The Combination of AI, Data Analytics and Medicaid Could Transform Healthcare Delivery
Unexpected, large-scale partnerships have the potential to spur dynamic change in the healthcare industry.
We need to look no further than the innovative venture announced earlier this year between Amazon.com, Berkshire Hathaway and JPMorgan Chase designed to improve healthcare delivery and lower costs for the U.S-based employees of these corporate giants.
Another unlikely 21st century trio with similar goals in mind is the developing relationship between Artificial Intelligence (AI) solutions, data analytics and Medicaid. While no one could argue the merits of blending these entities, some healthcare industry watchers wonder how cutting-edge technologies and advanced algorithms can be applied to an outdated Medicaid infrastructure ill-equipped to handle some of today’s most complex challenges.
Yet, data scientists are currently building algorithms exploring how AI can be applied to Medicaid. Further, comprehensive population health management platforms, which are already capable of aggregating and analyzing MMIS data, will be able to analyze Medicaid data in new and innovative ways, and discover important patterns as AI technologies evolve over the next few years.
For example, data scientists are testing models that make connections across the healthcare timelines of Medicaid patients. That is, what factors changed between the period when a patient was healthy and when they were diagnosed with a certain condition? How did patients respond to a particular procedure? What kind of treatments were given afterwards? What was the outcome?
Alternatively, by blending AI and data analytics, we’ll soon be able to look at a pool of Medicaid patients (let’s say 100 with similar weight, height, BMI, genetics, etc. diagnosed and treated for heart disease) – and analyze outcomes in dynamic ways.
Let’s say of those 100 patients, 50 showed improvement after treatment, 40 didn’t, and 10 died. By analyzing every bit of data along the timelines of the 50 patients with improved outcomes (including social determinants of health, medication, changes in diet and exercise), AI can find patterns within care protocols that could be applied to the other 40 patients who did not show improvement. AI could potentially turn around the whole treatment paradigm for Medicaid patients with specific health conditions – not based on clinical expertise, but data. And at the same time, lower costs.
Among the most promising developments between AI, data analytics and Medicaid is how new technologies can be used to help state agencies combat the growing opioid abuse crisis.
Top population health management companies that can access all available data from any disparate system, in any setting and in any format, are already a step ahead. That is, rather than relying on incomplete data sets that may rely only on prescription or claims data to discover general usage trends (as some platforms might), comprehensive PHM platforms will be able look at patient data that analyze the whole person in truly unique ways.
For example, AI and data analytics will analyze Medicaid patients who take opioids for spondylitis, and then apply additional algorithms to make connections that account for vast amount of other factors, including previous medical procedures, access to healthcare, high blood pressure, BMI over 30, etc.
As the opioid crisis grows, the combination of AI, data analytics and Medicaid can be a game changer for the healthcare industry. HealthEC, which already assists a number of state agencies with Medicaid initiatives, is developing our own opioid-abuse solution that we look forward to sharing with the marketplace within the next few weeks.
This article was originally published on HealthEC and is republished here with permission.
Download HealthEC’s white paper: Data-Driven: How ACOs and CINs are Using Analytics to Realign Care – Despite heavy investments in health IT, many organizations struggle to navigate the transition from patient data aggregation to analytics-driven action. This paper highlights steps that CINs, physician practices and other interconnected stakeholders can take to implement a data analytics infrastructure that supports ACO and other value-based care initiatives.