Evidence-based medicine is the most modern scientific approach to treatment. It combines a doctor’s individual clinical experience with the data from top medical research. An important part of this approach is making sure that medical practitioners are supplied the best research data in a structured format.
The process of distributing knowledge acquired through clinical research is complicated. For one, not everyone can access such publications. This is where IT technologies come to the rescue. This task can be simplified through the use of registry data collection (data about the results of treatments is added to a database and is later analyzed). This information is subject to a number of limitations in regards to its conclusiveness, while data collection is not controlled at all. With the increase in number of open state databases, nationwide analysis has become possible. For instance, the State Vital Records Office has been publishing openly accessible mortality data.
Alexei Yakovlev
Alexei Yakovlev spoke about the importance of data analysis at all stages: during the initial visit, when delivering the patient to a medical facility, in the operating room, in intensive care and after the patient’s release. He especially emphasized the importance of a digital clinical record. Various data analysis solutions have been applied at all of these stages and proved themselves effective.
One of the Almazov Center’s first projects in collaboration with ITMO University focused on analysis of data concerning patients with acute coronary syndrome and all patients admitted to the emergency room. Depending on the month of admittance, the patients’ transaminase levels, which tend to be higher among alcohol users, were analyzed. Turns out, with a delay of one to two weeks, the liver enzyme levels peak among all intensive care patients in the city center, and, accordingly, the longer the holidays are, the more likely this is to happen. The biggest peak occurs in late January, when the levels are 6 to 8 times higher than the norm. Other peaks occur in May and November holiday seasons.
At the Almazov Medical Research Center
Several years ago, a collaboration between Almazov Center, ITMO University and Yandex took place. The latter provided servers, which, over the course of a week, analyzed the time of patients’ arrival from 250 areas of the city to various medical facilities while accounting for traffic status in the city. Several conclusions were drawn: traffic has a significant effect on the transportation time, and that the minimum transportation time is only achieved in one third of all cases. If patients are delivered to facilities closest to their addresses, then the patient load among all hospitals would be unbalanced. For most of the addresses, alternative routes to other facilities were found that would take approximately the same amount of time. The system can accommodate unexpected complications: if, say, equipment goes out of order at a hospital, it will reroute patients to a nearby facility without significant effect on the time of transportation.
Using modern geoinformational technologies, a system can be developed that would minimize patient transportation time and can be used for decision support by emergency services operators, doctors and paramedics. It would display transportation time, taking into account the traffic situation, the nearest hospitals and their types.
Information systems
Information systems that exist today can track patients’ movements between wards, medical expenses and issued invoices (in other words, can handle administrative tasks), yet are not very user-friendly for doctors, for whom they are intended.
Comprehensive systems also exist that help set up a patient’s treatment in a way that would make the best use of all the opportunities for data collection and to minimize the time it takes to document a patient’s status and conduct routine procedures. The information from various devices is cycled through a network; connections between several networks occur at workstations, where their data is combined.
