Resumo Este estudo pretende identificar a prevalência de interações medicamentosas potenciais (IMP) em idosos institucionalizados e seus fatores associados. Realizamos um estudo observacional, multicêntrico e transversal, durante o último trimestre de 2010. Selecionamos uma amostra de 275 sujeitos (≥ 65 anos) de 10 instituições para idosos de Murcia (Espanha) mediante amostragem aleatória complexa em duas etapas. As IMP foram identificadas usando a base de dados do College of Pharmacists. Estimamos a prevalência de IMP de relevância clínica e analisamos os fatores associados com análise de regressão uni e multinível. Identificamos 210 IMP, das quais 120 foram consideradas clinicamente relevantes (57,1%) e afetaram 70 idosos (25,8%). Oito grupos farmacológicos constituíram 70,2% das IMP clinicamente relevantes. A prevalência de IMP esteve associada à multimorbidade (OR = 2,3; IC 95% = 1,4-4,5) e tomar dez ou mais medicamentos diariamente (OR = 9,6; IC95% = 4,8-19,1) e uso de medicamentos anti-inflamatórios (OR = 3,9; IC 95% = 1,4-10,4). Este estudo revela que as IMP clinicamente relevantes são muito comuns em idosos institucionalizados e que os serviços devem melhorar seus processos para reduzir a prevalência deste fenômeno.

Abstract This study purposes to determine the prevalence of potential and clinical relevant Drug-Drug-Interactions (pDDIs) in institutionalized older adults and to identify the pertinent factors associated. We conduct an observational, multicenter and cross-sectional study during the last quarter of 2010. We selected a sample of 275 subjects (aged ≥ 65 years) from 10 nursing homes of Murcia (Spain) by a two-stage complex sampling. pDDIs were identified using the College of Pharmacists Database. We only considered pDDIs of clinical relevance, and thereafter the relevant factors were identified through uni-level and multi-level regression analyses. A total of 210 pDDIs were identified, 120 of which were considered clinically relevant (57.1%), affecting a total of 70 elderly (25.8%). Eight pharmacological groups made up 70.2% of the clinically relevant pDDIs. More clinically relevant DDIs were found in people suffering several pathologies (OR = 2.3; 95%CI = 1.4-4.5), and also in people who take ten or more drugs daily (OR = 9.6; 95%CI = 4.8-19.1), and people who take anti-inflammatory drugs (OR = 3.9; 95%CI = 1.4-10.4). This study reveals that clinically relevant pDDIs are very common in institutionalized elderly people, and that caregivers should aim at improving their practice in order to reduce the prevalence of this phenomenon.

Resumo Este estudo pretende identificar a prevalência de interações medicamentosas potenciais (IMP) em idosos institucionalizados e seus fatores associados. Realizamos um estudo observacional, multicêntrico e transversal, durante o último trimestre de 2010. Selecionamos uma amostra de 275 sujeitos (≥ 65 anos) de 10 instituições para idosos de Murcia (Espanha) mediante amostragem aleatória complexa em duas etapas. As IMP foram identificadas usando a base de dados do College of Pharmacists. Estimamos a prevalência de IMP de relevância clínica e analisamos os fatores associados com análise de regressão uni e multinível. Identificamos 210 IMP, das quais 120 foram consideradas clinicamente relevantes (57,1%) e afetaram 70 idosos (25,8%). Oito grupos farmacológicos constituíram 70,2% das IMP clinicamente relevantes. A prevalência de IMP esteve associada à multimorbidade (OR = 2,3; IC 95% = 1,4-4,5) e tomar dez ou mais medicamentos diariamente (OR = 9,6; IC95% = 4,8-19,1) e uso de medicamentos anti-inflamatórios (OR = 3,9; IC 95% = 1,4-10,4). Este estudo revela que as IMP clinicamente relevantes são muito comuns em idosos institucionalizados e que os serviços devem melhorar seus processos para reduzir a prevalência deste fenômeno.

