OBJECTIVE:
to estimate the incompleteness of Mortality Information System records on deaths from external causes in 2000-2002 and 2008-2010, by Health Region (HR) in Pernambuco, Brazil.
METHODS:
this was a descriptive study calculating the proportion (%) of selected variable incompleteness by HR.
RESULTS:
the 'medical care' (60.2%; 84.4%) variable showed higher incompleteness in Pernambuco and in the HR (eight in 2000-2002; 11 in 2008-2010); the least incomplete variables were 'sex' (0.06%) and 'location of occurrence' (2.0%; 2.6%); incompleteness reduced for 'place of birth' in 11 HR and 'medical care' in nine HR; 'race/color' incompleteness decreased in all HR and in Pernambuco state (-53.9%).
CONCLUSION:
the incompleteness of some variables was reduced, however, the persistence of high incompleteness in variables like 'schooling', 'medical care' and 'necropsy' can distort information about deaths from external causes, precluding action and policy planning intended to reduce them.
External Causes; Information Systems; Death Certificates; Epidemiology, Descriptive
