RESUMO
A statistical survey to determine the recording and reporting procedures practiced at the local level of the health and civil registry offices was undertaken in four different areas which were randomly selected. It revealed some of the causes of the delay in reporting of vital and health statistics and the degree of accuracy, adequacy, and reliability of such reports. A significant observation is the information that deaths were under-registered by 27.3% in these areas where the people could be considered better informed. Evidences were also presented on why medical certification of cause of death can not be fully relied upon. Ultimately, this survey exposed the urgent need to improve the present system of collection of vital and health statistics in order that the information gathered be useful and meaningful. Suggestions on how to meet part, if not all of this need were offered. Standardization of forms and procedures to render the data sbumitted comparable is called for. That the "analysis is only as good as the data collected" is an indisputable truth. (Summary and conclusions)