Abstract This study purposes to determine the prevalence of potential and clinical relevant Drug-Drug-Interactions (pDDIs) in institutionalized older adults and to identify the pertinent factors associated. We conduct an observational, multicenter and cross-sectional study during the last quarter of 2010. We selected a sample of 275 subjects (aged ≥ 65 years) from 10 nursing homes of Murcia (Spain) by a two-stage complex sampling. pDDIs were identified using the College of Pharmacists Database. We only considered pDDIs of clinical relevance, and thereafter the relevant factors were identified through uni-level and multi-level regression analyses. A total of 210 pDDIs were identified, 120 of which were considered clinically relevant (57.1%), affecting a total of 70 elderly (25.8%). Eight pharmacological groups made up 70.2% of the clinically relevant pDDIs. More clinically relevant DDIs were found in people suffering several pathologies (OR = 2.3; 95%CI = 1.4-4.5), and also in people who take ten or more drugs daily (OR = 9.6; 95%CI = 4.8-19.1), and people who take anti-inflammatory drugs (OR = 3.9; 95%CI = 1.4-10.4). This study reveals that clinically relevant pDDIs are very common in institutionalized elderly people, and that caregivers should aim at improving their practice in order to reduce the prevalence of this phenomenon.

Abstract Objective: To identify predisposing, inhibitory and facilitating factors that may affect participation in colorectal cancer screening programs in the Roma population. Method: Qualitative and exploratory study by focus group technique. Three focus groups of men and three groups of women were carried out, with a total of 16 men and 18 women from the Roma population, aged 50-69 years, from the province of Alicante. A discourse analysis was performed with the PRECEDE model as an analysis framework. Results: Several barriers to participation were identified, such as the aversion of the Roma population to talk about cancer, refusal to anticipate a diagnosis that can cause suffering to the person and their family, poor knowledge of the disease and the preventive programmes, refusal to collect and handle samples, fear and shame about the colonoscopy, acceptance of divine will, difficulties in understanding and reading, and the perception of being discriminated by their ethnicity in the health sector. However, predisposing factors to participate in the screening programme also were identified, for example willingness to receive relevant information through more appropriate pathways, as well as their confidence in professional counselling. Conclusion: A willingness to receive the recommendation to participate and understanding information from health professionals have been identified. This will enable us to envisage potential strategies for approaching this population group. This could contribute to improved participation of the Roma population in colorectal cancer screening programmes and to open up new ways to promote preventive behaviours.

Resumen Objetivo: Identificar los factores predisponentes, inhibidores y facilitadores que pueden afectar a la participación en los programas de cribado del cáncer colorrectal en la población gitana. Método: Estudio cualitativo exploratorio mediante técnica de grupo focal. Se realizaron tres grupos focales de hombres y tres de mujeres, con un total de 16 hombres y 18 mujeres participantes de población romaní, de 50-69 años de edad, de la provincia de Alicante. Se realizó un análisis del discurso con el modelo PRECEDE como marco de análisis. Resultados: Se identificaron diversas barreras a la participación, como la aversión de la población gitana a hablar de cáncer, el rechazo a anticipar un diagnóstico que puede provocar sufrimiento de la persona y su entorno familiar, escaso conocimiento de la enfermedad y del programa preventivo, rechazo hacia la recogida y manejo de las muestras, miedo y vergüenza ante la posible colonoscopia, aceptación de la voluntad divina, dificultades de comprensión y lectura, y percepción de estar discriminados por su etnia en el ámbito sanitario. Sin embargo, se identificaron factores predisponentes a la participación en el programa de cribado, como la predisposición favorable a recibir información pertinente a través de vías más apropiadas, así como su confianza en el consejo profesional. Conclusión: La predisposición a recibir la recomendación de participar en los programas de cribado colorrectal y una información comprensible de los profesionales sanitarios permiten entrever posibles estrategias de aproximación a este colectivo. Esto podría contribuir a la mejora de la participación de la población gitana en el programa de cribado de cáncer colorrectal y abrir nuevas vías para promover conductas preventivas.

Abstract Objective: To identify predisposing, inhibitory and facilitating factors that may affect participation in colorectal cancer screening programs in the Roma population. Method: Qualitative and exploratory study by focus group technique. Three focus groups of men and three groups of women were carried out, with a total of 16 men and 18 women from the Roma population, aged 50-69 years, from the province of Alicante. A discourse analysis was performed with the PRECEDE model as an analysis framework. Results: Several barriers to participation were identified, such as the aversion of the Roma population to talk about cancer, refusal to anticipate a diagnosis that can cause suffering to the person and their family, poor knowledge of the disease and the preventive programmes, refusal to collect and handle samples, fear and shame about the colonoscopy, acceptance of divine will, difficulties in understanding and reading, and the perception of being discriminated by their ethnicity in the health sector. However, predisposing factors to participate in the screening programme also were identified, for example willingness to receive relevant information through more appropriate pathways, as well as their confidence in professional counselling. Conclusion: A willingness to receive the recommendation to participate and understanding information from health professionals have been identified. This will enable us to envisage potential strategies for approaching this population group. This could contribute to improved participation of the Roma population in colorectal cancer screening programmes and to open up new ways to promote preventive behaviours.

