With the ageing population on the rise, the need to improve the quality and efficiency of healthcare for older adults has never been more pressing. This demographic is incredibly diverse, often dealing with a multitude of diseases. Designing effective intervention strategies for such a varied group is a significant challenge. To address this, researchers utilised unsupervised machine learning techniques to categorise individuals aged 65 and older who had recently entered long-term care. The study, conducted in Tsukuba City, Ibaraki Prefecture, and Sammu City, Chiba Prefecture, focused on the relationship between these classifications (referred to as ‘clinical subtypes’) and their subsequent health outcomes.
Six clinical subtypes were identified in Tsukuba City: i. musculoskeletal and sensory disorders, ii. cardiac disorders, iii. neurological disorders, iv. respiratory disorders and cancers, v. Insulin-dependent diabetes, and vi. other conditions. This classification system was similarly validated when the data from Sammu City were analysed.
Regarding health outcomes, individuals with cardiac diseases, respiratory diseases/cancers, and insulin-dependent diabetes showed a higher risk of mortality compared to those with musculoskeletal and sensory disorders. Additionally, those with cardiac diseases, respiratory diseases/cancers, and other conditions experienced an increase in the severity of care needs.
The implications of this research extend beyond the individuals in need of care to their families and the healthcare staff involved. By developing targeted interventions for each identified clinical subtype, this research could significantly influence healthcare policies and practices, leading to more effective and personalised care for older adults in long-term care.
More information: Yuji Ito et al, Clinical subtypes of older adults starting long-term care in Japan and their association with prognoses: a data-driven cluster analysis, Scientific Reports. DOI: 10.1038/s41598-024-65699-6
Journal information: Scientific Reports Provided by University of Tsukuba
