This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn.

Combined step-down prednisolone, methotrexate and sulphasalazine versus sulphasalazine alone in the treatment of early rheumatoid arthritis (RA).

Treatment.

Cost-effectiveness analysis and cost-utility analysis.

Patients with early, active RA.

Hospital and primary care; the clinical study (COBRA) was conducted in 10 clinical centres, one in Belgium and the remainder in The Netherlands. The economic study was conducted in Maastricht, the Netherlands.

Resource use and effectiveness analysis data were collected as part of the study protocol between 1993 and 1995. The price date was 1995.

Effectiveness data were derived from a single study.

Costing was undertaken on the same patient sample as that used in the effectiveness study.

Patients were randomly assigned to the two therapies but the method of randomization was not specified. The sample consisted of a total of 156 patients (76 in the combined therapy group and 78 in the sulfasalazine group), but no evidence was provided as to how the sample size was calculated or whether the study sample was appropriate for the clinical question.

A randomised controlled trial (RCT) was conducted and the study was multi-centred (10 sites). The duration of follow-up was 56 weeks. Two patients were lost to follow-up (one in each group). Assessors were blinded for effects of high-dose prednisolone for the first six weeks of the study.

The analysis was based on intention to treat. The clinical outcome was a pooled index, that is a composite measure to assess patient's improvements in erythrocyte sedimentation rate (ESR), tender joint count, grip strength, observer's global assessment and improvement in functional ability and this was scored from the MacMaster Toronto Arthritis patient preference interview. A secondary endpoint involved radiographic damage to joints in hands and feet. Physical function was evaluated via the validated Dutch version of the Health Assessment Questionnaire. The two groups were similar in important demographic and prognostic characteristics.

At 28 weeks (half way through the study) the mean improvement measured by the pooled index was 1.4 for the combined-treatment group and 0.8 for the comparator (p<0.0001) and at 56 weeks these were 1.1 and 0.9 respectively, (p=0.0001). Regarding the radiographic damage, the groups were comparable at baseline and by 56 weeks it was 2 points and 6 points for combined-treatment versus sulphasalazine, (p=0.004).

Clinical, radiographic and functional outcomes favoured combined therapy at 28 weeks (and these differences were statistically significant) and radiography remained statistically significant at 56 weeks.

In the economic analysis quality-adjusted life years (QALYs) were generated. Health states were valued by direct measurement using the Maastricht Utility Measurement Questionnaire which is a reliable and valid Dutch adaptation of the MacMaster Utility Measurement Questionnaire. Patient values were obtained at interview and the rating scale and standard gamble techniques were used.

Costs were not discounted as they were incurred in a period of 56 weeks. Costs were provided in some detail whereas very little detail was provided on the resources used. Costs measured included the costs of the intervention, non-protocol drugs, costs of out-patient care, those of in-patient care and direct non-medical costs. The quantity/cost perspective was stated to be societal. Quantities and costs were based on actual data. The price date was 1995.

Costs were treated in a stochastic way. The Student's t test was used to assess means per patient group. Secondary analysis on the differences in volumes were conducted using the non-parametric Mann Whitney U test.

Not included.

The costs were originally calculated in Dutch Guilders but were converted to US dollars ($).

Sensitivity analysis was conducted to test the robustness of the cost estimates and the parameters analysed included the price of hospital admissions and home help and monitoring schedules.

When assessed by the rating scale, the area under the curve calculation of the utility scores showed a significantly better gain of 0.06 QALY for the combined treatment over the sulphasalazine alone treatment (p=0.01; 95% CI: 0.02 - 0.10). In terms of the clinical indicators, combined therapy generally showed greater improvements than the sulphasalazine alone group. No extra toxicity as a result of extra medication was found in the combined treatment group.

The mean total costs per patient by week 56 was $5,519 for treatment with combined therapy and $6,511 for sulphasalazine alone, (p=0.37).

In this trial combined treatment was demonstrated to be more effective and equally or less costly than the sulphasalazine treatment. As the combined treatment was shown to be dominant, incremental analysis was not required. No cost per QALY was given.

Due to better efficacy and lower, or equal, direct cost, the combined treatment was argued to be more cost-effective than sulphasalazine alone.

In this paper there is no explanation of why sulphasalazine was used as the comparator to the combined treatment.

The estimate of benefit used in the economic analysis was likely to be internally valid. For the purposes of the cost-utility analysis, patient values were obtained at interview and the rating scale and standard gamble techniques were used. It is not clear whether the sample size and the duration of follow-up were appropriate for the study question.

Limited detail was provided on resource use, but this may well be because of space limitations. In general adequate details of methods of quantity and cost estimation were provided. The costing methodology was good. The perspective was said to be societal, but as the authors noted, only direct costs were included. Productivity changes were not explored - though these treatments may only have had a small impact on final outcomes.

This was a well conducted study. The sensitivity analysis enhances the generalisability of the results. Although the study was purportedly conducted from a societal perspective, patients' utility values were used to generate utility values for the QALYs and their values may not reflect those of society.

The COBRA trial was funded by grant 92-045 from Ontwikkelingsgeneeskunde, Ziekenfondsradd, The Netherlands of which A C Verhoeven was a research fellow. Pharmacia and Upjohn, Sweden, provided the sulphasalazine free of charge.