Improving patient outcomes through medical data analytics

Megan D. Williams

It’s estimated that by 2020, healthcare will operate in a world of 44 zettabytes of data (a zettabyte is comparable to the number of grains of sand on all the world’s beaches). We’re currently creating that data at a rate of 2.5 quadrillion bytes daily.

Data and medical data analytics can drive ongoing transformation across the industry if they’re used to inform business and clinical decisions and support teams, such as anesthesia. This begins with addressing the “four Vs” in healthcare data analytics: volume, velocity, variety and veracity.

In his talk “Improving Patient Outcomes With Data Analytics,” now available for CME credit through AudioDigest, Dr. Christopher A. Troianos, professor and chair of the Anesthesiology Institute of the Cleveland Clinic and Lerner College of Medicine of Case Western Reserve University, explains how the industry has progressed in its use of big data, offering insight into what this means for healthcare as a whole and for physicianship itself.

Listen to a sample of Dr. Troianos’s lecture on AudioDigest and subscribe to learn more. ⟶

Solving complex problems with data

Healthcare challenges are complicated. They include a complex reimbursement environment, multiple intersecting markets, a shifting payer mix, workforce challenges and how to use skill sets to efficiently and effectively deliver care.

Additionally, demographic changes (for instance, baby boomers aging into high-consumption phases of life) continue to drive up the cost of care. For example, between 2014 and 2024, there will be a drop in commercial payers and a shift to government reimbursement, a dynamic that loses Cleveland Clinic 1% of its revenues each year, Dr. Troianos explains.

Arriving at a solution for these challenges starts with reviewing the legislation and efforts currently taking place in healthcare data. This includes an ongoing transformation from a fee-for-service approach that has involved minimal to no data collection or outcome reporting to a point where there is risk involved if positive outcomes aren’t achieved.

Meanwhile, the Institute for Healthcare Improvement is attempting to strengthen healthcare in general through the “triple aim” approach, which involves “improving the health of populations, enhancing the experience of care for individuals, and reducing the per capita cost of health care.” This timeline of recent developments highlights how the industry has used medical data to help achieve these aims:

• 2009: The Health Information Technology for Economic and Clinical Health Act attempted to increase the use of economic and clinical data. This went a step beyond EHR adoption and was intended to inform doctors and administrators in how they could improve care.
• 2010: 2010 saw the passage of the Affordable Care Act (ACA), which included legislation on collecting and using data. One of the major challenges data is supposed to address is the cost of healthcare. The ACA includes the Meaningful Use program (we are technically still in stage 3), which rewards providers if they produce and use data—and eventually penalizes them if they do not.
• 2015: The Medicare Access and CHIP Reauthorization Act of 2015 made it easier to collect data and support outcome-based reimbursement, starting the industry shift from a focus on volumes to value.
• 2017: While the ACA was designed to reduce cost of care, costs continued to rise, so the American Healthcare Act, or AHCA, was passed.

What data can do

Data gives insight into how a hospital is performing, how much staff is spending and how patients are doing. It’s the key to reducing medical and operating costs and shows how clinical resources are used so that those resources can be deployed more efficiently. Healthcare professionals can use a variety of data when working to improve patient outcomes, including claims data, clinical data, pharma data, patient behavior and patient perception data.

Those working in healthcare still haven’t taken full advantage of what this data can do, but others have. Players like Amazon, JP Morgan, Berkshire Hathaway and CVS have transformed how they deliver products and services in the industry by using this data. Healthcare veterans will need to do the same and rethink how care is delivered.

New pressures on providers

Expect consolidation across the industry to become a norm. We’re already seeing this with hospitals merging with insurance companies and other new approaches to provider relationships. For example, Cleveland Clinic has partnered with employers to provide cardiac services. Their work with Boeing in flying patients to Ohio for service has pushed other local Seattle providers to change how they provide care. On the physician front, the role of the provider is shifting. The National Health Service in England is testing AI to answer medical questions for patients, removing providers from the equation entirely. Some providers now use medical imaging analytics to interpret MRIs, taking the practitioner out of interpretation work.

In pharma, digital pills approved by FDA have been used to determine whether patients are taking their drugs. This has positive implications for anesthesia, where in outpatient settings, it’s impossible to reliably know if a patient has actually been taking vasoactive drugs. Similar technology is being used to automate scheduling and even improve pain management. One of the most promising applications involves predicting patient needs. Cleveland Clinic is using medical data analytics to better understand what’s driving patients and predict healthcare resource consumption patterns. This information also allows those who use it to decrease the future costs of taking care of patients by focusing on preventable admissions and conditions that predict readmission.

Dr. Troianos offers an interesting example of this. He discusses a case in which files were pulled from a gene bank on a cardiovascular cohort. The information in the files was then matched with EPIC-derived phenotypic data to identify patients who were at risk of colon cancer, ultimately optimizing the use of testing resources.

Improving patient outcomes through medical data analytics still mostly lies in the future, but we are already seeing results in U.S. healthcare. The industry has experienced an increase of $70 to $100 billion in value as a result of business decisions and data mining, with an estimated additional revenue of $300 to $450 billion in the future.

While every issue in healthcare won’t be solved by data, there is a growing opportunity to reposition clinical acumen as a partner to analytics. This strategy can be used to find solutions to healthcare dilemmas as margins continue to tighten.