Manejo clínico de la EPOC en situación de vida real. Análisis a partir de big data / Clinical Management of COPD in a Real-World Setting. A Big Data Analysis

OBJETIVO: El objetivo de este estudio es evaluar la calidad del diagnóstico y el tratamiento de la EPOC utilizando metodología de big data mediante la plataforma clínica Savana Manager 2.1. MATERIAL Y MÉTODOS: Sobre una población de 1.219.749 sujetos mayores de 40 años se incluyó a 59.369 pacientes con un diagnóstico de EPOC. RESULTADOS: El 78% de ellos eran varones. Solo 26.453 (43,5%) disponían de espirometría. En 18.172 pacientes se hizo una aproximación a la gravedad de su proceso: 4.396 leves, 7.100 moderados y 6.676 graves, aunque solo disponían de espirometría obstructiva el 27, el 34 y el 28%, respectivamente. El manejo clínico de la EPOC recae fundamentalmente en Atención Primaria y Neumología, con un papel relevante de Medicina Interna y, en menor medida, de Geriatría. El tratamiento farmacológico está basado en el uso de broncodilatadores y corticoides inhalados (CI). Se observa un marcado descenso en la utilización de los beta-2-agonistas de larga duración (LABA) en monoterapia y una leve reducción de combinaciones de CI/LABA, asociados a un LAMA en el 74% de los casos. La mortalidad hospitalaria por cualquier causa de la población global fue del 5,60% frente al 1% de la población general mayor de 40 años. El 35% presentó un ingreso hospitalario, con una estancia media de 6,6 días y una tasa de mortalidad hospitalaria en este grupo del 10,74%. DISCUSIÓN: Este estudio identifica cuáles son las principales características de una población no seleccionada de EPOC y cuáles son los principales errores en el manejo de la enfermedad

OBJECTIVE: The aim of this study was to evaluate the quality of diagnosis and treatment of COPD using Big Data methodology on the Savana Manager 2.1 clinical platform. MATERIALS AND METHODS: A total of 59,369 patients with a diagnosis of COPD were included from a population of 1,219,749 adults over 40 years of age. RESULTS: In total, 78% were men. Spirometry data were available for only 26,453 (43.5%) subjects. Disease severity was classified in 18,172 patients: 4,396 mild, 7,100 moderate, and 6,676 severe, although only 27%, 34%, and 28%, respectively, presented obstructive spirometry. The clinical management of COPD is mainly the responsibility of the primary care and pulmonology departments, while internal medicine and, to a lesser extent, geriatrics also participate. Drug treatment was based on bronchodilators and inhaled corticosteroids (ICS). A marked decline in the use of long-acting beta-2 agonists (LABA) in monotherapy and a slight reduction in ICS/LABA combinations, associated with a LAMA in 74% of cases, was observed. All-cause in-hospital mortality among the overall population was 5.6% compared to 1% of the general population older than 40 years. In total, 35% were admitted to hospital, with an average stay of 6.6 days and a rate of hospital mortality in this group of 10.74%. DISCUSSION: This study identifies the main features of an unselected COPD population and the major errors made in the management of the disease

Clinical Management of COPD in a Real-World Setting. A Big Data Analysis. / Manejo clínico de la EPOC en situación de vida real. Análisis a partir de big data.

OBJECTIVE: The aim of this study was to evaluate the quality of diagnosis and treatment of COPD using Big Data methodology on the Savana Manager 2.1 clinical platform. MATERIALS AND METHODS: A total of 59,369 patients with a diagnosis of COPD were included from a population of 1,219,749 adults over 40 years of age. RESULTS: In total, 78% were men. Spirometry data were available for only 26,453 (43.5%) subjects. Disease severity was classified in 18,172 patients: 4,396 mild, 7,100 moderate, and 6,676 severe, although only 27%, 34%, and 28%, respectively, presented obstructive spirometry. The clinical management of COPD is mainly the responsibility of the primary care and pulmonology departments, while internal medicine and, to a lesser extent, geriatrics also participate. Drug treatment was based on bronchodilators and inhaled corticosteroids (ICS). A marked decline in the use of long-acting beta-2 agonists (LABA) in monotherapy and a slight reduction in ICS/LABA combinations, associated with a LAMA in 74% of cases, was observed. All-cause in-hospital mortality among the overall population was 5.6% compared to 1% of the general population older than 40 years. In total, 35% were admitted to hospital, with an average stay of 6.6 days and a rate of hospital mortality in this group of 10.74%. DISCUSSION: This study identifies the main features of an unselected COPD population and the major errors made in the management of the disease.

Improving Adherence to Clinical Pathways Through Natural Language Processing on Electronic Medical Records.

This paper presents a pioneering and practical experience in the development and implementation of a clinical decision support system (CDSS) based on natural language processing (NLP) and artificial intelligence (AI) techniques. Our CDSS notifies primary care physicians in real time about recommendations regarding the healthcare process. This is, to the best of our knowledge, the first real-time CDSS implemented in the Spanish National Health System. We achieved adherence rate improvements in eight out of 18 practices. Moreover, the provider's feedback was very positive, describing the solution as fast, useful, and unintrusive. Our CDSS reduced clinical variability and revealed the usefulness of NLP and AI techniques for the evaluation and improvement of health care.