ObjectiveTo deal with the increasing long-term care (LTC) needs of elderly people in Taiwan, the government launched the Ten-year Long-term Care Project (TLTCP) in 2007, and through the care management system, care plans for those in need were distributed and implemented by care managers according to the single assessment process. Based on the emphasis of linking the right need assessment to the care plan, this study aimed to explore the need profiles of LTC recipients with regard to their health indicators to serve as a validity check on the identified dependency levels and care plans in the current care management system. DesignA model based on latent class analysis (LCA) was used for dealing with the issue of health heterogeneity. LCA provides an empirical method that examines the interrelationships among health indicators and characterizes the underlying set of mutually exclusive latent classes that account for the observed indicators. The analysis included a total of 2901 elderly care recipients in the LTC dataset from a southern city, 1 of the 5 major metropolitan areas in Taiwan. The identified dependency levels of the samples and their care plans in need assessment were compared and discussed. ResultsFour need profiles were explored in the LTC dataset. Apart from the low (LD) (32.95%) and moderate dependent groups (MD) (17.48%), there were 2 groups identified among the high-dependency levels, including the severe physical and psychological dependency (SPP) (26.37%) and the comorbidities and severe dependency (CSD) groups (23.20%), which in sum were approximately identified as high dependency (HD) by care managers in the LTC dataset. In addition, the CSD group currently costs more for their care plans on average in LTC services (NT. 277,081.15, approximately 9200 USD) than the SPP group (NT. 244,084.21) and the other groups. ConclusionNeed assessment is a key to success in care management in LTC. The results of this study showed the importance of focusing on multifacet indicators, especially the mental and social health indicators in need assessments by improving the unified assessment process to sensitively detect those with various needs and then link them to the right care plan.