The Centers for Medicare & Medicaid Services (CMS) is launching an artificial intelligence (AI)-based competition, calling for innovative solutions that can predict unplanned hospital admissions and other adverse events.
“The CMS Artificial Intelligence Health Outcomes Challenge will unleash innovative solutions as CMS continues to move the healthcare system towards value,” agency officials said in a March 27 announcement.
Specifically, the challenge aims to develop artificial intelligence-driven predictions that healthcare providers and clinicians participating in CMS Innovation Center models could use to reduce the burden to perform quality improvement activities and make quality measures more impactful, officials noted.
The competition has three stages, starting with the launch stage in which participants will submit an application via ai.cms.gov and provide information about their proposed solution. Up to 20 participants will be selected to advance to Stage 1. CMS and its partnering organizations—the American Academy of Family Physicians and the Laura and John Arnold Foundation— will then award up to $1.65 million in total to selected participants in Stage 1 and Stage 2.
If selected for Stage 1, participants will develop algorithms that predict health outcomes from Medicare fee-for-service data, and strategies and methodologies to explain the artificial intelligence-driven predictions to frontline clinicians and physicians while building trust in the data. Participants in Stages 1 and 2 of the competition will use Medicare claims data sets provided by CMS to develop their algorithms and solutions, according to CMS officials.
“The Artificial Intelligence Health Outcomes Challenge is an opportunity for innovators to demonstrate how artificial intelligence tools—such as deep learning and neural networks—can be used to predict unplanned hospital and skilled nursing facility admissions and adverse events,” CMS Administrator Seema Verma said in a statement. “For artificial intelligence to be successful in healthcare, it must not only enhance the predictive ability of illnesses and diseases, but also enable providers to focus more time with patients. The power of artificial intelligence will truly be unleashed when providers understand and trust the data and predictions.”
The deadline for submitting applications for the Launch Stage is June 18.
Similarly, the Agency for Healthcare Research and Quality (AHRQ) has also announced a competition to explore how predictive analytics may be applied to existing databases to forecast future trends in healthcare utilization and spending.
For this challenge, applicants will be asked to use predictive analytics and related methods to estimate hospital inpatient utilization for selected counties in the U.S. for 2017. They must also provide the predicted values of the number of hospital inpatient discharges and the average length of stay for selected U.S. counties in 2016 by applying the model, methods and analytic approach used to obtain the 2017 estimates, according to AHRQ officials.
In the end, challenge judges will require a brief report describing the model, methods, analytic approach and rationale for the estimates so that they may try to replicate the results, they added. The winners will be awarded a total prize pool of $225,000.
Applicants will receive access to customized analytic files that includes information on hospital inpatient discharges for years 2011 to 2016; they will be judged on two core areas—reliability and validity.
The deadline to apply for AHRQ’s challenge is June 28.
“Healthcare decision makers need better access to estimates that define current and future healthcare issues,” said Gopal Khanna, director of AHRQ. “AHRQ’s challenge competition will help demonstrate how predictive analytics can use existing data to provide those kinds of estimates and make new resources available for real-time decisions about policies and use of healthcare resources.”
