Proactively identifying such rising risk patients is important to mitigate future health related costs.
It is well documented that a small percentage of patients account for most of the healthcare expenditures in the United States; the costliest 10% of patients account for more than two-thirds of the healthcare expenditures in the United States (1). What is less well known is that, while some high-cost patients have consecutively high-cost years, the majority experienced what has been termed a ‘cost bloom’—a dramatic year-over-year cost increase that moved them from lower expenditure deciles into the upper decile of spending. These previously low-cost patients account for roughly 68% of all high-cost patients annually (2).
Proactively identifying such rising risk patients is especially important, as they may disproportionately benefit from interventions designed to mitigate future costs. Such efforts can be an effective way to simultaneously improve quality and reduce costs, and are distinct from programs that address the needs of existing high-cost patients. Unfortunately, accurately identifying these rising risk patients can prove challenging for healthcare organizations (HCOs). Nearly 50% of cost bloomers are likely to have no inpatient hospital costs, and when compared to persistently high-cost patients, are diagnosed with four times fewer chronic conditions (2).
In addition, and to the extent that standard prediction tools and reimbursement formulas have failed to accurately predict or account for ‘cost bloomers,’ HCOs pay the price. Such models leave HCOs that are held accountable for the total cost of care vulnerable to inadequate funding, financial penalties, and unfair performance assessments, which can also ultimately affect the health outcomes for patients (2).
As HCOs invest to improve patient outcomes and avoid unnecessary costs, it is paramount they have the ability to anticipate and address the needs of rising risk patients. AI can help by accurately identifying patients likely to be cost bloomers. Using AI-based models can also avoid the drawbacks of traditional risk scoring by allowing providers to easily integrate a wide range of distinct and dynamic patient data that would otherwise have been excluded. By leveraging AI, HCOs can launch proactive interventions designed to address the needs of rising risk patients in ways that drive a disproportionate benefit in quality, costs and future health outcomes.