Modelo epidemiológico discreto para a transmissão de Acinetobacter baumannii em UTIs brasileiras

• Ano de defesa: 2014
• Autor(a) principal: Jamielniak, Josemeri Aparecida [UNESP]
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Estadual Paulista (Unesp)
• Palavras-chave em Português: Infeccão hospitalar; Acinetobacter; Epidemiologia; Análise estocástica; Analise de sensibilidade; Markov, Processos de; Epidemiology
• Link de acesso: http://hdl.handle.net/11449/110366

Nosocomial infections are detectable infections that generally occur 48 hours after patient admission. Acinetobacter baumannii is a bacteria related to nosocomial infections, which is the fourth Gram-negative bacteria frequently in brazilian ICUs. The absence of control strategies increases the permanence of patients and the cost of treatment, therefore intervention and control measures have become extremely important. Mathematical models approach for epidemiology has been growing and can be used to support decisions. Thus, in this work we proposed a compartimental mathematical model in discrete time to describe A. baumannii transmission in a brazilian ICU with parameters from the literature. We considered that contamination occured because of either the environment (e.g. invasive methods, surgical techniques) or contact between patient and health care workers that play the role of disease vectors. We chose a discrete mathematical model since surveillance cultures are collected per day. In deterministic version we analysed stability and sensitivity of equilibrium points to model parameters in two situations in which probability of isolation for colonized patients was varyied, supposing the hospital accomplishes a active search to identify colonized patients by using oropharynx swab and oropharynx swab plus axillae swab, without active search, supposing a more colonized environment. In stochastic version, we used Discrete Time Markov Chains to simulate and assess the effectiveness of some control measures like colonization by the environment and contact and isolation probabilities. We verified the great importance of the hygienization of hospital environment and the identification of patient colonization