Predictive Analytics: A New Way Forward for Healthcare

Innovation in the private sector has unleashed forces that have fundamentally transformed daily life in the twenty-first century. The world’s largest transportation service, Uber, doesn’t own a single car. Airbnb, which offers hotel services globally, doesn’t own a single hotel. Even one of the biggest retailers, Amazon, doesn’t own a single store. Each company has taken advantage of a growing trend of consumerization by personalizing services through new web and mobile technologies.

Consumerization is also playing an increasingly important role in healthcare as legislation changes the landscape for patient care and creative solutions to healthcare’s most pressing problems are being developed by entrepreneurs. The HITECH (Health Information Technology for Economic and Clinical Health) Act, signed in 2009, invests over $35 billion in improving the healthcare sector’s information technology and interoperability through the use of electronic health records (EHRs) across a number of systems. The Patient Protection and Affordable Care Act, signed in 2010, led to the creation of Accountable Care Organizations (ACOs), which have financial incentives for ensuring patients — particularly those with chronic illnesses — are healthy and that overall costs for care are kept low. Novel ideas are also on the rise and materializing in digital health startups and similar healthcare ventures.

Determining the most effective method of treatment, recommending the most successful prescriptions, and preventing the hospitalization and readmission of patients are just a few decisions that predictive analytics has the capacity to fundamentally change.

Yet, for too long, a variable has been absent in the equation of game-changing progress in healthcare. While Uber, Airbnb and Amazon each utilizes a probabilistic approach toward providing services for their consumers, healthcare does not — at least not entirely yet. Predictive analytics can change that. It offers a great deal of potential in bridging the gap between what we know and what is possible in healthcare.

Predictive analytics, the process of learning from aggregate historical medical data in order to make effective predictions about the future that translate to improved decision making, opens the door to novel insights for doctors, patients, and healthcare professionals. A hybrid approach that combines the expertise of clinicians with advancements in computer science and data collection will provide the tools necessary to solve some of healthcare’s hardest problems. Determining the most effective method of treatment, recommending the most successful prescriptions, and preventing the hospitalization and readmission of patients are just a few decisions that predictive analytics has the capacity to fundamentally change.

Source: “The Future of Personalized Healthcare: Predictive Analytics”, Rock Health report.

The disparity in patterns isolated from a multitude of sources forms a complexity that is difficult for computers to identify and interpret. Aggregate data collection used for analysis is enormous and requires a human element to discern patterns translatable into actionable insights for the healthcare sector. The culmination of human effort in detecting patterns across large swaths of data will help in the fulfillment of what the Institute for Healthcare Improvement describes as the “Triple Aim Initiative.”

Above: An illustration detailing the optimizing healthcare by the IHI.

The Triple Aim Initiative provides the conceptual framework for building an optimal healthcare system through a multidimensional approach focused on: 1) Improving the quality and satisfaction of care for individual patients, 2) Improving the collective health of populations, and 3) Lowering the per capita cost of healthcare. These three central tenets provide the structure for leading systemic changes in healthcare. Predictive analytics, through a probabilistic approach of problem-solving, is already shaping to have an immense impact towards the fulfillment of these initiatives by improving care-coordination, triaging, and chronic disease management.

Improved Individual Care

Artificial intelligence powered predictive analytics startups, like Lumiata, have developed innovative methods for triaging patients in order to increase the overall efficiency of healthcare. The challenge of determining the distribution of healthcare resources to deliver successful care to as many people as possible is one that has lingered since the inception of the healthcare system. Technology like Lumiata’s “Risk Matrix” brings the chance of potentially solving this problem.

Located in the Silicon Valley, Lumiata has collected over 170 million data points and 25,000 hours of physician expertise to help fuel their analytics platform. The company leverages data science to provide payers and providers evidence of the health trajectory of patients based on medical studies, lab results, and other key information points. This information is used to provide actionable insights to accurately predict the course of a patient’s health with a comprehensive rationale for each prediction. In an interview with Oliver Wyman’s Health & Life Sciences division, Ash Damale — the Founder & CEO of Lumiata — describes Lumiata as facilitating “truly personalized care at scale by leveraging data to enable automation that the industry desperately needs. If you boil it down, we try to answer the core questions: what to do for whom, when, and why?” Thus far, Lumiata has partnered with Google, the Independence Blue Cross, between ten to twenty ACOs, people on the Medicare Advantage Plan, and Intellivisit for their healthcare analytical services.

