It's often said that all healthcare is local.
But healthcare executives regularly compare their organizations nationally against aspirational competitors or those that rank highly in well-known lists put out by U.S. News & World Report and IBM Watson Health.
Healthcare analytics company Trilliant Health has rolled out an application that layers machine learning over its massive data sources to help health systems more accurately benchmark against market peers.
These benchmarks can guide an organization's strategies for new specialties and services lines, M&A activity and expansion into new markets, Trilliant Health executives said. Leveraging Trilliant Health's technology, healthcare organizations can see which U.S. markets are most similar to each other across multiple index factors and then benchmark against the relevant competitive leaders in the relevant markets, according to the company.
"In the increasingly complex health economy, the need for evidence-based strategies among every stakeholder—from policymakers and health plans to providers and life sciences—has never been more critical. The starting point for evidence-based strategy is an accurate benchmark," Hal Andrews, president and CEO of Trilliant Health, told Fierce Healthcare.
"Healthcare is local, which means that every market has unique attributes that influence the delivery and consumption of healthcare goods and services. Accurate benchmarking at the market level is essential to compete effectively and efficiently," he said.
Historically, healthcare executives have looked at national benchmarks, but these lists can be irrelevant to health systems operating in radically dissimilar environments, resulting in suboptimal decision-making and capital allocation.
"As a society, we have been primed to determine what is 'best' using third-party rankings and abstract measures. Healthcare is no exception. What are the top five restaurants in New York? What are the best colleges in the country? What are the best hospitals in America? Who are the best orthopedic surgeons in Tampa?" Andrews said. "Subsequently, the entire health economy pursues aspirational and arbitrary benchmarks to guide critical decisions."
Looking at well-known lists, U.S. News & World Report ranks one organization as the best U.S. health system while IBM Watson Health has a different ranking. And Forbes will list certain payers as the best health insurance companies even as others are rated highest on NCQA’s Health Plan Report Card, Andrews noted.
“Not every health system can be like Mayo Clinic or Intermountain Healthcare or Geisinger; in fact, none of them can be, for many different reasons," he said.
In contrast, Trilliant Health's machine learning and similarity modeling, called SimilarityIndex, enables healthcare organizations to benchmark accurately by identifying the most similar markets, facilities, providers and patient populations.
The benchmarking tool is based on Trilliant Health’s proprietary all-payer claims database representing more than 320 million Americans, provider directory and proprietary demand forecast methodology. The company's machine learning technology determines the most similar markets, or metropolitan U.S. core-based statistical areas (CBSAs), based on population growth, economic, demographic and consumer data.
Numerous market forces—a declining number of commercially insured consumers, increasing deductibles, new market entrants, increasing healthcare consumerism and now inflation—are creating massive challenges for healthcare organizations.
Companies across the healthcare and life sciences industries face unique threats from significant shifts in patient populations and disease burden to competition from nontraditional market entrants–all at a time when demand for healthcare services has remained relatively flat, said Sanjula Jain, Ph.D., Trilliant Health chief research officer and senior vice president of market strategy.
"While markets can be characteristically similar, they can have vastly different healthcare demand and supply," Jain noted.
"You cannot fix a bad market, so understanding which markets are which is foundational to evidence-based strategy," Andrews said. "All health economy stakeholders should analyze and understand market similarities by applying the same mathematical rigor and algorithmic intelligence used by technology leaders in other industries to determine which market characteristics are most alike."
Trilliant's benchmarking model analyzes five-year projected population growth and excess mortality per capita, data pertaining to estimated household income, debt, home values and discretionary income as well as population-level data on health insurance status, age and sex, generational distribution, race/ethnicity and primary language.
The benchmarking tool also looks at consumer psychographic profiles, use of wearables and prescription use and drug adherence and analyzes a more representative measure of market concentration using inpatient surgery, outpatient surgery and inpatient medical volumes, according to the company.
The model then combines all of the above information to show the markets most similar to a selected CBSA.
Organizations can use the model for internal benchmarking to identify the strongest and weakest markets. "With that understanding, health systems can begin to compare the level of competition—and the opportunities—in each of those markets in order to allocate scarce capital more efficiently," Andrews said.
He added, "Another use case, which is not limited to providers, is to understand the characteristics of their 'high-performing' markets, enabling them to find the most similar markets which, in theory, would also be markets in which to invest."
Trilliant Health works with more than 75 of the top U.S. health systems that encompass more than 1,500 hospitals. The Nashville, Tennessee-based company synthesizes a variety of first-party and third-party data sets to power its data analytics tools. The company's all-payer claims data set combines commercial, Medicare Advantage, traditional Medicare and Medicaid claims, representing about 70 billion medical claims.
The company aims to use its data sets and analytics technology to build on its predictive analytics capabilities.
Last year, Trilliant Health rolled out a new predictive analytics tool that enables strategy teams to see a 10-year view of the healthcare market with insights all the way down to the local market level. The demand forecast analytics platform was designed to help providers allocate resources and understand how to create growth strategies.