“Patients are complicated. Health care providers are complicated. Driving a car is complicated. Deaths from driver errors are epidemic and may be influenced by the complexity of decision-making,” says Scott Weingarten, Senior Vice President and Chief Clinical Transformation Officer for Cedars-Sinai Health System.
But it doesn’t have to be that way.
Accident avoidance systems in self-driving cars, which work through decision support, could reduce fatal car crashes by 90% by the middle of the 21st century. Decision support processes a large number of inputs from sensors, such as traffic signals and other cars, and provides real-time guidance to the vehicle.
A myriad of studies show why we also need decision support in medicine. Weingarten cites the Johns Hopkins study that showed medical errors are the third leading cause of death in the U.S; the JAMA study that showed women physicians have significantly lower patient mortality rates than men physicians, potentially because they’re more likely to practice evidence-based medicine; and studies that physician decisions can influence as much as 80–90% of health care costs. He also cites a JAMA study that found one-third of health care expenditures are waste and 10% of health care expenditures are overtreatment; studies from the Rand Corporation that show patient care is consistent with evidence only about half the time; and studies that there could be a 17-year lag between the discovery of life-saving treatments and when they are uniformly adopted into practice.
“This is very hard to believe. Physicians are highly trained, intelligent people who care deeply about their patients. Why could this be?” Weingarten asks. It’s because of information overload. “The volume of information potentially available to help each patient’s care, along with relevant or potentially relevant scientific evidence, overwhelms our capacity to process all of this information in real time.”
The information technology to enable decision support already exists in the form of electronic health records. “They’re there today, they have the capability, and patients can and should benefit from this investment,” he says. “So if all of the patients’ information is available, can we process it and provide the best diagnostic tests and treatments to providers and patients? This is really precision medicine.”
The patient of tomorrow will have their omic, microbiome, social determinant, symptom, imaging, and laboratory information, along with personal preferences and level of engagement, enabling the best treatment and the optimal approach to that treatment. It’s not science fiction, Weingarten emphasizes. In a 21st-century medical accident avoidance system, we could have better quality of care, better value, lower costs — and more time spent talking to patients — enabling us to harness information that could save lives.
“Decision support is transforming the way that we will drive a car. Why shouldn’t it also transform the way that we care for patients or that patients care for themselves?”
From the NEJM Catalyst event The Future of Care Delivery: Relentless Redesign at Providence St. Joseph Health, January 19, 2017.