Latent classes of frailty and their association with intrinsic capacity in older cancer survivors: A study under the healthy aging paradigm

The aim of this study was to identify frailty profiles using latent class analysis (LCA) and examine their associations with intrinsic capacity (IC) among Chinese elderly cancer survivors. A total of 308 elderly cancer survivors were recruited from a tertiary hospital in Nantong, China between November 2023 and April 2024, and data were collected through questionnaires and clinical assessments. LCA was used to classify frailty subtypes, univariate analysis and multinomial logistic regression (reference: robust group) were used to identify associated factors, and one-way ANOVA was used to compare IC differences across subtypes. Three frailty profiles were identified-frail (31.5%), pre-frail (19.8%), and robust (48.7%)-with significant IC variations. Lower monthly household income (odds ratio (OR) = 16.00, <i>p</i> = 0.028), smoking (OR = 8.76, <i>p</i> = 0.013), malnutrition (OR = 5.25, <i>p</i> = 0.044), activities of daily living (ADL) (OR = 71.31, <i>p</i> < 0.001), depression (OR = 15.91, <i>p</i> = 0.048), and fatigue (OR = 33.43, <i>p</i> < 0.001) were independent risk factors. These findings indicate that Chinese elderly cancer survivors exhibit heterogeneous frailty profiles and that IC decline is positively associated with the severity of frailty. The identified risk factors and subtype characteristics provide a basis for devising tailored interventions to improve health outcomes in this population.