Austin, Texas-based MAP Health Management, which works with a variety of healthcare stakeholders to offer a remote patient monitoring and engagement platform focused on addiction treatment, is bringing cognitive computing into their toolset through a new partnership with IBM Watson Health.
Over the six years since MAP Health was founded, the organization has amassed numerous partnerships representing multiple points in the treatment of substance abuse disorder. Their large network includes treatment centers, providers, payers, clinics and technology companies, which altogether work to offer analytics geared at improving clinical and financial outcomes for chronic behavioral health illnesses. At this point, MAP Health Management CEO Jacob Levenson said, it was time to bring in cognitive computing.
“There is an immediate value of unstructured data that we are otherwise not able to fully realize,” Levenson told MobiHealthNews. “For example, we have millions and millions of words from case notes which have what I personally think is some of the best data available. A lot of our data comes in binary, black and white forms, and Watson will enable us to turn the lights on and understand the value of it.”
MAP Health primarily seeks to fill gaps in treatment programs that can arise from a number of factors, such as a lack of standards in data collection and interoperability, or insufficient follow-up support after treatment. The company has embarked on several initatives as of late to improve their population health management abilities. Last month, they teamed up with Lief Therapeutics to pilot the use of their stress-monitoring patch in people with substance abuse disorder. In February, the company announced a partnership with Soberlink, makers of a high-tech, FDA-cleared remote Breathalyzer, and just earlier this month teamed up with relapse reduction app WeConnect.
While the Watson integration will be global for MAP Health, one of the first partners to deploy the cognitive computing abilities will be Aetna Behavioral Health, which will use the tool to help predict substance abuse relapses among its members. MAP Health and Aetna are working with addiction treatment providers to collect and analyze the patient data with the goal of develop treatment protocols and long-term strategies to support each patient in the way that works best for them.
“IBM Watson Health and MAP have the potential to positively impact the tens of millions of people and families suffering from addiction in the United States,” IBM Watson Health VP of Partnerships and Solutions, Kathy McGroddy-Goetz, said in a statement. “MAP Health Management is widely recognized as having a robust addiction outcomes database. IBM’s Watson cognitive computing technology is a natural fit to further empower what MAP is doing to help improve qualitative and quantitative outcomes in the behavioral health and addiction treatment fields.”
With Watson, Levenson said they saw a way to advance their mission of laying down the tracks for data to move around between stakeholders on their network, and on a much bigger scale than MAP Health has previously been able to.
“This is an extremely fragmented space, even though the front page of most news these days includes at least one opioid addiction story,” said Levenson. “I’ve never seen a big brand like IBM move into this space, which shows that there is a recognition that this kind of population health management platform is starting to lead people to expect positive results.”
The jury may still be out on whether IBM’s cognitive computing can deliver on its promise to transform healthcare – Social Capital CEO Chamath Palipitiya recently called IBM Watson “a joke” – but that doesn’t seem to be hindering the company’s ability to take up new partners. The research arm of IBM is working with Sutter Health to develop methods to predict heart failure based on under-utilized EHR data.
In order to predict heart failure, doctors currently document signs and symptoms in the EHR for the purpose of ordering diagnostic tests that could ostensibly give them enough warning to prevent an acute attack. However, most patients are admitted to the hospital following exactly that, and often at the point where there is irreversible damage to their organs. So IBM and Sutter Health spent three years employing the latest artificial intelligence capabilities such as IBM’s natural language processing, machine learning and big data analytics tools to train computer models to help them more accurately predict heart failure. In an article published in the American Heart Association’s journal Circulation, the researchers outline how they were able to learn about the types of data needed to train models and potential new application methods.
