Healthcare has had higher barriers to adopting data science than other industries. And while state-of-the-art analytics solutions are already available, few of them are actually in use by clinicians.
That’s something that the industry could start to see change in 2020, as a culture of data-driven decision-making among clinicians begins to mature, and the quality improvement process improves.
Data science will increasingly guide clinicians in finding opportunities for improvement, designing and implementing interventions, and evaluating impacts.
Jason Cooper, chief analytics officer at HMS, said AI-driven prescriptive analytics and other advanced analytical techniques can process what can easily be 1,000 or more predictors from claims, self-reported consumer data, electronic health records, census, and social determinant data.
“All that data can provide the right insights to proactively engage members and offer care management services preemptively,” he said.
In 2020, he said a greater emphasis will be placed on shifting care management’s current focus off of critical, high-risk patients because it’s just not producing the desired outcomes.
Instead, health systems need to work towards preventing patients from ever getting to the critical state by shifting their focus to well care versus just sick care.
“By leveraging predictive and prescriptive analytics, care management can proactively engage patients and coordinate the right care intervention at the right time to address hidden and rising risk, arresting risk progression and getting ahead of issues before they become less manageable,” Cooper said.
He pointed to a recent report detailing how the U.S. is now showing a steady decrease in life expectancy — and an increase in mortality rates among relatively young individuals aged 25 to 64.
Cooper said drug overdoses, suicides and chronic conditions are the underlying factors of this phenomenon.
“Though our healthcare system has profound and systemic shortcomings, we clearly need to better understand and address behavioral and social health across populations,” he said. “Moreover, there is a rising need to apply predictive and prescriptive analytics across healthcare organizations of all sizes.”
Fortunately, prescriptive analytics are no longer limited to large organizations with deep pockets and an army of PhD-level data scientists on-staff.
“Oftentimes, smaller organizations won’t seem to have the budgets allocated for predictive and prescriptive analytics, but once the cost savings is demonstrated, the ROI speaks for itself,” Cooper explained.
He said one of the big keys to success in value-based care is getting the treatment plan right in the first place since failures will now be paid for at the healthcare organization’s expense.
“Organizations must know what they can do to avoid these costly issues by leveraging prescriptive analytics,” he said. “Robust analytics can help organizations achieve triple aim by improving healthcare outcomes, reducing cost and improving quality. All in all, this can improve medical loss ratio for risk-taking entities.”
Darren Schulte, CEO of AI analytics company Apixio said he sees one of the top prescriptive analytics trends of 2020 including the ability to predict the optimal care site, such as a home or a post-acute care facility, for a patient at a given time in their care trajectory.
Two others he’s optimistic about are the ability to predict the best care program or intervention method–such as telehealth versus a clinic appointment versus a home visit–that is best suited for a patient given their current health needs, and the ability to predict the most efficient resource allocation for hospitals to make smarter staffing decisions.
“The use of cloud-based prescriptive analytics that impact few existing clinical workflows requires little staff training time and can increase efficiency and operating margins,” Schulte said. “These types of analytics are more likely to be adopted and used by smaller health systems.”
He also noted prescriptive analytics could help move plans and providers even further toward a value-based care model by providing estimates on patient care alternatives, such as post-acute care options, that result in better outcomes and lower costs.
“Prescriptive analytics could also be of great help offering predictions on utilization and cost variance year-over-year and modifiable cost opportunities,” he explained.
Nathan Eddy is a healthcare and technology freelancer based in Berlin.
Email the writer: [email protected]
Source: Read Full Article