BACKGROUND: Falls are the leading cause of injuries in older adults.Some medications (e.g., antipsychotics, antidepressants) have beenassociated with increased fall risk. Standardized coding of falls inhealthcare claims and/or Minimum Data Set assessments provides lit-tle information regarding the events surrounding falls. UnstructuredEHR data may offer a novel opportunity to obtain this informationand improve fall prevention strategies.
OBJECTIVES: To evaluate the extent to which nurses' EHR notesrelated to fall incidents provide contextual and environmental infor-mation regarding the nature of falls in nursing homes.
METHODS: Data from 51 nursing homes in a large EHR system includedin the National Institute on Aging Long-Term Care Data Cooperative2018–2021) were used for the analysis. Nurses' notes describing fallevents were identified though a keyword search for“fall”or“fell”.Natural language processing (NLP) techniques were used to extractprominent latent constructs derived from the co-occurrence of words.Weighted unigram/bigram frequencies were calculated for each note(after removing“the”,“a”, etc.), then inversely weighted by the totalnumber of notes in which they occur (TF-IDF method). This processcalculates the informativeness of a word for a specific note. Latentconstructs were identified through an iterative optimization processknown as non-negative matrix factorization. This method assignsscores to both words and notes. Notes scoring high on one constructare likely to contain words also scoring high on that construct and viceversa. Two reviewers independently examined high-scoring wordsand read the corresponding full nurses' notes. After discussing theirinterpretations, they were able to achieve consensus on the constructlabels.
RESULTS: A total of 9104 unique fall events were identified. The first10 constructs corresponded to three stages of fall response events: 1)fall context (i.e., wheelchair, bed); 2) post-fall checks (i.e., injury, neu-rological, well-being); and 3) post-fall actions (i.e., assisted transfers,escalation to supervisors, root-cause analysis). Medications werenotably absent from the top 10 constructs identified.
CONCLUSIONS: Nurses' notes provide valuable contextual informationabout fall events, which may be useful in developing novel fall preven-tion strategies in nursing homes. Future research includes extendingthe use of NLP to further isolate and categorize fall response stages.Topic categories can also be compared with structured data(e.g., muscle weakness, medication) to further demonstrate the value-add of EHR nurses' notes.
