An evaluative, descriptive, time series study was conducted to evaluate the completeness of variables from death certificate. The study population consisted of all deaths of children less than one year old living in Pernambuco State, Brazil, from 1997 to 2005, contained in the database of the Mortality Information System (SIM), provided by the Data Processing Department of the Unified Health System (DATASUS). We calculated each variable's incompleteness (blank/unknown data) by municipality of residence. Based on these data, we calculated the percentage of municipalities with excellent completeness (> 95%) and the tendency qui-square (p>95%). There was a difference in the filling behavior of the variables of the death certificate. While sex, age, race/color, place of occurrence, type of delivery and weeks of gestation had an increase in the percentage of municipalities with excellent filling, death confirmed by autopsy, completion of surgery, complement and medical examination showed a decrease. The study emphasizes the importance of educational actions and a better control in the filling of the variables in death certificates.
Evaluation; Information systems; Vital statistics
