Rising workload and costs in diagnostic departments must be contained

EDITOR—Winkens and Dinant have highlighted some issues regarding the rising workload in pathology.1 In 1985 the workload in most diagnostic departments in the United Kingdom was reported to have been rising 10% a year whereas the number of inpatients and outpatients increased by less than 2% a year2; it is roughly similar now. A review of laboratory audits showed that the number of inappropriate tests requested by clinicians varies from 5% to 95%.3

The common perception among physicians is that these tests are cheap. Their unit cost may be low, but they have a high cumulative cost.3 The annual bill for operating laboratory tests is greater than the annual cost of operating computed tomographic scanners.3

Several methods to modify clinicians' use of diagnostic tests have been reported. The most potent interventions are methods that facilitate the preferred behaviour through blocking inappropriate requests or defaulting to the intended practice.4 In a study in the United States several characteristics were associated with a low level of laboratory use: being a leader, being part of a service group whose leader was a low user, clinical experience, being board certified, and being a graduate from “established” medical schools in the north east of America, Chicago, or California.5

The two most important reasons for the rising workload and costs in laboratories is the ease with which tests can be requested and lack of ownership by clinicians, as the problem is viewed largely as a laboratory problem. Good leadership and medical training are important. Thus consultants should play a key part as leaders, and a course on cost containment should be made compulsory in the medical curriculum.

The concepts of “profile” and “routine” should be abolished and investigations tailored to individual needs. It must be made mandatory for all junior doctors to get a certificate of competence in laboratory use from their consultants based on the information produced by the laboratory.

The question we have to grapple with is how we want to use our resources: whether to have more investigations or to fund more nurses, doctors, or such like to improve patient care. I suspect that the response would be similar to that of those people who say that they would prefer higher taxes to fund public services but vote otherwise in the polling booth. The decision we make will dictate the quality of NHS we have. Let's have more doctors and nurses.

Article gave unbalanced view of overuse of diagnostic tests

EDITOR—Winkens and Dinant report that diagnostic tests are overused by medical practitioners, and they propose various measures to curtail this problem.1 We believe, though, that their view is unbalanced. Investigations should only be done if they have potential therapeutic implications, but patients are entitled to be assessed adequately. Even experts' recommendations on appropriate diagnostic testing may be outdated by the time they are reported.

Patients with rheumatoid arthritis and systemic lupus erythematosus have many more cardiovascular events than other patients. 2 3 These events are a major determinant of the long term morbidity and mortality in these diseases. Abnormal cardiovascular risk profiles in these diseases show traditional risk factors such as dyslipidaemia 2 3 and non-traditional ones such as a raised acute phase response2 and raised serum homocysteine concentrations.3

These risk factors are associated with atherosclerosis in the two diseases. 2 3 Recent guidelines on the evaluation and management of rheumatoid arthritis and lupus, reported by the American College of Rheumatology, do not deal with the issue of cardiovascular disease. We believe that not evaluating cardiovascular risk—both clinically and by laboratory testing—may no longer be appropriate, and we start treatment accordingly in rheumatoid arthritis and lupus.

Similarly, gout is associated with atherosclerosis.4 Hyperuricaemia is a documented manifestation of the metabolic syndrome or insulin resistance syndrome, while the latter predicts a threefold increase in the incidence of cardiovascular events.4 Most patients with gout have dyslipidaemia among other treatable cardiovascular risk factors. 4 5 Snaith recently commented: “There is much work to be done before writing a prescription for allopurinol.”5

Because of the costs involved in treating cardiovascular disease, we think that the assessment and treatment of cardiovascular risk factors in rheumatoid arthritis, lupus, and gout is cost effective, at least in the long term.

Primary care organisations must take charge of laboratory expenditure

EDITOR—Winkens and Dinant offer a gloomy but realistic assessment of the many attempts to change doctors' behaviour related to laboratory testing.1 The experience documented is similar in many respects to that in New Zealand. In one area here, however, comprehensive sustained strategies achieved an appreciable reduction in laboratory expenditure, and this has been maintained over several years.

Pegasus Health in Christchurch is a primary care organisation similar in many respects to the primary care groups and trusts in England. It has a membership now of 230 general practitioners with a global budget of around NZ$80m (£25m; US$39m; €38m), and it established a comprehensive laboratory budget holding programme in 1994. This was through a contract with the then funding authority, which enabled it to keep nearly all savings.

An evaluation after one year showed that savings of 23% had been achieved and that the pronounced variation between groups with high and low costs per consultation had been greatly reduced. 2 3 The study conclusively showed that general practitioners, within the incentive of a defined budget and the ability to use savings for improving patient services, were able to make major savings with no evidence of any reduction in the quality of care.

A subsequent study in Pegasus showed that savings were being maintained but that variation was still inappropriately high.4 There was some evidence that better quality care was associated with lower expenditure. Since then per capita expenditure on laboratory services has been maintained at between NZ$20 (£6.20) and NZ$25 (£7.75), whereas the national cost weighted figure per capita has risen to NZ$37 (£11.46).

Primary care organisations have generally sought to engage in budget holding of laboratory services, but this has been inhibited by a confused and conflicting contracting process between funders and primary care organisations. There have been disagreements about setting budgets and what levels of savings could be retained by the primary care organisation. The experience is a prime example of the inability of bureaucrats to collaborate effectively and constructively with professional aspirations. It is the main reason behind the failure to extend the successful experience of Pegasus to a wider constituency.

Much more constructive action will now be needed as the new population funded district health boards begin to grapple with reducing their wide underfunding and overfunding on laboratory and related services. Having laboratory budgets held by primary care organisations along the Pegasus model seems to be the only answer.