OBJECTIVE: To evaluate the quality of the Mortality Data System of the Ministry of Health and the improvement that could be obtained using a simple technique of data recovery. METHODS: Two Brazilian databases were compared regarding the deaths of residents of Pariquera-Açu, in the 1979/1995 period. Household interviews and examination of hospital/medical and forensic records were done for deaths that occurred in the city in the 2nd semester of 2001 which met the following criteria: death certificates with an underlying cause classified as an ill-defined condition, incomplete diagnosis or an external cause. Presumable cases of aids and all deaths in 10 to 49 year-old women were also investigated. After gathering information, a new death certificate was filled out and the new and the original underlying causes of death were compared. RESULTADS AND CONCLUSIONS: There was a small difference in the number of deaths in both official systems, from 1979 to 1995, and an increase in the proportion of deaths with an ill-defined condition. After the research, taking into account the 103 deaths that met the criteria, 58.3% of the ill-defined underlying causes and 59% of the incomplete diagnoses became a well defined cause; 73.3% of the external causes with an incomplete diagnosis became a well defined violence. One new case of aids and one unknown maternal death were discovered. It was possible to conclude that an improvement in mortality data can be obtained by means of a simple and feasible technique.
Mortality Information System; Health statistics; Evaluation of the quality of information
