- Social Responsibility
- Social Responsibility
- Our Giving
- Corporate Social Responsibility
- Be The First Tobacco-Free Generation
- Community Stories
- Thought Leadership
- Hiring Areas
- Why CVS
- Important Security Alert
- Investor Story
- Results Center
- 2016 In Review
- Financial Information
- SEC Filings
- Events & Presentations
- Stock Information
- Corporate Governance
- Investor Resources
Predicting High Cost Patients to Optimize Health Care Resources
Millions of Americans suffer from chronic conditions that require lifestyle modifications and daily maintenance medications, but only a small percentage actually adopts healthier behaviors and takes medication as prescribed. Unfortunately, this can worsen a patient’s existing condition, resulting in additional (and often preventable) medical interventions including hospitalization.
It’s not surprising that this scenario is a major contributor to the rapid increase in health care spending. But what if we could look ahead and predict who these costly patients will be? This would allow us to direct health care resources to address individual patients’ needs and challenges sooner and could lead to greater cost efficiencies and improved health outcomes.
According to research from the CVS Health Research Institute and our research partners, predictive analytics is bringing us closer to this reality. In a recent study published in Medical Care, the researchers compared predictive cost-modeling strategies to understand which approach can best classify and predict future high health care spenders. Study findings showed that group-based trajectory modeling – a method that uses health insurance claims data to classify patients based on patterns of health spending across several measures including inpatient, pharmacy and outpatient costs – had better predictive accuracy compared to the more conventional approach that defines high-cost health care users as those in the top 5th percentile of overall spending.
The findings also showed that group-based trajectory modeling allowed for greater precision in understanding future potential health care costs. The researchers concluded that understanding the nuances in patients’ spending patterns helps to more accurately predict long-term costs and pinpoint who could most benefit from certain medical and pharmacy interventions.
At CVS Health, we are using predictive analytics to help shape patient care for all those we serve and are working to develop innovative strategies that tackle rising health care costs.
“This research is an important step in helping to better understand and validate group-based trajectory modeling, which will ultimately help us more effectively optimize health care interventions, improve patient outcomes and produce cost-savings for patients, payers and ultimately the entire health care system, ” said study author Troyen A. Brennan, MD, Chief Medical Officer of CVS Health.