Resumen Objetivo: Identificar los factores predisponentes, inhibidores y facilitadores que pueden afectar a la participación en los programas de cribado del cáncer colorrectal en la población gitana. Método: Estudio cualitativo exploratorio mediante técnica de grupo focal. Se realizaron tres grupos focales de hombres y tres de mujeres, con un total de 16 hombres y 18 mujeres participantes de población romaní, de 50-69 años de edad, de la provincia de Alicante. Se realizó un análisis del discurso con el modelo PRECEDE como marco de análisis. Resultados: Se identificaron diversas barreras a la participación, como la aversión de la población gitana a hablar de cáncer, el rechazo a anticipar un diagnóstico que puede provocar sufrimiento de la persona y su entorno familiar, escaso conocimiento de la enfermedad y del programa preventivo, rechazo hacia la recogida y manejo de las muestras, miedo y vergüenza ante la posible colonoscopia, aceptación de la voluntad divina, dificultades de comprensión y lectura, y percepción de estar discriminados por su etnia en el ámbito sanitario. Sin embargo, se identificaron factores predisponentes a la participación en el programa de cribado, como la predisposición favorable a recibir información pertinente a través de vías más apropiadas, así como su confianza en el consejo profesional. Conclusión: La predisposición a recibir la recomendación de participar en los programas de cribado colorrectal y una información comprensible de los profesionales sanitarios permiten entrever posibles estrategias de aproximación a este colectivo. Esto podría contribuir a la mejora de la participación de la población gitana en el programa de cribado de cáncer colorrectal y abrir nuevas vías para promover conductas preventivas.

ABSTRACT Background: Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) are widely used but they increase the risk of gastrointestinal haemorrage among other adverse effects. The objective of this study was to compare potentially inappropriate medications (PIM) by NSAIDs using the original Beers Criteria, a global reference for evaluating elderly people’s prescriptions, and the Spanish adaptation of the same; and the relation between PIM of NSAIDs and gastrointestinal bleeding. Methods: The study was a retrospective observational study carried out located in a primary care district in the province of de Murcia, south-eastern Spain. The study population (n=7.856) were citizens aged 65 and above, with at least one drug prescribed in a Spanish health district during the study period . We analized illnesses and treatments registered in the primary care’s electronic medical history of patients and hospital admissions, during the 12 month study period (2012). The original Beers Criteria and their Spanish adaptation were used to evaluate PIM of NSAIDs in patients considering the medication globally and also each active substance. Gastrointestinal bleeding events recorded in the data bases studied were evaluated. Results: Detection of PIM of NSAIDs was 5,6% with the original version and 7,0% (Δ=25,5%; p<0,001) with the adapted one. PIM of NSAIDs was related with an increased incidence of gastrointestinal bleeding without significant differences between PIM exposed and NSAID exposed patients AINE (RR=1,6; IC:0,2-14,5). Conclusions: The Spanish adaptation of the Beers criteria identified a greater degree of PIM of NSAIDs than the original version, and in both versions the detection of PIM was not related with a significant increase of gastrointestinal bleeding compared to patients exposed to NSAIDs.

RESUMEN Fundamentos: Los Antiinflamatorios No Esteroideos (AINE) son un grupo de medicamentos con uso muy extendido en la población, su uso genera un mayor riesgo de hemorragia digestiva. El objetivo de este trabajo fue evaluar y comparar la prescripción potencialmente inadecuada (PPI) de AINE según los criterios de Beers en su versión original con su adaptación española y la relación de esta PPI con los eventos de sangrado gastrointestinal. Métodos: Estudio observacional longitudinal retrospectivo de 12 meses (año 2012) realizado en un área de salud de la Región de Murcia. La población estudiada fue los mayores de 65 años a los que se les había prescrito, al menos, 1 receta médica de AINE durante el periodo de estudio (7.856). Se utilizaron ambas versiones de los criterios de Beers para evaluar la PPI por AINE. Para evaluar el papel etiológico de la exposición a AINE potencialmente inadecuados, respecto a haber recibido AINE, en la hemorragia gastrointestinal se calculó la incidencia acumulada y el Riesgo Relativo. Resultados: La detección de PPI por AINE pasó de 5,6% con la versión original, a 7,0% (Δ=25,5%; p<0,001) En los sujetos con prescripción de AINE la exposición a PPI por AINE presentó una mayor incidencia de sangrado gastrointestinal pero sin diferencias significativas respecto a la población que recibió AINE (RR=1,6; IC:0,2-14,5). Conclusiones: El uso de la adaptación española de los criterios de Beers posibilita una mayor detección de PPI por AINE en comparación con el uso de la versión original, en ambas versiones, la PPI por AINE no genera un incremento significativo en el sangrado gastrointestinal respecto a recibir AINE.

