Using the database obtained from the Campaigns for the elimination of congenital syphilis in the City of Rio de Janeiro in 1999 and 2000, we performed a reliability study on the National Disease Surveillance Data System (SINAN). We also evaluated the quality of the data provided. The percentage of agreement for the variable number of consultations in prenatal care was 91.0% in 1999 and 92.1% in 2000, and for syphilis during pregnancy was 62.7% in 1999 and 76.3% in 2000. Kappa statistics for the variables age of mother, date of treatment and therapy regimen provided an almost perfect agreement for age, in both years, but a slight to fair agreement for the other two. Looking at the quality of data, important variables as first VDRL during pregnancy, second VDRL and gestational age at syphilis treatment had more than 20% of blank or unknown answers. The reliability study of the SINAN using the campaigns' primary databases corroborates the need of establishing strategies to increase the quality of the SINAN at the various levels of the health systems. The 2004 change in case definition and the implementation of the Basic Course in Epidemiological Surveillance on syphilis in pregnancy and congenital syphilis will contribute to the improvement of the system.
Reliability; Epidemiological surveillance; Congenital syphilis
