Aplicação do modelo ARIMA na previsão de atendimentos em pontos de atenção com alta demanda da Rede de Assistência à Saúde do município de Monte Carmelo, MG

• Autor(a) principal: Brito, Franciele Guimarães de
• Data de Publicação: 2019
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFU
• Texto Completo: https://repositorio.ufu.br/handle/123456789/26631; http://dx.doi.org/10.14393/ufu.te.2019.20
• topic: Séries temporais; Modelo de média móvel integrado autorregressivo; Necessidades e demandas de serviços de saúde; Time series; Autoregressive integrated moving average model; Needs and demands of health services; Engenharia elétrica; Serviços de saúde; Análises de séries temporais; Modelagem de processos; Modelos matemáticos; CNPQ::ENGENHARIAS::ENGENHARIA BIOMEDICA

The Integrated Delivery Systems within the Unified Health System should promote the systemic integration of actions and health services. Assistance in specialized care services and is a critical factor in the network due to high demand, which generates long waiting times for care. The context possible to cite the extensive queues in the area of cardiology, being diseases of the circulatory system the first cause of death in Brazil in both sexes. The urgency and emergency department is also an important component of the Health Care Network and requires the organization of local health systems and the articulation between the different levels of care to provide qualified and resolutive care. The management of these service demands can be aided by the observation of the behavior of the needs of the population and acquisition of knowledge about the probable evolution of the time series. The objective of this study is to apply the Autoregressive integrated moving average (ARIMA) model to predict care at secondary care points with high demand in the Integrated Delivery Systems of Monte Carmelo, MG. A retrospective study was carried out through the analysis of the time series of the cardiology visits in the period of January/2014 - March/2019 and of Urgency and Emergency department visits classified as green by the Manchester Protocol in the period of January/2014 - February/2019. The time series were analyzed in the temporal domain for the construction of a parametric model with the purpose of performing the forecast of demand. Data processing was performed using Software R Version 3.4. The ARIMA model (3,1,1) presented a better fit for predictiong cardiac care. In relation to the patients classified as green during the reception by the Nurse, according to the Manchester Screening Protocol, in the emergency and emergency room visits, the ARIMA model (2,1,1) presented a better fit. Thus, the application of the models must be seen by the managers as a tool to aid decisions, so it must support processes of planning, management and evaluation of public policies. However, mathematical models for forecasting demand are a tool for managing care and services