Identifying high-risk patients and matching them with the most appropriate interventions is a challenge. Risk-stratification tools can help, but they have limitations.
Predicting financial and adverse outcome risks of individuals or groups involves the systematic use of key information. By leveraging such information, risk-stratification tools can help identify high-risk patients and enable providers to match them with the most appropriate level of intervention. Key data include past utilization of services, number and types of diagnoses, demographics, medications, and factors regarding patients' functional and psychosocial status.
This report provides an overview of common risk-stratification tools, how the health care industry uses them to prevent avoidable hospitalizations and treat illnesses, and their many limitations. Through a literature review, case studies, and other techniques, the authors sought to answer these primary questions:
- Which tools are most commonly used?
- How do these tools differ in their ability to predict future utilization and health outcomes in populations and in individuals?
- In which population does each tool work best?
- What are the limitations of each tool?
- How can these tools be applied most successfully to link patients with appropriate care management interventions?
The authors conclude that existing predictive tools have limitations, and the more variables included in a risk-stratification approach, the better its predictive ability. Because these tools do not address psychosocial issues, obtaining input from clinicians and health assessment surveys will enhance their effectiveness.
The complete report is available under Document Downloads.