The association between intrinsic capacity and quality of life in community-dwelling older adults: a cross-sectional study

The concept of Intrinsic Capacity (IC) represents a paradigm shift in healthy aging research, yet the association between IC and Quality of Life (QOL) remains insufficiently explored, particularly using advanced analytical methods. To examine the association between IC and QOL, explore the network structure of IC components, and develop an exploratory post-LASSO ordinal logistic regression model for IC-level classification among community-dwelling older adults. This observational, cross-sectional study included 310 community-dwelling older adults from Chongqing, China, recruited via convenience sampling. We assessed Intrinsic Capacity (IC) with the WHO ICOPE screening tool and Quality of Life (QOL) with the WHOQOL-BREF (World Health Organization Quality of Life Brief Version). The main analytic strategy was threefold: first, we used multiple linear regression to determine the primary association between the IC composite score and QOL. Second, we employed network analysis to examine interrelationships among IC domains and characterize their statistical connectivity. Finally, we developed and internally validated an exploratory post-LASSO ordinal logistic regression model to test whether QOL and demographic data could distinguish individuals across low, moderate, and high IC levels with SHAP values used for feature importance interpretation. The median IC score was 4.00 (IQR: 3.00–4.75) with locomotion (65.16%) and cognition (58.71%) showing the highest decline rates. The exploratory network analysis suggested statistical interrelationships among IC components, with hearing showing relatively high statistical connectivity, as indicated by the highest strength centrality and expected influence in the estimated IC network. IC was significantly positively associated with QOL (B = 2.68, β = 0.355, P < 0.001), with significant QOL differences across IC level groups. The post-LASSO ordinal logistic model showed moderate discriminative ability for adjacent IC-level classification, with AUC values of 0.776 for low versus moderate IC and 0.794 for moderate versus high IC. IC was significantly associated with QOL among community-dwelling older adults. Environmental conditions, age, marital status, and education-related variables may help characterize differences in ordered IC levels. The exploratory post-LASSO ordinal logistic model showed moderate performance for adjacent IC-level classification, but the findings remain preliminary and require validation in larger independent longitudinal samples.