Avoidable causes of death in a low-income area in Rio de Janeiro Sate, Brazil: A ten-year longitudinal record linkage study

Background: Record linkage methodology represents a useful tool for monitoring health outcomes in small areas. Purpose: To evaluate the proportion of avoidable causes of death among total deaths identified trough a ten-year longitudinal record linkage study carried out in a low-income area where a major sanitation programme was partially implemented. Methods: A census of 29,807 households was conducted from June to December 1997 in Duque de Caxias County (Rio de Janeiro, Brazil). The census database (29,807 households members) was linked to the Rio de Janeiro State's mortality database (1997 to 2006; with 801,587 records), using an automatic probabilistic approach combined with an extensive clerical review. Avoidable causes of death were classified according to the list proposed by a Brazilian Ministry of Health work group. We restricted the analysis to the following age ranges: adolescents and young adults (10 to 24 years), adults (25 to 59 years), elderly (60 years or more) . Results: A total of 681 deaths were identified with 91 (13.4%) presenting an ill-defined cause. Avoidable causes accounted for 88.1% , 76.1% and 72.1% of deaths with a defined cause among adolescents and young adults, adults and the elderly, respectively. External causes were predominant in the younger age strata (70.2% homicides and 14.9% traffic accidents). The share of avoidable deaths due to non-transmissible causes rises with age, reaching 58.1% in the elderly group. Conclusion: The proportion of deaths due to avoidable causes in the area of the study is high, probably due to lacking health conditions and care.