Alzheimer Caregiving Problems According to ADLs: Evidence from Facebook Support Groups

Background and study goal: Social media are a societal phenomenon today, including the oldest generation, yet they are seldom used in current health research to identify the needs of persons with Alzheimer's disease (PADs) and their carers. There is an even bigger research gap in the analysis of caregivers' communication in online support groups and its classification according to activities of daily living (ADLs). For this, the goal of this study is to identify real-life practices of informal caregivers who care for PADs based on the analysis of their communication in Facebook groups. Methods: A sample of 1603 contributions was obtained from support groups by keyword search, manual coding, and verification; thus, the contributions in the sample are relevant for the individual basic ADLs of PADs. Next, five main conversation topics were identified for each ADL. This was done using the topic extractor based on simple parallel threaded implementation of LDA with a sparse sampling scheme and data structure. Results: The qualitative dimension of research identified discussion topics as well as specific caregiver behavior patterns for each ADL. The quantitative dimension determining the level of engagement of group members in individual ADLs was also measured. The highest engagement was found in activities of feeding and drinking, followed by bathing. In contrast, the activities of dressing, continence, and toileting attract the lowest interest. Moreover, the causal links between the topics discussed within the areas of ADLs were identified. Conclusions: The acquired knowledge can help further research focus on the most problematic areas relevant for people with AD in order to increase their quality of life and at the same time reduce the caregiver burden. The study expands the information of the demands posed by the individual caregiver activities, specifically in the context of activity-based costing or time-based activity costing.