Developing prognostic models for life expectancy and geriatric outcomes

Northern California Institute/Res/Edu, San Francisco CA

Investigators

Linked publications, trials & patents

Abstract

DESCRIPTION (provided by applicant): Prognosis is the risk of an outcome for a condition or disease, such as mortality of mobility impairment. A prognostic index assembles the risk from multiple risk factors to predict outcome more accurately than a single factor would predict on its own. Current prognostic indexes suffer either from being inaccurate, un-interpretable, or both. Consequently, clinicians are unable to accurately determine prognosis in clinical decision making for older adults, often leading to poor outcomes for patients. Many older adults are referred to hospice in the last few hours or days before death, or never at all Other patients are exposed to the harms of cancer screening without the possibility they will live long enough to benefit. Currently available prognostic models estimate risk over a given time frame, such as risk of death over a 4 year period. Clinicians and patients do not think of risk this way, however. They want to know how long do I have? This project will address this question by making novel use of the exceptionally long follow up time in the nationally representative National Institute of Aging-funded Health and Retirement Study. The objectives of this project are to use the Health and Retirement Study (HRS) to develop and validate easy-to-use prognostic tools that estimate life-expectancy and time to the onset of disability, inability to manage medications or finances, or mobility impairment. Prognostic tools will be internally validated using the HRS and externally validated using the English Longitudinal Study on Aging (ELSA). This approach leverages the fact that the ELSA was based on the HRS. This project is expected to have a positive impact by informing, individualizing, and improving clinical decision-making in older adults.