Better Population Health

Improving the collective health of populations is key towards reducing the cost of healthcare. Reducing the impact of chronically ill patients and aging populations, the costliest patients, on the healthcare system is critical in mitigating further increases in costs in the future. As lifespans increase and as chronic illnesses remain exorbitantly expensive to treat, it is imperative for a paradigmatic shift in healthcare towards a value-based model. Focusing on attaining optimal outcomes at the lowest possible cost instead of the status quo’s supply-driven system — which is designed to maximize the volume of care instead of its value — is necessary towards improving the health of our population altogether.

In fact, Lumeris — the largest population health management company — has exclusively focused on developing value-based care in partnerships with healthcare organizations through predictive analytics. Lumeris has distinguished itself by its cloud-based Accountable Delivery System Platform (ADSP) which integrates EMRs, pharmaceutical and lab data, claims, and other essential points of information. As a result, partnered organizations have greater insights that can lead to action that moves healthcare to a value-based system.

Per Capita Cost Reductions

Echo Labs is an example of a company that has used predictive analytics to manage patients with chronic illnesses. The company has developed actionable insights from monitoring key indicators in a patient’s health. In turn, hospitalizations are potentially prevented and costs to our healthcare system are reduced.

Echo Labs has engineered non-invasive blood monitoring and analysis that detects essential physiological indicators like nutrition, hydration, energy levels, and sleep to identify the early stages of trouble in order to enact a timely intervention before a hospitalization or emergency occurs. Their methods of detection using spectrometry and optical sensors lead to a better decision-making calculus that focuses on prevention and timely intervention to optimize the situational needs of patients. By measuring essential vitals towards a patient’s health, Echo Labs is able to analyze patient data on a long-term basis to predict the probability of events where intervention at an early stage can significantly improve care and significantly reduce per capita costs. Both providers and patients stand to gain from the potential of a mutually beneficial partnership aimed at lowering costs and improving care.

Looking Ahead

The enormous potential that predictive analytics has in changing the paradigm of healthcare cannot be overlooked or ignored. As our healthcare system undergoes seismic change due to the Affordable Care Act, the Health Information Technology for Economic, the Clinical Health Act, and continual advancements in the healthcare sector, predictive analytics will take on an increasingly important role in delivering the necessary change to improve quality of care, impact health outcomes and reduce costs.

Follow us on Twitter for the latest insights, updates and announcements

1Baum, Stephanie. “Lumiata Raises $10M to Ramp up Product Development for Population Health Risk Assessment — MedCity News.” MedCity News. N.p., 26 May 2016. Web. 28 Aug. 2016. 2Gandhi, Malay, and Teresa Wang. “The Future of Personalized Healthcare: Predictive Analytics | Rock Health.” Rock Health. Rock Health, n.d. Web. 28 Aug. 2016. 3“HealthIT.gov.” Health IT Legislation. N.p., n.d. Web. 28 Aug. 2016. 4“The IHI Triple Aim.” The IHI Triple Aim. Institute for Healthcare Improvement, n.d. Web. 28 Aug. 2016. 5“Innovation Diffusion: The Lumiata Team on the World’s First Medical Graph & the Promise of AI.” Oliver Wyman, n.d. Web. 28 Aug. 2016. 6Loeb, Steven. “Meet Lumiata, a Provider of Predictive Health Analytics.” VatorNews. N.p., 11 Feb. 2015. Web. 28 Aug. 2016. 7Parikh, Ravi B., Dr., Ziad Obermeyer, Dr., and David Westfall Bates, Dr. “Making Predictive Analytics a Routine Part of Patient Care.” Harvard Business Review. Harvard Business School, 21 Apr. 2016. Web. 28 Aug. 2016. 8Porter, Michael E., and Thomas H. Lee. “The Strategy That Will Fix Health Care.” Harvard Business Review. Harvard Business School, 01 Oct. 2013. Web. 28 Aug. 2016. 9Secretary, HHS Office of the. “Read the Law.” HHS.gov. N.p., n.d. Web. 28 Aug. 2016. 10“What Is Value-Based Care?” Dartmouth-Hitchcock. Dartmouth School of Medicine, n.d. Web. 28 Aug. 2016.


Originally published at elevarco.com on September 21, 2016.