Clinical tool includes five characteristics, including activities of daily living and history of nonhip fracture<\/p>\n<\/h3>\n
By Lori Solomon HealthDay Reporter<\/b><\/p>\n
<\/b><\/p>\n
TUESDAY, March 14, 2023 (HealthDay News) — A new risk prediction model can identify residents of nursing homes at greatest risk for fall-related injuries, according to a study published online March 8 in the Journal of the American Geriatrics Society<\/em>.<\/p>\n Matthew S. Duprey, Pharm.D., Ph.D., from the Brown University School of Public Health in Providence, Rhode Island, and colleagues developed and validated a series of models to predict the absolute risk for fall-related injuries in nursing home residents. Models relied on data from Medicare claims and Minimum Data Set v3.0 clinical assessments.<\/p>\n The researchers found that within two years of follow-up, 6.0 percent of residents experienced one or more fall-related injuries. The model included 70 predictors, and the discrimination of the two-year prediction model was good (C-index = 0.70), with excellent calibration. For the six-month model, calibration and discrimination were similar (C-index = 0.71). Five characteristics to predict two-year risk in the clinical tool included independence in activities of daily living (hazard ratio, 2.27) and a history of nonhip fracture (hazard ratio, 2.02). Validation results showed similar performance. <\/p>\n “Our models can be used to accurately estimate the six-month and two-year risk of fall-related injuries among long-stay nursing home residents. The full models can be automatically calculated with existing clinical data, while the short tool could be used by clinicians with similar performance,” the authors write. “We believe these models will provide researchers, clinicians, and policymakers with a useful tool to improve the quality of care in nursing homes.”<\/p>\n One author disclosed financial ties to the pharmaceutical industry.<\/p>\n Abstract\/Full Text (subscription or payment may be required)<\/a><\/p>\n <\/i><\/p>\n