Aplicación de un modelo matemático predictivo para apoyar la identificación de factores de riesgo asociados a la mortalidad infantil en las entidades de salud de la ciudad de Barranquilla
Cargando...
Fecha
2019
Autores
Rodríguez Puente, Leonardo
Guerrero Villanueva, Juan
Título de la revista
ISSN de la revista
Título del volumen
Editor
Ediciones Universidad Simón Bolívar
Facultad de Ingenierías
Facultad de Ingenierías
Resumen
La investigación desarrolla las características de las principales variables sociodemográficas asociadas a la mortalidad infantil, es decir, aquellos factores como tiempo de gestación, sexto, etnia, tipo de parto, multiplicidad de partos, peso, estado civil y régimen de seguridad social, como una forma de estudiar si estos tienen algún tipo de influencia sobre el evento. Para ello, fueron utilizadas las cifras reportadas por el Departamento Administrativo Nacional de Estadística (DANE) con respecto a los nacidos vivos y fallecimientos no fetales ocurridos entre 2013 y 2018.
Una vez determinados estos aspectos, se identificaron los factores de riesgo más significativos en la mortalidad infantil por medio de pruebas de asociación estadística. Esto permitió sentar un primer paso que permitiera comprender la naturaleza de estas relaciones como un insumo para el modelo predictivo. En tal sentido, la regresión de Cox y la técnica de visualización que se desarrolló para mostrar sus resultados es la propuesta central de la investigación, como herramienta clave para la gestión de las entidades de salud en favor de mitigar las consecuencias negativas que deja la mortalidad infantil en la sociedad.
The research develops the characteristics of the main sociodemographic variables associated with infant mortality, that is, those factors such as gestation time, sixth, ethnicity, type of birth, multiplicity of births, weight, marital status and social security regime, such as way to study if they have some kind of influence on the event. For this, the figures reported by the National Administrative Department of Statistics (DANE) were used with respect to live births and non-fetal deaths occurred between 2013 and 2018. Once these aspects were determined, the most significant risk factors in infant mortality were identified through statistical association tests. This allowed to establish a first step that would allow understanding the nature of these relationships as an input for the predictive model. In this sense, the Cox regression and the visualization technique that was developed to show its results is the central proposal of the research, as a key tool for the management of health entities in favor of mitigating the negative consequences of mortality child in society.
The research develops the characteristics of the main sociodemographic variables associated with infant mortality, that is, those factors such as gestation time, sixth, ethnicity, type of birth, multiplicity of births, weight, marital status and social security regime, such as way to study if they have some kind of influence on the event. For this, the figures reported by the National Administrative Department of Statistics (DANE) were used with respect to live births and non-fetal deaths occurred between 2013 and 2018. Once these aspects were determined, the most significant risk factors in infant mortality were identified through statistical association tests. This allowed to establish a first step that would allow understanding the nature of these relationships as an input for the predictive model. In this sense, the Cox regression and the visualization technique that was developed to show its results is the central proposal of the research, as a key tool for the management of health entities in favor of mitigating the negative consequences of mortality child in society.
Descripción
Palabras clave
Factores de riesgo, Mortalidad infantil, Gestión del riesgo, Risk factors, Infant mortality, Risk management