendency to apply any innovation, however minimal its health benefit, regardless its real cost: the health care system carries out a benefit analysis, not a cost-benefit analysis. Indeed, it should be suspected that the stringency of the benefit analysis is globally minimal if we take into account that there is only solid evidence about the effectiveness of 25 per cent of technologies and health procedures presently at work.

49. As a result, any debate or proposal on the costs of health care technology must face two types of problems which are different but interrelated:

a) Evaluation of technology from the point of view of efficacy, safety and opportunity cost.
b) The incentives that encourage the creation and adoption of cost-reducing innovations, as opposed to those which increase quality at whatever price.

(We let apart other problems which are foreign to the field of health care, related to the economic interests of innovators, equipment manufacturers and pharmaceutical companies.)

50. With respect to item a), the most reasonable suggestion is that there should be State Agencies for health care technology, made up by NHS professionals, scientists, and cost-benefit analysts to decide which technologies should be applied in the NHS, based on their effectiveness and therapeutical safety and on comparing implementation costs versus health improvements.

51. The idea of incentives is extremely important because, as it has been pointed out, NHSs tend to award incentives to innovations without consideration for costs.

On the one hand it is clear that if the NHSs begin by covering hospital treatment, there is a tendency to favour procedures which treat illness versus those that prevent it, because the sale of technology which treats the illness is guaranteed beforehand. If, in addition, the payment system for clinics and hospitals is by reimbursing incurred costs, there is a clear stimulus for health care R&D to develop technologies which increase quality, albeit at disproportional costs.

Two types of proposals can be suggested in this field:

(i) Government should incentivate in an active and distinct manner R&D expenditure for technologies which reduce costs, guaranteeing their implementation in the NHS. This is particularly important in preventive measures (e.g. immunization).

(ii) Substitution of the incurred cost system for one of prospective reimbursement, which involves to cease covering total marginal costs and to guarantee instead a fixed cost by type of admission or diagnosis, so as to generate incentives to work below standard costs.

52. A brief final reference to the cost of drugs which represent an important item of total health care expenditure (an average 12 per cent in OECD countries). Apart from the use of reference prices, and the exploitation of the NHS negotiation power as monopsonist some aspects of drug prescribing and marketing to consumers deserve to be taken into account.

Regarding the first aspect, it seems recommendable to enforce the prescription of generics and compel pharmacy outlets to dispense the most inexpensive remedy available. Regarding distribution channels, the general recommendation is that it should be competitive, in the sense that pharmacy outlets and other selling centres may offer lower prices and additional services.