Solution for the management of several clinical practice guidelines in a domain-independent decision support system

Abstract

Malnutrition is a very frequent and serious problem in humans, even more in the elderly. Advanced age brings with it a series of physiological (e.g., swallowing or chewing problems) and psychological changes that can be considered risk factors for malnutrition. It is triggered by loss, dependency, loneliness, and chronic illness, and potentially impacts on higher morbidity, mortality and the worsen of the quality of life. Without intervention, it presents as a downward trajectory leading to poor health and decreased quality of life. That is why it is essential to assess whether a risk situation exists and to evaluate to what extent it can be evitable. Therefore, the main objective of this work is to provide nutritional recommendations through a decision support system considering not only the different nutritional needs, also the whole environment of an elderly patient, such as socio-demographic and economic factors (sex, marital status, education...), psychosocial factors (social relationships, family, physical exercise...), and morbidity factors (diseases). Having this in mind, the aim of this work is to provide the most personalized nutritional recommendations. For this purpose, Clinical Practice Guidelines have been formalized along with experienced nutritionists on the domain within the NUTRIGEP project. The project consists of a product that, in addition to predicting the risk of malnutrition, can prevent it in the geriatric environment, contributing to the good nutritional management of the elderly in order to improve their health condition. Physically it consists of a back-end and a frontend for clinicians and nutritionists, which conceives an integrated solution to support the healthcare professional on the one hand, and to guide the nutritionist in developing personalized diets and preventing malnutrition on the other hand. This work has been focused on the back-end part of this application for the creation, evaluation and management of different rules, using the Drools rule engine and leaning on decision flows, which helped us generating personalized recommendations for patients affected by either one or more pathologies at the same time.