ABSTRACT Background: Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) are widely used but they increase the risk of gastrointestinal haemorrage among other adverse effects. The objective of this study was to compare potentially inappropriate medications (PIM) by NSAIDs using the original Beers Criteria, a global reference for evaluating elderly people’s prescriptions, and the Spanish adaptation of the same; and the relation between PIM of NSAIDs and gastrointestinal bleeding. Methods: The study was a retrospective observational study carried out located in a primary care district in the province of de Murcia, south-eastern Spain. The study population (n=7.856) were citizens aged 65 and above, with at least one drug prescribed in a Spanish health district during the study period . We analized illnesses and treatments registered in the primary care’s electronic medical history of patients and hospital admissions, during the 12 month study period (2012). The original Beers Criteria and their Spanish adaptation were used to evaluate PIM of NSAIDs in patients considering the medication globally and also each active substance. Gastrointestinal bleeding events recorded in the data bases studied were evaluated. Results: Detection of PIM of NSAIDs was 5,6% with the original version and 7,0% (Δ=25,5%; p<0,001) with the adapted one. PIM of NSAIDs was related with an increased incidence of gastrointestinal bleeding without significant differences between PIM exposed and NSAID exposed patients AINE (RR=1,6; IC:0,2-14,5). Conclusions: The Spanish adaptation of the Beers criteria identified a greater degree of PIM of NSAIDs than the original version, and in both versions the detection of PIM was not related with a significant increase of gastrointestinal bleeding compared to patients exposed to NSAIDs.

RESUMEN Fundamentos: Los Antiinflamatorios No Esteroideos (AINE) son un grupo de medicamentos con uso muy extendido en la población, su uso genera un mayor riesgo de hemorragia digestiva. El objetivo de este trabajo fue evaluar y comparar la prescripción potencialmente inadecuada (PPI) de AINE según los criterios de Beers en su versión original con su adaptación española y la relación de esta PPI con los eventos de sangrado gastrointestinal. Métodos: Estudio observacional longitudinal retrospectivo de 12 meses (año 2012) realizado en un área de salud de la Región de Murcia. La población estudiada fue los mayores de 65 años a los que se les había prescrito, al menos, 1 receta médica de AINE durante el periodo de estudio (7.856). Se utilizaron ambas versiones de los criterios de Beers para evaluar la PPI por AINE. Para evaluar el papel etiológico de la exposición a AINE potencialmente inadecuados, respecto a haber recibido AINE, en la hemorragia gastrointestinal se calculó la incidencia acumulada y el Riesgo Relativo. Resultados: La detección de PPI por AINE pasó de 5,6% con la versión original, a 7,0% (Δ=25,5%; p<0,001) En los sujetos con prescripción de AINE la exposición a PPI por AINE presentó una mayor incidencia de sangrado gastrointestinal pero sin diferencias significativas respecto a la población que recibió AINE (RR=1,6; IC:0,2-14,5). Conclusiones: El uso de la adaptación española de los criterios de Beers posibilita una mayor detección de PPI por AINE en comparación con el uso de la versión original, en ambas versiones, la PPI por AINE no genera un incremento significativo en el sangrado gastrointestinal respecto a recibir AINE.

Objective. To analyze the relationship of mental health problems in Spanish population with the economic recession (2006-2012), and find out how it affects the self-perception of health status. Materials and methods. Cross-sectional study using the National Health Survey of Spain, 2006/2007 and 2011/2012. Using logistic regression models, three indicators linked to mental health and perceived health were analyzed. Results. In 2011/2012 the consumption of anti-anxiety medications and sleeping pills increased in men and women. Mental dysfunction increased during the economic crisis in the male population. The perception of optimal health did not suffer significantly in either men or women. Conclusions. The economic recession showed a changing relation to the mental and general health of the population, coinciding with an increase in mental health disorders, such as anxiety.

