Abstract
OBJECTIVE: To evaluate the validity of information in the rubella surveillance system in Quebec.DATA AND METHODS: Cases of rubella in the provincial registry of notifiable diseases, "Maladies à declaration obligatoire" (MADO), from 1994 to 1996 were matched with laboratory-identified cases and with cases in a reference file created from all case investigation records of regional departments of public health for the same period. Sensitivity and the proportion of cases in agreement were calculated.RESULTS: Compared with laboratories, the sensitivity of the provincial registry was 56%. Compared with the reference file, global sensitivity (confirmed cases plus clinical cases) was 58% and the positive predictive value was 50%. Of the 356 cases reported to regional public health departments, 65% were classified in the same diagnostic category (confirmed case, clinical case, excluded case) by public health professionals and a group of experts (weighted kappa=0.32). Information on rubella vaccination status was missing in 25% of cases in the MADO file for rubella.CONCLUSIONS: Notification of positive results for immunoglobulin M antibodies and viral cultures should be required of all laboratories. Uniform procedures should be adopted and applied for the validation of cases that are reported to regional departments of public health. In the context of the rarefaction of rubella, any immunoglobulin M-positive result should be interpreted using all available epidemiological information.