Análise de dados da ambiência em Unidades de Terapia Intensiva

Abstract

The Intensive Care Unit is an essentially artificial environment, as it concentrates critically ill patients, highly specialized professionals and state-of-the-art equipment for continuous diagnosis, treatment and monitoring. Its operating dynamics can result in an unwelcoming environment, affecting the health and well-being of its occupants. Environmental variables are often related to various health problems, such as: exposure to high levels of noise for long periods can result in increased blood pressure and heart rate, in addition to hearing disorders; inadequate lighting can cause eyestrain, headache, sleep disturbances and irritation; ambient temperature associated with relative humidity can cause dryness of the skin, eyes and throat; and temperature and humidity can influence the proliferation of fungi, mites, viruses and bacteria. Excessive noise and inadequate lighting are related to sleep disturbance in patients admitted to Intensive Care Units. This can disrupt the circadian cycle and, ultimately, potentiate the emergence of delirium: a severe state of mental confusion that makes it difficult for the patient to recover, interferes with the prognosis and can leave cognitive sequelae. For professionals, the environment can be an amplifier of stress and fatigue levels, contributing to the development of symptoms of Burnout syndrome. Impaired performance can lead to an increase in the error rate and culminate in poorer quality of care and impaired patient safety. In addition, the lack of control over the temperature and humidity of the air can affect from the operation of equipment to the quality of inputs. Variables such as noise, illuminance, temperature and humidity are rarely monitored together and continuously. Without this information, hospitals hardly have indicators and effective protocols to help create a productive, welcoming and humane space – the core of the current health management model. Motivated by this delicate environment, this work presents an embedded system for monitoring environmental variables, collecting and evaluating data captured in Adult Intensive Care Units (ICU) and Neonatal Intensive Care Units (NICU). This system monitors noise, luminosity, temperature and humidity variables 24 hours a day and 7 days a week. We collected data on environmental variables (noise, light, temperature and humidity). Data were collected at a frequency of 30 seconds for 1 month at each site. Monitoring took place in a NICU and an Adult ICU, which at the time was focused on treating patients with Covid-19. This information served to the monitored ICUs as a portrait of their environment during, demonstrating what is going well and what needs to be improved to reach an optimal result. For data evaluation, descriptive statistics metrics such as mean, maximum, minimum and standard deviation were used. In addition to these metrics, in order to identify patterns and characteristics more related to the dynamic systems that describe the data, we use more sophisticated metrics such as Shannon’s classical entropy.