Value-based care will catalyze providers’ tech expectations
Organizations will look to achieve actual ROI from artificial intelligence, but will need consultative help from solution providers such as ClosedLoop.ai
This article is part of a CEO leadership series
The shift to value-based care will have significant technology implications for healthcare organizations during the 2020s, including the need to use tools such as artificial intelligence to optimize the effectiveness and efficiency of care.
While the shift to these types of contracts have moved slowly, it’s expected that reimbursement for Medicare and Medicaid services is expected to move almost completely to value-based contracts by 2030, and that will increasingly put pressure on providers to use more technological prowess to improve efficiency. Value-based care will shift more risk to providers for caring for people assigned to them.
The growing risk means providers will need to better predict which patients will need pre-emptive care and then intervene effectively, says Andrew Eye, CEO of ClosedLoop.ai, a developer of services that enable providers to apply artificial intelligence to care delivery. KLAS Research has recognized the company as a Best in KLAS as a top vendor of those providing AI services.
“You’ve got to spot adverse events before they happen, and you have to have interventions that are effective in avoiding those adverse events – that’s how you bend the cost curve,” Eye notes. He expects more movement toward preventive care, “and that means I need to precision target those limited interventions.”
AI is just good math to predict a future event - identify need for intervention in advance.
Shifting technology focus
Research indicates that healthcare organizations have three top areas of focus to prepare for this shift in care reimbursement, particularly in coming out of the COVID-19 pandemic – these are telehealth and virtual care, electronic health records systems, and artificial intelligence and analytics, Eye says.
But even though AI has been widely anticipated to play a role in healthcare, many organizations feel ill-equipped to use such advanced computing. That’s where ClosedLoop can provide not only its platform but its expertise in applying AI to healthcare in ways that can achieve demonstrable results, he contends.
“We like to say that if we predict the future, you can change it. At the end of the day, artificial intelligence is just good math. Our job is to predict a future event based on really good math. But at the end of the day, that doesn’t matter unless we help our customers actually intervene and prevent that avoidable hospitalization or stop that fall and the related injury, or that opioid prescription for someone who might become addicted.”
After a lot of initial hype, there has been spotty deployment of AI in healthcare, and vendor promises have often involved “a lot of smoke and mirrors,” says Ryan Pretnik, director of analytics research at KLAS Research. “It created a sense of nervousness when (providers) would engage with an AI provider or solution provider.”
Where to start with AI - make your current success pathways more intelligent.
ClosedLoop seeks to dissipate those concerns by working closely with providers, Eye notes. “We can provide a custom predictive model for any outcome (a provider) is interested in. The reason we’re able to do that is through technology differentiation – not focusing on building algorithms, but rather building a machine that builds algorithms, which allowed us to quickly customize things.”
The company can provide consultation to help determine the best potential for achieving rapid and actionable results from its analysis.
“If you have 10 use cases that a provider thinks they want to go towards and apply AI to help drive an outcome, the reality is that probably only a half of those can be used for AI,” Eye explains. “If we apply a model or prediction here, will it drive an outcome? We want to make sure that we’re aligned with outcomes that users really want, and then allowing it to use a tool to get a quick value.”
Eye says ClosedLoop can support organizations that already have data scientists on board, or can supply those capabilities to organizations that need them.
The ClosedLoop model enables quicker turnarounds for organizations that want to build predictive models. “That ability to have a low-risk, really fast turnaround avoids instances where we’re not six months into a project only to find out there’s not enough signal in our data to predict the event of interest,” Eye contends. “For that same reason, we can do that first project in days or weeks.”
Applying AI requires a strategic partnership. ClosedLoop scores high as a Best in KLAS solution partner
Demonstrating capabilities
ClosedLoop anticipated rapid growth after receiving investor funding in January 2020, but had to shift gears as the healthcare industry was disrupted by the COVID-19 pandemic. To demonstrate its capabilities, it developed an open-source COVID Vulnerability Test, a predictive model based on proxy endpoints, Eye says.
Built in partnership with several chief medical officers across the country, the model was distributed for free, he said. “There are over 10 million people affected by that model that we know of, not to mention all of the open-source adoption that we couldn’t track. We were just getting scrappy and trying to help.”
View listing of articles in this CEO leadership series