Modelos de fragilidade e multiestado na avaliação do tempo de internamentos hospitalares por Covid-19 em Minas Gerais

Abstract

Since December, 2019, many patients were interned due a pneumonia of unknown cause and discovered that the infection was caused by a new virus type from the family Crown virus, denominated SARS-CoV-2 and the disease originated by the virus was called Covid-19, the one which caused the pandemic. Since then, various researches and data basis have been made available and the statistic analysis have been proposed, focusing either on the clinic and the administrative aspects due the internment demand raises. Therefore on this paper, it was an objective to adjust models of surviving analysis including the fragility effect and multi-state models on the internment time of patients carriers of Covid-19, in the state of Minas Gerais. The obtained data variables were associated to each of the 24.946 patients diagnosed on Covid-19. Fragility models were adjusted to analyze the internment time considering the public and private management of the hospitals, considering the public or private management of hospitals, besides the covariates of gender, age, necessity of hospitalization in Intensive Care Unit (ICU) and presence of secondary diagnosis, the death was considered a censored data. From the hospitalized patients carriers of Covid-19, 53,54% were male, 33,81% needed the ICU, 8,86% of the patients had at least a secondary diagnosis, being the patient’s average age equal to 61,35 years. In the case of patients that died the average age raised to 70,29 years. Considering the fragility model adjusted to the log- normal distribution, it was estimated that the recovering risk decreases 1,33% to each year raise on the patient’s age. Patients with more hospitalization times were the ones in which there was the presence of secondary diagnosis, necessity of ICU and they were from the male gender. In the multi-state model, it was observed that the variable ‘age’, all the transitions were significant. Besides that, patients which were treated in public hospitals had longer internment times than the patients that were treated private hospitals. Therefore, the fragility and multi-state models study in the evaluation of the hospital internment times, turns to be an important auxiliary resource to the competent organs.