Dengue haemorrhagic fever (DHF) is a severe viral illness with significant morbidity and mortality especially among children in Southeast Asia. The tourniquet test is recommended by the World Health Organisation (WHO) as an initial clinical screening procedure for patients suspected to have DHF, particularly in patients with DHF grade I. The aim of this study was to evaluate the validity of the tourniquet test as a predictor of DHF and also to assess the usefulness of repeated, serial tourniquet testing as a diagnostic indicator of DHF. Individuals included into this study were children from the Institute of Paediatrics, Kuala Lumpur who were admitted on a clinical suspicion of DHF based on the WHO criteria and who had serology for Dengue IgM performed. A standard method of tourniquet was performed on these patients on a daily basis following admission. A total number of 60 patients were considered for analysis and of these the tourniquet test was positive in 50 patients and negative in the remaining 10 patients. Sensitivity of the test was found to be 85.4% and the specificity was 25%. Further assessment of the predictive value of the test showed that the positive predictive value (PPV) was 82% while the negative predictive value (NPV) was 30%. In conclusion, a positive tourniquet test, serially performed on a daily basis was found clinically to be a useful preliminary screening tool for DHF as recommended by WHO. However its specificity was low and consequently led to a high false positive rate.