Mining social media posts: An alternative approach to understanding home health care workers’ experiences

Home healthcare workers (HCWs) play a critical role in maintaining the quality of life of older adults and people with disabilities. Aging baby boomers and rising life expectancy are quickly increasing the demand for HCWs. However, both low mental health and high turnover are widely reported among HCWs. Prior research on HCWs relies heavily on interviews and surveys, but the large number of authentic posts on social media from HCWs about their work experiences has been largely neglected. To address the above research gaps, our study employs both AI-powered automatic text analysis and computer-assisted human coding to mine and analyze HCWs’ social media posts. We hypothesized that work-environment-related stressors, lack of job support, and client-related challenges contribute to HCW stress. We found that HCWs often face hazardous work environments and challenging clients while lacking sufficient support, which contributes to increased stress in their jobs. The posts were predominantly negative, raising an alarm for the mental health of HCWs, as well as the conditions in which they work. Moreover, we demonstrated that the AI-powered approach and traditional human coding are complementary methods that can be used in tandem for mining social media posts. We also provided recommendations regarding how to improve HCWs’ well-being and social media mining methods.