Informal caregivers of people with Alzheimer's disease and related dementias (ADRD) are at risk of poor mental health. This study aimed to investigate the feasibility and validity of studying caregivers' mental stressors using online caregiving forum data (March 2018-February 2022) and natural language processing and machine learning (NLP/ML). NLP/ML topic modeling generated eight prominent topics, which we compared with qualitatively defined themes and the existing caregiving framework to assess validity. Among a total of 60,182 posts, 5848 were mental distress-related; for the ADRD patients (symptoms, medication, relocation, care duty share, diagnosis, conversation strategy) and the caregivers (caregiving burden and support). While we observed novel topics from NLP/ML-defined topics, mostly those were aligned with the existing framework. For feasibility assessment, qualitative title screening was done. The findings shed new light on the potential of NLP/ML text analysis of the online forum for informal caregivers to prepare tailored support for this vulnerable population.
Download full-text PDF |
Source |
---|---|
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11561247 | PMC |
http://dx.doi.org/10.1038/s44184-024-00100-y | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!