Objectives. To determine whether DAS28 measurements by a specialized nurse, before the rheumatologist visit, in combination with the advice to rheumatologists to reach a DAS28 4 3.2, had beneficial effects on disease activity and medication prescription in patients with RA and to explore possible predictors for variation in medication changes and reasons for non-adherence to the advice to reach a DAS28 4 3.2.Methods. In this pilot study, rheumatologists were randomized to 'usual care' (n = 3) or DAS28 measurement by a nurse prior the rheumatologist visit (n = 4). In the usual care group, the DAS28 was measured but not provided to rheumatologists. Mixed model analyses were used for analysing between-group differences and for the prediction model. Rheumatologists in the intervention group were asked to provide reasons in cases of non-adherence to the advice.Results. After 18 months, DAS28 was reduced by À 0.69 and À 0.66 (P = 0.70) in, respectively, the intervention (144 patients) and the usual care (104 patients) groups. In the intervention group, medication was changed by rheumatologists in 35% of the visits with a DAS28 > 3.2; in the usual care group this was 33% (P = 0.99). Baseline DAS28 (OR 1.6; P40.0001) and HAQ (OR 1.3; P = 0.03) were positively related to a medication change. The most frequently mentioned reason not to change medication was patient refusal (26%).Conclusions. DAS28 measurement by a nurse was as effective as usual care; however, this intervention without protocolized treatment adjustments is not sufficient to lead to a considerable reduction in disease activity compared with trials with protocolized treatment adjustments.