Saving lives and money with machine learning
By Tom Lawry, Director, Worldwide Health, Microsoft Corporation on September 28, 2016
Filed under Health
My last blog touched on use of advanced analytics to improve Clinical Effectiveness in ICUs across Brazil. In this blog I want to highlight how cloud-based analytics and Machine Learning (predictive analytics) are being used in healthcare to improve Operational Effectiveness.
Based in Singapore, Fullerton Health is a leading provider of corporate healthcare solutions across Asia Pacific. With 200 facilities in five countries they serve more than 9 million lives by offering a scalable delivery system that focuses on high quality care that is affordable. This model is helping to drive increased coverage and accessibility in the markets served by Fullerton Health.
Clinical and operational leaders at Fullerton Health view advanced analytics as an investment that empowers them to save more lives by automating processes, improving care and actually reducing operating costs.
Fullerton Health recently teamed up with Seattle-based partner KenSci to deploy agile predictive analytics across the care and cost continuum. KenSci is a Risk Prediction platform for Healthcare based on Microsoft technologies. The initial focus at Fullerton was to utilize predictive analytics to better identify and manage fraudulent claims which cost resources that otherwise would be going into health and wellness services. Medical Fraud has become increasing sophisticated and the sheer volume of claims and complexity of fraud tactics makes it difficult for auditors to effectively identify and prevent fraud.
Utilizing Machine Learning and Advanced Analytics they immediately identified over a million dollars in fraudulent/questionable claims in the data sets routinely reviewed and audited by over 20 Claims specialists.
Another early implementation involved helping a corporate client make better use of the funds it spent to care for 3,000 employees. In this situation they discovered that ten percent of the covered lives were consuming 70% of the resources due to chronic conditions and other issues with care management.
Once they the problem was defined and quantified, prescriptive analytics helped to create a plan to tailor processes for this employer to include better care pathways resulting in improved care for employees with chronic conditions. At the same time, they decreased absenteeism while reducing costs for managing the health of this population by 60%.
Upcoming initiatives at Fullerton Health include applying machine learning to improve its Evidence-based medicine initiative to improve clinical practices across its network of clinics and providers.
For a first-hand look at the impact Advanced Analytics is having at Fullerton Health across five countries check out the short video below.
KenSci footnote: In addition to utilizing Microsoft-based data and analytic solutions their growing success is due to its collaborative model that includes practicing physicians, data scientists, computing and biomedical researchers, and software developers. Spun out of University of Washington after 4+ years of industry-academic research, KenSci is building one of the first vertically integrated predictive analytics platforms for Healthcare. KenSci was recently chosen as a part of Microsoft’s Seattle Accelerator program for Machine Learning and Data Science startups.