The anti-Sm test is thought to be specific for the diagnosis of systemic lupus erythematosus (SLE) and is one of the American College for Rheumatology (ACR) criteria for the classification of SLE. Locally, the sensitivity and specificity of this test for the diagnosis of SLE are unknown. We therefore aim to study the sensitivity and specificity of this test in our local context. A total of 1034 patient samples that were sent for routine testing of anti-Sm antibodies over a 30-month period, were included in the study. However, only 1031 patient samples were included in the final analysis as 3 casenotes were not traceable. Clinical diagnoses were obtained through a lupus database and chart review. Quantification of the anti-Sm antibodies was by the enzyme-linked immunosorbent assay (ELISA) technique. Positive anti-Sm test results were present in 165 samples, comprising 156 (94.5%) samples from lupus patients and 9 (5.5%) having diagnoses other than SLE. These diagnoses ranged from arthritis, overlap syndrome, Raynaud's phenomenon and cardiac-related diseases. The calculated sensitivity and specificity of the anti-Sm test, using a positive cut-off value of 20 units/ml, were 39.7% and 98.6%, respectively. If the positive cut-off value is raised to 30 units/ml, the specificity rises marginally to 99.5% but the sensitivity will drop by more than 10 percentage points to 27.2%. The anti-Sm test is not useful as a screening test for lupus but a positive result is highly specific for SLE.