Prediction Model Using VA Data May Help Identify Primary Care Patients at Increased Risk for Hospitalization or Death
BACKGROUND:
Given the increasing emphasis on managing population health, it is important to identify patients at greater risk for adverse outcomes. From a clinical perspective, it is desirable to identify not only which patients are at high risk for hospitalization, but also those who are at risk of death or other major clinical event. This information can guide the use of enhanced care management programs and services that have potential to improve patients' outcomes. In an attempt to identify high-risk patients, investigators in this study developed statistical models using health information from VA's clinical and administrative databases to predict the risk of hospitalization or death among all Veterans (n=4,598,408) who were assigned to a primary care provider as of 10/1/10. Potential predictor variables from six categories were examined: demographics, medical conditions, vital signs, prior year use of VA health services, medications dispensed, and laboratory results. The outcome variable was the first occurrence of hospitalization for any cause, death without hospitalization, and combined hospitalization or death within 90 days and 1 year following 10/1/10. If a Veteran died after he or she was hospitalized, investigators counted only the first event of hospitalization.
FINDINGS:
- Prediction models using electronic clinical data accurately identified Veterans receiving VA primary care who were at increased risk of hospitalization or death. Of the top 5% of Veterans in terms of predicted risk, 51% were hospitalized or died within the following year.
- Predictors of death were quite different from predictors of hospitalization. In general, clinical and demographic characteristics (i.e., increasing age, metastatic cancer) were most predictive of death, while recent use of health services was most predictive of hospitalization.
- The authors suggest that in clinical settings, these values can be used to identify high-risk patients who might benefit from care coordination and special management programs, such as intensive case management, telehealth, home care, specialized clinics, and palliative care.
IMPLICATIONS:
These risk predictions have been implemented in VA's electronic medical record system, enabling real-time viewing by primary care clinicians for their panels of patients.
LIMITATIONS:
- Study data did not include non-VA hospitalizations that were not paid for by VA and are thus likely to omit a greater proportion of hospitalizations for Veterans over age 65 who use Medicare. Nonetheless, the model appears to perform as well at predicting high-risk Veterans over age 65 as for those under age 65.
AUTHOR/FUNDING INFORMATION:
This work was conducted entirely for Operations and not as part of research. Drs. Maynard and Bryson are part of HSR&D's Northwest Center for Outcomes Research in Older Adults, Seattle, WA. Dr. Fihn is Director of VA's Office of Analytics and Business Intelligence.
Wang L, Porter B, Maynard C, Evans G, Bryson C, Sun H, Gupta I, Lowy E, McDonell M, Frisbee K, Nielson C, Kirkland F, and Fihn S. Predicting Risk of Hospitalization or Death among Patients Receiving Primary Care in the Veterans Health Administration. Medical Care April 2013;51(4):368-73.
