Skip navigation
Link to Back issues listing | Back Issue Listing with content Index | Subject Index

Planning for anti-TNF therapy


Bandolier 99 reported on the evidence about anti-tumour necrosis factor (TNF) antibody treatment for treatment of rheumatoid arthritis examined by the UK's National Institute for Clinical Excellence (NICE). NICE adopted the British Society of Rheumatology (BSR) guidelines for treatment of rheumatoid arthritis, that anti-TNF agents should be used if the following criteria were met:

Treatment costs are presently about £8,000 a year, and there is clearly a budgetary impact involved in introducing the agents. But how many patients need to be treated, and how much needs to be budgeted to implement the NICE recommendations and to follow the BSR guidelines? A good study from the West Midlands [1] provides some useful figures.


The West Midlands has 38 consultant rheumatologists in 14 centres serving 5.3 million people. In a two-week period in summer 2001, 12 centres and 32 consultants reviewed all patients attending outpatient clinics. A standard questionnaire was attached to notes for doctors or nurse specialists to complete. It confirmed that patients had rheumatoid arthritis according to ACR criteria, history of response to methotrexate and disease modifying drug (DMARD) use, and any contraindications to anti-TNF therapy. For those patients failing methotrexate and at least one DMARD, and with no contraindications, assessment of disease activity was made using painful and swollen joint counts, patient global assessment of disease activity and laboratory tests.


Over the two weeks, 1,441 patients with RA were assessed. Their average age was 58 years and 72% were women. Three of the patients were already using anti-TNF therapy.

Table 1 shows the numbers and percentages of patients who met different criteria for possible anti-TNF use. Thus 233 patients failed methotrexate therapy because of adverse effects (61%) or lack of efficacy (31%), and 177 had also failed at least one other DMARD. Of this 177, 19 had a contraindication for anti-TNF use, and of the remaining 158, 80 (5.6%) had disease activity above the threshold for anti-TNF use.

Table 1: Candidate rheumatoid arthritis patients for anti-TNF therapy, according to different criteria

RA by ACR criteria
Failed methotrexate therapy
Failed further DMARD therapy
No contraindication to anti-TNF therapy
Disease activity score above threshold for anti-TNF therapy
Additional criteria applied
Failed methotrexate +2 or more DMARDs
Failed methotrexate +3 or more DMARDs
Failed methotrexate + more than 3 other DMARDs

There were thus 80 candidates for anti-TNF therapy, but only three patients actually receiving anti-TNF therapy. The use of additional criteria of failure of methotrexate, and two, three or more than three failed DMARDs, would reduce the numbers of candidates for anti-TNF therapy, with the most stringent criteria giving only 27 candidates (1.9%).


This is useful. When new guidelines are introduced, we need a handle on their impact, and this shows how information can be speedily collected. Yes, there is a necessary period of organisation, but having a group of clinics collecting information comprehensively over a short period minimises the duration of a study while maximising the number of patients studied. It also much reduces the likelihood of patients attending twice and being double counted. Here the participation of 12 of 14 centres, and the inclusion of nurse-led clinics meant that the information was comprehensive.

We can be fairly sure that about 6% of patients with rheumatoid arthritis attending rheumatology outpatient clinics in the UK will be candidates for anti-TNF treatment using NICE guidelines. That may be too much of a budgetary jump at one go, but we know that even if we restrict access for those who have failed more than three DMARDs, 2% of patients are still candidates. We have a series of bottom lines, each of which can come with budgetary requirements, and can apply it to our own area.

While this study was not designed to produce incidence or prevalence figures, the authors go one step further. Using some literature data and assuming that 80% of patients with rheumatoid arthritis attend hospital outpatient clinics, they arrived at an estimate of 36 patients eligible for anti-TNF therapy per 100,000 population.


  1. CS Yee et al. The prevalence of patients with rheumatoid arthritis in the West Midlands fulfilling the BSR criteria for anti-tumour necrosis factor therapy: an out-patient study. Rheumatology 2003 42: 856-859.

previous or next story