Big data, or big data, is a collection of data that cannot be captured, managed, and processed in an affordable timeframe with conventional software tools. It requires the application of new processing models in order to become an information asset with greater decision-making, insight discovery and process optimization capabilities. Industries are now making every effort to improve their ability to process data and, through processing, add value to it. The healthcare industry is no exception, and researchers are actively looking for ways to mine useful clues from the wealth of previous case information. The Wall Street Journal reported in depth in June on the efforts of U.S. healthcare organizations that are working to measure the level of risk in the intensive care unit at any given point in time.
The intensive care unit (ICU) may soon be a less scary place for patients.
Intensive care intensive care units (ICUs) in hospitals are designed to specialize in the treatment of critically ill patients, with varying success rates. Now, researchers are looking at a wealth of critical care intensive care unit data for clues to help patients achieve better outcomes.
Studies have shown that more than 5 million patients (in the United States) enter the ICU every year, with a mortality rate of 10?29 percent. While some patient deaths are inevitable, others are due to avoidable complications caused by equipment and treatment infections. Other complications include blood clots and psychosis from excessive sedation and prolonged immobilization.
In the past, intensive care units have had success in reducing patient risk with more basic techniques such as checklists. Following a step-by-step checklist for patient care can combat pneumonia caused by prolonged ventilator retention or infections caused by catheters and various hoses.
The age of data
Some hospitals are now testing ways to apply big data. Sifting through years of medical records from multiple sources, including data that may never be included in a single analytics method, to find previously unknown correlations that can lead to more trouble spots, and more possible solutions.
"Checklist reconciliation efforts can work on predictable risks, but it's the less predictable contingencies that are our biggest concern." , said Kenneth Sands, chief quality officer and senior vice president at Harvard Medical School's Beth Israel Deaconess Medical Center (BIDMC). The medical center is collaborating with integrated systems scientists from MIT and human factors experts from APTIMA, Inc. to form a project team, "State of Risk," to measure the level of risk in a given intensive care unit at any point in time.
BIDMC has seven intensive care units, including units for surgical and coronary patients. By analyzing data on patients in all of the ICUs between 2012 and 2014, the hospital project team defined a variety of conditions that would increase risk, such as high volume of admissions, an increase in the number of critically ill patients, an increase in the proportion of nurses with less than one year's nursing experience starting in the ICUs, and an excessively high patient-to-nurse ratio in the ICUs.
"We're using data that people haven't previously imagined to predict harm in the ICU." Daniel Talmor, interim chair of anesthesia, critical care, and pain medicine at BIDMC, said, "For example, people don't generally factor in the experience level of nurses as a risk."
The project team also identified 30 hazards such as bleeding, medication errors, cardiac arrest, readmission to the intensive care unit after discharge, falls, and errors in communication during and after the risk state.
The app developed by the team automatically grabs data from EHR-type medically-specific software and allows doctors and nurses to enter additional questions about specific patients and wards. A visual "dashboard" is then used to calculate risk scores, which are displayed in real time on ICU staff monitors and handheld devices.
Dr. Sands said, "Now we can give early warning of patients who are going to be at risk."
Patricia Folcarelli is a nurse and senior director of patient safety at the hospital. Instead of relying solely on a general list that may not apply to every patient, she spoke of how critical care intensive care units (CCIUs) can change staffing levels, delay elective procedures, or move patients from overloaded CCIUs to less burdensome ones. As a result, the team is also developing personalized lists that are "context-specific," focusing on the needs of individual patients.
BIDMC's State of Risk team is a program funded by the Gordon and Betty Moore Foundation. The BIDMC "State of Risk" team is a program funded by the Gordon and Betty Moore Foundation. The foundation has also provided research and development funding to several other teams in similar fields to help hospitals improve outcomes in critical care. Other teams funded include the Johns Hopkins University School of Medicine in Baltimore, which is improving the efficacy of critical care intensive care units by rapidly analyzing and diagnosing current patient data as opposed to historical data. Peter Pronovost, senior vice president of patient safety and quality at Johns Hopkins University School of Medicine, said that the inability to interconnect medical devices in the intensive care unit has left doctors and nurses scrambling to piece together patient data from individual medical devices, which is not only a waste of manpower and resources, but also potentially puts patients at risk.
"We need software that can connect medical record data to devices to predict risks that could endanger patients, recommend appropriate treatments based on the risk, show whether the treatment has been implemented, and then monitor the patient's condition," said Dr. Pronovost.
Making the Connection
Johns Hopkins assigned an expert from the Applied Physics Laboratory to help it design a system called Project Emergence, which captures data from electronic records and bedside sensors and shows whether an individual patient should receive a treatment and how to prevent complications. and how to prevent related complications.
Alan Ravitz is the head engineer on the project. He explains that the project team began by asking the ICU staff what information they needed most and how it should be displayed. Nurse Rhonda Wyskiel suggested a dial-shaped display interface, which became the basis for the ward's "hazard monitors" and caregiver tablet devices. The monitors remind staff when certain care is due, which is especially useful when staff are not on time. The interface has a seven-color dial, with red for imminent danger, yellow for developing problems, and green for completed tasks.
One of the sensors installed in the display continuously marks the angle of the patient's bed. Studies have shown that when beds in intensive care units are kept at a 30-degree angle, it prevents patients from developing pneumonia from the ventilator. Therefore, if the nurse forgets to set the bed back to 30 degrees after changing the angle, that part of the monitor turns red. In addition, other sensors in the monitor include a device that can be hooked up to an IV pole, which measures the distance the patient walks.