Objective:

To describe epidemiological characteristics, as well as maternal deaths registration characteristics in the population registered in a subsidized healthcare service provider in the state of Cauca (Colombia).

Methodology:

Descriptive, retrospective study. Univariate analysis was performed for the characterization of deaths, a mapping of their spatial location, and a graphical presentation of the number of deaths per year. Bivariate analysis with the Mann-Whitney U test and the Kruskal-Wallis test was also implemented. The concordance in records was assessed using Cohen's Kappa coefficient.

Results:

There were 34 maternal deaths between 2009 and 2014, with a rate of maternal mortality of 193, 141, 231, 67 and 110 deaths per 100,000 live births for the period from 2010 to 2014. The incidents occurred in women who mostly resided in scattered rural areas (67.6 %), had a low level of education (63.3 %), were housewives (73.5 %) and did not attend antenatal care or did it lately (67.7 %). Poor concordance was identified between institutional documents and reports to the Colombian Public Health Surveillance System (K = -0.0282, p = 0.8116).

Conclusions:

The maternal deaths compiled from the entity information system during the study and which had a profile linked to risky conditions were, in most cases, preventable. Although the entity used a strategy to identify delays in three aspects of each maternal mortality case, this assessment is to be completed for all events, accompanied by complete records to facilitate follow-up processes.

        · abstract in Spanish | Portuguese     · text in Spanish     · Spanish ( pdf )