| Popis: |
The purpose of this study was to develop a model to predict falls with major injury (FMI) among elderly nursing home (NH) residents. Identifying risk factors for FMI can be a preliminary step toward building a model and developing an assessment tool that will help create specific preventive interventions to enhance quality of life and reduce healthcare costs. It focused on two basic research questions: 1) What are the intrinsic factors and extrinsic factors that contribute to FMI in NH residents? 2) Which factors contribute to the best model for predicting FMI among NH residents? The framework for the study was Lawton’s Ecological Model of Aging (EMA) (Lawton & Nahemow (1973) which had been used in previous studies of falls and fall risk. After IRB approval from Kent State University, a request was made to the Centers for Medicare and Medicaid Services (CMS) to obtain the Long-Term Care Minimum Data Set (MDS) 3.0 collected between January 1, 2014 and December 31, 2014. Study inclusion criteria were (1) age 65 years or over, (2) having had at least one fall incident in 2014. Study design was an exploratory and retrospective cohort. The whole dataset (N= 442,729) was split into two parts—one for building the model (n= 221,365) and the other for testing the model (n= 221,364) for cross-validation. Binary logistic regression was used to build and test the model. All statistical analyses were conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). 34,611 residents (7.82%) out of the 442,729 included in the data analysis had sustained major injury upon falling. The FMI model demonstrates that residents most likely to have FMI are white females with advanced age of 85 or above who have some hearing difficulty, adequate vision, behavior problems, more independent and/or mobile functional status, continence, up to three comorbidities, and weight change (intrinsic factors). Furthermore, the FMI model reveals these extrinsic factors to be risk factors for FMI: use of a walking aid, no wheelchair use, and more medication burden (the number of medications taken.) |
