[Socioeconomic inequality and its association with mortality indicators in the departments of Colombia in 2000].

OBJECTIVES: To study the inequalities in various mortality indicators for the departments of Colombia with respect to national figures, and to identify associations between the departmental mortality indicators and departmental socioeconomic indicators.
METHODS: To determine mortality rates and the Gini coefficient for mortality for the departments, data from the death registry were adjusted by the estimated registry coverage for each of the departments. Five socioeconomic indicators were selected: Gini coefficient for income distribution, Human Development Index, per capita gross domestic product, per capita social investment (in health care, etc.), and percentage of the population with health care services from the country's social security system. The differences among the departments were then studied and compared to the national averages. The Spearman's rank correlation coefficient was calculated to find associations between the mortality indicators and each of the five socioeconomic indicators.
RESULTS: For Colombia overall, death registry coverage was estimated at 76%, with Chocó department having the lowest coverage (35%), and Caldas department the highest (88%). The associations between the Gini coefficient for mortality and four of the socioeconomic indicators studied were significant. The national mortality rate was significantly associated with one socioeconomic indicator. Death caused by diabetes mellitus was associated with all the socioeconomic indicators; death caused by undernutrition or by diarrhea, with four socioeconomic indicators; and death from traffic accidents, with two socioeconomic indicators. Homicide was not associated with any of the socioeconomic indicators studied.
CONCLUSIONS: Adjusting the death registry data produced mortality indicators that were more valid for drawing associations with socioeconomic indicators. The Gini coefficient of mortality, mortality from undernutrition, and mortality from diarrheal diseases were more suitable indicators for evaluating the inequalities among the departments because of their higher levels of association with the socioeconomic indicators. Regarding diabetes-related mortality, the associations with all the socioeconomic indicators could be due to systematic errors that lesser-developed departments made when the cause of death was being assigned. A department is a large unit for analysis, which can make it difficult to identify associations between socioeconomic indicators and deaths due to homicide or traffic accidents.