Objetivo. Analizar la relación de los problemas de salud mental en población española con la recesión económica (2006-2012) y establecer en qué sentido afecta a la autopercepción del estado de salud. Material y métodos. Estudio transversal comparativo utilizando la Encuesta Nacional de Salud de España, 2006/2007 y 2011/2012. Mediante modelos de regresión logística, se analizaron tres indicadores relacionados con la salud mental y la salud percibida. Resultados. En 2011/2012 aumentó el consumo de medicamentos ansiolíticos y somníferos en hombres y mujeres. La disfunción mental aumentó durante el periodo de crisis económica en la población de varones. La percepción de una salud óptima no sufrió cambios significativos en hombres ni en mujeres. Conclusiones. La recesión económica mostró una relación variable con la salud mental y general de la población, y coincidió con un aumento de los trastornos de salud mental, como la ansiedad.

Objective. To analyze the relationship of mental health problems in Spanish population with the economic recession (2006-2012), and find out how it affects the self-perception of health status. Materials and methods. Cross-sectional study using the National Health Survey of Spain, 2006/2007 and 2011/2012. Using logistic regression models, three indicators linked to mental health and perceived health were analyzed. Results. In 2011/2012 the consumption of anti-anxiety medications and sleeping pills increased in men and women. Mental dysfunction increased during the economic crisis in the male population. The perception of optimal health did not suffer significantly in either men or women. Conclusions. The economic recession showed a changing relation to the mental and general health of the population, coinciding with an increase in mental health disorders, such as anxiety.

Objetivo. Analizar la relación de los problemas de salud mental en población española con la recesión económica (2006-2012) y establecer en qué sentido afecta a la autopercepción del estado de salud. Material y métodos. Estudio transversal comparativo utilizando la Encuesta Nacional de Salud de España, 2006/2007 y 2011/2012. Mediante modelos de regresión logística, se analizaron tres indicadores relacionados con la salud mental y la salud percibida. Resultados. En 2011/2012 aumentó el consumo de medicamentos ansiolíticos y somníferos en hombres y mujeres. La disfunción mental aumentó durante el periodo de crisis económica en la población de varones. La percepción de una salud óptima no sufrió cambios significativos en hombres ni en mujeres. Conclusiones. La recesión económica mostró una relación variable con la salud mental y general de la población, y coincidió con un aumento de los trastornos de salud mental, como la ansiedad.

Objetive: to determine the prevalence of malnutrition in patients aged 65 years or more at admission and factors associated with its presence. Analyze excess hospital stay (EHS), economic impact and premature readmission rate associated with hospital malnutrition in elderly patient. Material and method: retrospective study conducted at the University Hospital Reina Sofía. All patients aged 65 years or older admitted to internal medicine in 2011. The sample size was calculated taking into account the income of the previous year, and considering a prevalence of malnutrition of 50% with a 95% and included error of 5%. To define the degree of malnutrition nutritional control tool (CONUT), which establishes a score based on albumin, total cholesterol and lymphocyte determination was used. To determine the factors associated with the presence of moderate to severe malnutrition analysis of multivariate logistic regression was performed. For each patient the EHS, premature readmissions and the associated cost to EHS was calculated. A threshold of statistical significance of 0.05 was used for all analyzes and were performed with SPSS v15.0. Results: 310 patients, of whom 54.2% were women were included, the mean age was 80.1 years (SD: 6.8), ranging between 65 and 95 years. Regarding diagnosis at admission 27.4% were respiratory diseases, 22.6% of the circulatory and digestive 11.6%. The median Charlson index was 2.0, found that 36.8% of patients had high comorbidity. The most prevalent chronic diseases were diabetes mellitus (44.2%), chronic kidney disease (25.2%) and dementia (10.6). Regarding the CONUT, 75.8% of patients met the criteria of malnutrition: 42.6% mild, 28.7% moderate and severe 4.5%, of which only 46.6% had some nutritional support during admission. Factors associated with the presence of moderate to severe malnutrition were female gender (OR: 1.7; 95%: 1.1 - 2.8), age over 80 years (OR: 2.0, IC 95%: 1.2 - 3.5), and dementia (OR: 2.4; IC 95%:1.2 - 5.2). No association with comorbidity or with other chronic diseases was found. Regarding the EHS (days) differences between patients with moderate to severe malnutrition (4.7; IC 95%: 2.3 - 7.1) and normally nourished (-0.1; IC 95%: -1.4 - 1.2) (p = 0.001) were found, but were not for cases of mild malnutrition (1.6, 95%: 0.5-2.8) (p = 0.07).Regarding the rate of premature readmission in malnourished patients was 28/235 (11.9%). The cost associated with EHS / 100 patients was € 195 479.4 for moderate malnutrition, mild malnutrition € 73 484.8, and normally nourished patients represented a saving of € 12 353. Conclusions: hospital malnutrition in elderly patients remains an unsolved problem, given the high prevalence found, associated to an excess of hospital stay and increased hospital costs, especially in patients with moderate to severe malnutrition. The CONUT is a nutritional screening tool very useful for the speed and validity of their results, and allows detecting patients at risk or nutritional alert without lead to increased costs.

