Aug 30, 2017
The U.S. healthcare industry is broken: It’s too expensive, there aren’t enough primary care doctors, and care is reactive instead of proactive. A salve for many of these issues is data analytics. Important answers can come from crunching large data sets: an initiative in Paris, for example, involves four hospitals that are crunching a decade’s worth of hospital admissions records to predict daily and hourly patient demand. That insight will help the hospitals more accurately match staff and resources to demand, eliminating waste and, hopefully, improving care.
Analyzing “small data” is also a powerful idea, as panelists at the 2016 emids Healthcare Summit discussed. Reviewing a patient’s non-clinical data, such as activities, diet, income, shopping and even voting habits can illuminate signals for risk. “It’s not just treating Bob, who has diabetes and is 72 years old,” said Brian Garcia, Summit panelist and chief product and technology officer for wellness technology provider Welltok. “Bob’s problem is that he has high anxiety, he eats poorly and he doesn’t get enough exercise. If we can start looking at those pre-identifiers, then you can start reducing the overall cost of treating Bob and his diabetes.”
Welltok has conducted research on consumers to predict healthcare outcomes, such as discerning the likelihood that a patient will be readmitted to the ER based on income levels, geographic information and other nontraditional data.
What else can data intelligence do for patients? For one, it can save time and reduce anxiety from dealing with an inefficient healthcare system. Older patients tend to visit the doctor lot, which can entail several hours getting to the appointment, sitting in the waiting room, and then getting back home. This can be a miserable quality of life, when repeated several times monthly. A real-time scheduling analytics system could instead predict wait times for the day and alert patients to show up 30 minutes later than their appointment.
Tapping the Database of Intentions
Then there’s the time involved researching a condition and finding a quality provider whom is covered by your insurance. “Imagine a world where my health plan has uploaded all of their emails to the Google system, with no identifying information, and I search for pain in the left side of my head and a symptoms box would come up and say here’s what it is and, by the way, there is an urgent care center that takes my medical insurance 0.3 miles away from me,” said Addie Braun, Google’s head of health insurance industry, during the Summit panel.
Crunching multiple data sets is also imperative for uncovering and preventing waste. A health system could review admissions and discharge data to analyze the top reasons why people visit the ER. In a neighborhood where there are high percentages of visits for patients with colds or other non-emergent conditions, further research could indicate a need for more urgent care or primary care centers. Healthcare leaders could also introduce community education programs on cold and flu prevention and treatment and which facilities to visit for which conditions.
Deploying the best IT solutions for collecting, storing, and analyzing data within a healthcare system or payer organization is a work in progress. There many technologies on the market, but the effective ones will be able to quickly integrate both structured and unstructured data from many sources to create a rich, searchable repository of patient and operational data. Reporting and sharing that data must be handled carefully, given concerns around data quality and privacy. Finally, integrating data insights into the workflow of physicians and healthcare executives must be seamless, Summit panels agreed. Despite the challenges, we are entering an age in which useful, everyday analytics will positively affect healthcare organizations and patients everywhere.
Want to learn more about how data analytics and integration can disrupt healthcare for the better? Download the charter from the 2016 emids Healthcare Summit.