In research published in the Journal of the American Geriatrics Society, investigators developed and validated models that can predict the risk of fall-related injuries (FRI) in nursing home residents based on routinely collected clinical data.
The researchers conducted retrospective cohort study of long-stay US nursing home residents (mean age 85 years, 69.6% female) between January 1, 2016 and December 31, 2017 (n = 733 427) using Medicare claims and Minimum Data Set v3.0 clinical assessments. Predictors of FRIs were selected through statistical methods, from an original set of 70 predictors. To come up with a useful clinical tool, they calculated a score using the five strongest predictors in the model.
Within 2 years of follow-up, 6% of residents experienced one or more FRI. The prediction models achieved good discrimination and excellent calibration for accurately estimating individuals’ six-month and two-year risk of fall-related injuries. In the clinical tool to predict 2-year risk, the five characteristics included independence in activities of daily living (ADLs) (HR 2.27; 95% CI 2.14–2.41) and a history of non-hip fracture (HR 2.02; 95% CI 1.94–2.12). Performance results were similar in the validation sample.
“These models can be used by researchers and clinicians to accurately determine patient risk for fall-related injuries using routinely collected clinical assessment data,” the authors wrote. “In nursing homes, these models should be used to target preventive strategies.”
Source: Wiley
