Análise da mortalidade por causas mal definidas e por diagnósticos incompletos na região do Médio Paraíba, Estado do Rio de Janeiro, de 2005 a 2009

Abstract

A high frequency of deaths due to ill-defined causes and incomplete diagnoses compromise the validity of cause specific mortality indicators, constituting an obstacle to the rational allocation of health resources based on epidemiological profile. This study evaluates the quality of information regarding the underlying cause of death in the Médio Paraíba region, state of Rio de Janeiro, Brazil, in the years 2005 to 2009 for the entire population. Data were obtained from the Mortality Information System (SIM) provided by DATASUS / MS. The analysis was based on two indicators of proportional mortality, the proportion of deaths due to ill-defined causes (IDC - Chapter XVIII, ICD-10) and incomplete diagnosis (ID) according to the classification presented in the Burden of Disease Project in Brazil, 2002. The association between quality of information of the underlying cause of death and demographic, socioeconomic and related to the occurrence of deaths covariates was assessed by calculating odds ratios of deaths due to IDC and ID in relation to the remaining causes of death. Proportional mortality due to IDC in the Médio Paraíba was found to be 4.54% in the period 2005-2009, values similar to those expected in developed countries. However, following a national trend, the proportion of incomplete diagnosis in the Médio Paraíba region, in the same period was high (20.59%). Together, the proportions of deaths from IDC and ID in the Médio Paraíba region in the quinquenium from 2005 to 2009, reached 25.13%, above the median value of 12% estimated for the world population. The odds of deaths due to IDC and ID decreased at higher levels of education. As for the variable race, the deaths of the black subjects had higher odds of having IDC. Deaths of white individuals were more likely to be listed as the underlying cause ID. Among deaths without medical assistance chances of IDC and ID were higher in relation to deaths with assistance. The deaths in hospital had lower odds of IDC and greater chances of ID. The variables had ignored or not reported were associated with higher odds of IDC and ID. The results suggest that in the Médio Paraíba quality of mortality data regarding IDC is well above the national level, resembling the values of developed countries. Still, the proportion of residual causes is quite high, showing that despite the significant improvement of SIM persist limitations that restrict the wider use of the system and prevent that advances policies and programs in health are greater.