Conectar
To identify a frame of reference for resident safety management in nursing homes.
Q-methodology.
This study was conducted using Q-methodology to identify shared perspectives about resident safety management among nursing home professionals. Data were collected from 13 May 2023, through 29 August 2023. Thirty-four professionals, including nurses, care workers, social workers and physical therapists, classified Q-samples into a normal distribution grid through Q-sorting. Data analysis was performed using the PQmethod programme. Q-factors were interpreted by integrating interview transcripts, demographic data and factor arrays that organised the analysis results.
The analysis included the Q-sort of 33 professionals, with an average age of 46.03 years and 6.53 years of nursing home experience, after excluding one individual who did not fit any Q-factor. Four Q factors explaining 63% of the total variance were identified: constructing individualised possible risk trajectories, utilising ingrained safety principles, creating supportive safety environments and coordinating safety principles with individual needs.
Understanding the diverse subjectivities of professionals can help develop strategies that promote collaboration among nursing home professionals and support preventive safety management practices.
The frame of reference derived from nursing home professionals' perspectives suggests a resident-tailored framework.
This study supports the development of interprofessional education tailored to the specific needs of nursing home settings by identifying shared perspectives among nursing home professionals. The findings highlight the need for clear guidelines to help professionals balance resident autonomy with safety and assess the impact of family involvement.
Reporting involved qualitative and quantitative approaches, in compliance with the MMAT criteria for mixed-method research.
No Patient or Public Contribution.
To develop a method for computationally detecting fall events using clinical language models to complement existing self-reporting mechanisms.
Retrospective observational study.
Text data were collected from the unstructured nursing notes of three hospitals' electronic health records and the Korean national patient safety reports, totalling 34,480 records covering the period from January 2015 to December 2019. Note-level labelling was conducted by two researchers with 95% agreement. Preprocessing data anonymisation and English translation were followed by semantic validation. Five language models based on pretrained Bidirectional Encoder Representations from Transformers (BERT) and Generative Pretrained Transformer (GPT)-4 with prompt programming were explored. Model performance was assessed using F measurements. Error analysis was conducted for the GPT-4 results.
Fine-tuned BERT models with the English data set outperformed GPT-4, with Bio+Clinical BERT achieving the highest F1 score of 0.98. Fine-tuned Korean BERT with the Korean data set also reached an F1 score of 0.98, while GPT-4 achieved a competitive F1 score of 0.94. GPT-4 with prompt programming showed much higher F1 scores than GPT-4 with a standardised prompt for the English data set (0.85 vs. 0.39) and the Korean data set (0.94 vs. 0.03). The error analysis identified that the common misclassification patterns included fall history and homonyms, causing false positives and implicit expressions and missing contextual information, causing false negatives.
The clinical language model approach, if used alongside the existing self-reporting, promises to increase the chance of identifying the majority of factual falls without the need for additional chart reviews.
Inpatient falls are often underreported, with up to 91% of incidents missed in self-reports. Using language models, we identified a significant portion of these unreported falls, improving the accuracy of adverse event tracking while reducing the self-reporting burden on nurses.
Not applicable.
FreshRSS 