The literature on the fluorescent antinuclear antibody (FANA) test, commonly used in diagnosing systemic lupus erythematosus, was analyzed. The specificity of the test reported in early descriptive studies is much greater than the value obtained when the test is used in clinical practice. The probability of systemic lupus erythematosus in a specific patient was determined when different numbers of the classification criteria developed by the American Rheumatism Association are present. The predictive value of a positive or negative FANA test result was calculated using different pretest probabilities based on clinical criteria. The marginal benefit of the FANA test was determined as minimal at points of very large and very small pretest probability of systemic lupus erythematosus, and as maximal when five clinical criteria are present.