Objetivo: determinar la prevalencia de desnutrición en pacientes con edad igual o superior a 65 años al ingreso hospitalario y los factores asociados a su presencia. Analizar el exceso de estancia hospitalaria (EEH), el impacto económico y la tasa de reingresos prematuros asociados a la desnutrición hospitalaria en pacientes de edad avanzada. Material y métodos: estudio retrospectivo realizado en el Hospital Universitario Reina Sofía. Se incluyeron todos los pacientes con edad igual o mayor a 65 años que ingresaron en Medicina Interna durante 2011. Se calculó el tamaño muestral teniendo en cuenta los ingresos del año anterior, y considerando una prevalencia de desnutrición del 50% con un IC 95% y un error del 5%. Para definir el grado de desnutrición se empleó la herramienta Control Nutricional (CONUT), que establece una puntuación basada en la determinación de albúmina, colesterol total y linfocitos. Para determinar los factores asociados a la presencia de desnutrición moderada-grave se realizó un análisis de regresión logística multivariante. Para cada paciente se calculó el EEH, los reingresos prematuros y el coste asociado al EEH. Para todos los análisis se utilizó un dintel de significación estadística de 0,05 y se realizaron con el paquete estadístico SPSS v15.0. Resultados: se incluyeron 310 pacientes, de los cuales el 54,2% fueron mujeres, la edad media fue de 80,1 años (DE: 6,8), con un rango entre 65 y 95 años. En cuanto al diagnóstico al ingreso, el 27,4% correspondían a enfermedades del aparato respiratorio, 22,6% del circulatorio y 11,6% del digestivo. La mediana del Índice de Charlson fue de 2,0, encontrando que el 36,8% de los pacientes presentaban una comorbilidad alta. Las patologías crónicas más prevalentes fueron la diabetes mellitus (44,2%), la enfermedad renal crónica (25,2%) y la demencia (10,6). En relación con el CONUT, el 75,8% de los pacientes presentaban criterios de desnutrición: el 42,6% leve, el 28,7% moderada y el 4,5% grave, de estos, solo un 46,6% tuvo algún tipo de soporte nutricional durante el ingreso. Los factores asociados a la presencia de desnutrición moderada-grave fueron el sexo femenino (OR: 1,7; IC 95%: 1,1 - 2,8), edad mayor de 80 años (OR: 2,0, IC 95%: 1,2 - 3,5), y la demencia (OR: 2,4; IC 95%:1,2 - 5,2). No se encontró asociación con la comorbilidad ni con otras patologías crónicas. Respecto al EEH (días), se encontraron diferencias entre los pacientes con desnutrición moderada-grave (4,7; IC 95%: 2,3 - 7,1) y normonutridos (-0,1; IC 95%: -1,4 - 1,2) (p = 0,001), no siendo así para los casos de desnutrición leve (1,6; IC 95%: 0,5-2,8) (p = 0,07). En relación a la tasa de reingresos prematuros en pacientes desnutridos fue de 28/235 (11,9%). El coste asociado al EEH/100 pacientes fue de 195.479,4 € para la desnutrición grave-moderada, 73.484,8 € desnutrición leve, mientras que en los pacientes normonutridos supuso un ahorro de 12.353 €. Conclusiones: la desnutrición hospitalaria en el paciente anciano sigue siendo un problema sin resolver, dada la elevada prevalencia encontrada, asociándose a un exceso de estancia hospitalaria y a un aumento del gasto hospitalario, especialmente en pacientes con desnutrición moderada- severa. El CONUT es una herramienta de cribado nutricional de gran utilidad por la rapidez y validez de sus resultados, y permite detectar pacientes con riesgo o alerta nutricional, sin suponer un incremento de costes.