Researchers of the St. Petersburg Federal Research Center of the Russian Academy of Sciences (SPC RAS) have developed software able to support socially-oriented decision-making at emergency hospitalization of great number of patients in the context of a pandemic. The above development will definitely increase the efficiency of medical institutions fighting against the pandemic, avoid queues of the ambulances and account for the uncertainty factor at hospitalization in the pandemic.
"We have proposed a computer model that supports the ambulance dispatch center in case of emergency hospitalization of a great number of patients in the pandemic. In its work, the model accounts for many factors, including the fact of reporting the disease symptoms, the number of available hospitals and their workload, the age of the patient, the speed of examination and patients’ admission, the patients' residential remoteness from hospitals. The need to solve the problem of making operational decisions at emergency hospitalization could be easily traced by news feeds and messages on social networks with photos of ambulance queues at the emergency rooms of hospitals at the increase in the incidence of coronavirus," says Nikolai Teslya, Senior Researcher of the Laboratory of Integrated Automation Systems at the St. Petersburg Institute for Informatics and Automation of the Russian Academy of Sciences (SPIIRAS is the SPC RAS structural division).
When a pandemic had only started, in order to help medical staff, the Ministry of Health developed the actions’ regulation for making decisions on hospitalization of patients with signs of coronavirus disease. According to this regulation, doctors had to assess the symptoms, determine the disease severity and decide on hospitalization, CT scan or home quarantine. However, the course of the pandemic is rapidly changing, and the existing guidelines for the hospitalization management may not keep up with the situation progress. In addition, operational information includes a great number of parameters, their tracking and evaluation is quite a complex task for any dispatcher.
"Throughout the study, we collected a set of data from ambulance stations in two districts of St. Petersburg. They cover completely depersonalized information about patients and doctors: age, symptoms, preliminary diagnosis for the patient and the total duration of the shift, time spent with the patient, age and gender of medical staff. It turned out that some of the factors studied can really affect the speed and efficiency of decision-making. Taking these factors into account at the moment of making a decision on hospitalization will allow for a better uniform patients’ dispatching," Nikolai Teslya notes.
So, the study showed the interdependence between the hospitalization duration and the age of the patient: on average, the older the patient, the longer the doctors who arrived on call make a decision. Moreover, an analysis of the hospitals’ work revealed that they spend different time for admitting patients with coronavirus, and therefore hospitals were divided into three categories according to the hospitalization’s speed: this factor can also be accounted for at making decisions, depending on the patient's condition severity. Upon collecting and analyzing operational information about the current situation, the program will propose a hospitalization scheme, that will include a computed tomography center, where, if additional examination is needed, the patient should be delivered, and a hospital where a patient should be transported to if the diagnosis were confirmed.
The computer model operation is designed to provide the dispatcher with a solution that most effectively coordinates actions of all participants in the hospitalization process (patients and their relatives, hospitals, CT centers, as well as ambulance crews). Calculations have shown that the model can calculate about 6 million solutions less than in 10 minutes, at that, choosing those that provide the fastest transportation and reception of the patient.
"Studies of these processes in other countries have shown that the general hospitalization of everyone who reveals at least some symptoms leads to unnecessary overcrowding of hospitals. There is simply nowhere to place people and at the same time the mortality of coronavirus increases. Therefore, the question arises how to distribute the sick, with due account for limited resources, and stays a key one for the health system in a pandemic. Our development will allow for removing from the dispatcher a heavy burden on the operational analysis of information about the resources availability in the healthcare system at making a decision on the transportation of a sick person, as well as to distribute ambulance crews and hospitals in such a way as to reduce the number of situations when cars stand for several hours waiting for their turn," explains Nikolay Teslya
Now scientists are setting special research task of implementing their computer model into the system of hospitalization of patients with coronavirus. By the end of the year, they plan to increase the computer model accuracy by attracting additional data from medical institutions in St. Petersburg. The project is supported by a Grant of the Russian Foundation for Basic Research in the field of The Pandemic Combating.