Résumé
A hospital-based trial to compare the clinical diagnosis of malaria; microscopy, and a rapid diagnostic antigen capture detection dipstick (ParaSight-F) was conducted in North-west Thailand. 301 people who presented themselves at the hospital were selected. 204 (68%) were presumptively diagnosed as having malaria by the triage nurses; 64 (21.3%) were P. falciparum parasite positive, and 94 (32%) tested positive for P. falciparum with the ParaSight-F test strips. There was no association between hemoglobin levels (<10g/dl and > or = 10g/dl) and malaria, and although there was a good statistical association between temperature and malaria the specificity, sensitivity and positive predictive values were all low, indicating that temperature alone is a poor indicator of the disease. Based on the microscopy results, we found that a presumptive clinical diagnosis dramatically over-diagnosed malaria, and similarly there were a large number of false positives using the ParaSight-F test. We believe that many of the patients had received some form of malaria treatment prior to presentation at the hospital, and that the high number of false positives are explained by persistent antigenemia and the possibility of there being sequestered parasites following incomplete chemotherapy.