Authored by: Laura Kanov, Senior Vice President, Product Strategy
Among the 50% to 70% of patients with a terminal illness who prefer to be cared for and die at home, only about 25% have a home death. Nearly a third of Americans who die after age 65 years will have spent time in an intensive care unit in their final 3 months of life, and almost a fifth undergo surgery in their last month. A disproportionate amount of health care resources and expenditures are spent on patients who are terminally ill. Health care experts estimate that one-quarter of all Medicare costs — US $150 billion annually — goes to treating patients in their last year of life. Worst of all, despite high costs, evidence demonstrates that these patients receive inadequate quality of care – characterized by fragmentation, error, overuse and poor quality of life.
Palliative care and hospice services improve patient-centered outcomes such as depression, pain, patient and family satisfaction and receipt of care in the place of patient choosing. How can providers and plans identify those individuals approaching end of life in a timely way to help improve quality of care, understand and fulfill end of life wishes?
In order to help clients address these issues, HBI Solutions applied advanced machine learning techniques to develop a predictive algorithm for all-cause 1-year mortality (n= 125,896) and prospectively validated its accuracy in a statewide population of individuals aged 65 and older (n= 153,199).
HBI Solutions began by aggregating over 14,000 facts and features from demographic, social determinant, laboratory, radiographic, medication, primary & secondary diagnoses and procedures. These were reduced to the most influential 99 features. We found our model (C-statistic = 0.91) to outperform other models (C-statistics = 0.65 – 0.79) that were derived from more limited sets of risk factors or data sources.
We believe this tool can help providers and plans identify individuals for palliative care and quality of life discussions to improve end of life quality and reduce medical spend futility in line with patient and family wishes.