Scoring System Helps Predict Post-Hospital Mortality
For now, researchers see this as a tool for hospital comparisons, not patient care
MONDAY, June 8, 2015 (HealthDay News) -- A fairly simple scoring system appears to accurately estimate patients' risk of dying within a year of hospitalization, according to research results reported online June 8 in CMAJ, the journal of the Canadian Medical Association.
For a Hospital-patient One-year Mortality Risk (HOMR) model, the research team used administrative data including age, chronic medical conditions, any emergency department visits in the past year, and whether they live independently at home or need care. The researchers assigned points to each piece of information and ended up with a system that estimated a patient's risk of dying in the next year. A score of 21, for example, translated to a 1 percent risk, while a score of 55 meant a 90 percent risk. Then they tested the score's accuracy using records from more than three million patients at hospitals in Boston and the Canadian provinces Ontario and Alberta.
The researchers found that HOMR score was strongly and significantly associated with risk of death within one year after admission to hospital for nonpsychiatric indications in all populations and was highly discriminative. There was only a minor difference between the score's predictions and the actual percentage of patients at each hospital who died within a year.
"Similar performance was seen when the score was used in geographically and temporally diverse populations," the authors write. "The HOMR model can be used for risk adjustment in analyses of health administrative data to predict long-term survival among hospital patients."