Difficulties in clinical diagnosis of measles: proposal for modified clinical case definition.

Abstract

To examine the accuracy of clinical diagnosis of measles and to develop an improved measles clinical case definition. Case survey. Eastern Sydney, December 1990 to August 1993. All cases of measles notified to the Eastern Sydney Public Health Unit without or before serological confirmation. Demographic and clinical details, measles- and rubella-specific IgM and measles complement fixation titres in acute and convalescent (when available) sera and epidemiological links with confirmed measles cases. Of 49 subjects reported and with complete follow-up, 24 were confirmed with measles, four with rubella and 21 had no definite diagnosis. Clinical diagnosis of measles had a false positive rate of 51%. Subjects with confirmed measles were significantly more likely to have a cough (23/24) than those with no definite diagnosis (15/21; P = 0.03) and to be febrile on the day of rash onset (23/24 versus 10/21; P = 0.001). The Centers for Disease Control definition had a sensitivity of 92% but specificity of only 24%. A modified case definition of rash, cough and fever present at onset of rash had equal sensitivity but greater specificity (57%). As measles is no longer common in Australia, clinical diagnosis is unreliable. When a public health response is needed, cases should be confirmed by serological tests or, if not available, we propose use of our modified clinical case definition.