Using electronic health records with predictive analytics algorithms could help identify long-term care residents who are at a high risk of harmful falls, a new study has found.
The study, conducted at 13 California nursing homes, analyzed data from the CMS Minimum Data Set and EHRs for more than 5,000 residents. On its own, the CMS Minimum Data Set collected data infrequently and didn’t account for all the factors that can contribute to falls.
Researchers found that adding clinical data from EHRs to the CMS assessment data helped create more thorough and up-to-date predictions for residents at risk for falls. Adding EHR data with the CMS data improved the fall prediction algorithm’s accuracy by 13%, and confirmed 32.2% of observed falls among residents in the highest 10% of risk. Prior analysis using only the CMS data confirmed 28.6% of falls.
The authors of the study predict the EHR-CMS combined analytics program could prevent six additional falls and save $43,842 per year.
The results of the study were published in the Journal of the American Medical Informatics Association.