Skip navigation

Costing Drug Treatments

Bandolier 15 contained a letter about costs of drug treatments. The implication was that it would be most cost effective to treat urinary infections with trimethoprim (about 76% effective with a cost of £0.38 per patient) as a first line with the more costly but more effective drug cephalexin (about 95% effective, £3.60 per patient) reserved for those who fail to respond. The thesis was that in a study of over 4,000 patients with UTI from Epsom & Ewell about £8,000 a year could be saved using such a scheme.

Costing GPs time

Professor John Mellerio of London and Dr David Jenkins, a GP from Cardiff, take a different view. They argue that a major cost that should be included is the time taken by the GP in seeing the 19% of patients who had not responded to trimethoprim. In money terms this would involve seeing about 800 patients again, for about 10 minutes each. This is about 130 hours of GPs' time a year, which, if costed at the BMA rate of £100 an hour amounts to about £13,000 and would have to be added to the costs of using trimethoprim as first line treatment.

Using this argument, they say, the use of cephalexin as a first treatment would be the most cost-effective. How, in addition, should the costs of the 4,000 or so days of continued unresponsive infection, pain or discomfort to patients unresponsive to trimethoprim be accounted for?

Taking a global view

This is an instructive dialogue. It is part of the process, as referred to earlier, of constructing models which describe the consequences of particular actions.

In this case it is difficult to be dogmatic. It could be argued that the GPs are there and are paid anyway, so that the additional time costs are illusory. It could also be argued that this is a true opportunity cost, and represents a lost opportunity to be doing something better and more effective with time already paid for.

Is this just an academic argument - is it real? Perhaps two ways of looking at the problem help make a decision.

Argument of extremes

The argument of extremes simply extrapolates each side of the argument to the point where they become unreal. Here, for instance, one extreme view would be that if GPs spend all their time fiddling with petty algorithms just to save a few pounds on drug costs, they'll never do anything really effective. The alternative argument might be that if fundholders manage their time and drug costs effectively, they will have cash to spend on other value-added services.

Argument of ranked priorities

Here one simply chooses the ranking of the priorities which one uses to make a judgement a priori , and then holds to the consequences. For instance, the ranking could be better for the patient, better for the doctor, better for the drug budget. That ranking would put cephalexin firmly in first place.

Further enlightenment from Bandolier's readers would be welcome.

previous or next story in this issue