We investigated the clinical utility of different strategies for antinuclear antibodies (ANA) and antibodies to extractable nuclear antigens (ENA) testing. All requests for ANA and ENA (n = 485) in a 20-week period were tested by immunofluorescence (FANA) and immunodiffusion (strategy 1), enzyme-linked immunosorbent assay (ELISA) techniques (strategy 2) or a combination of FANA and ELISA (strategy 3). Results of strategy 1 were positive by FANA in 8% (by immunodiffusion in 2%). By ELISA, 11% of the samples tested positive. In 12% (n = 60) of the cases the two strategies did not agree. The positive predictive value (PPV) for autoimmune disease of strategy 1 was significantly higher than that for strategy 2, but after exclusion of rheumatoid arthritis this difference was abolished. In strategy 2 reagent costs were high but working time comparably shorter. With strategy 3 PPV results were not better, whereas costs and working time were higher. The most frequently occurring reasons for ANA/ENA test requests were: joint symptoms (37%), follow up (30%) or abnormal laboratory result (7%). In a survey of the clinicians 66% replied that the test result did not have any consequences, irrespective of the result or the strategy used. We conclude that FANA and immunodiffusion are superior to ELISA techniques. However, the clinical value of ANA/ENA testing is low and more selective test ordering is strongly recommended.