Family external social support as a bridge to humanistic care: a cross-sectional network analysis with exploratory...

While family health is widely recognized as fundamental to college students' psychological development, the specific associative patterns through which different family health dimensions relate to humanistic care ability-and how these patterns differ by gender-remain incompletely mapped. This study employed cross-sectional network analysis and network comparison methods to examine 2,357 Chinese college students (ages 17-23; 1,325 females) who completed the Family Health Scale-Short Form and Humanistic Care Ability Questionnaire. Gaussian Graphical Models with LASSO regularization (EBIC, γ = 0.5) revealed three key findings. First, family external social support emerged as the node with the highest bridge expected influence in the observed network, statistically connecting family health to humanistic care dimensions, particularly trust and hope. This pattern is consistent with predictions derived from social capital theory but should not be interpreted as evidence of a developmental mechanism given the cross-sectional design. Second, a small negative association was observed between family health resources and trust (edge weight = -0.07), the interpretation of which is constrained by the operational content of the resource subscale (which assesses access to practical and healthcare resources rather than affluence) and by the small effect size. Third, patience and humility exhibited the highest network centrality, functioning as the most centrally connected nodes in the observed network; following recent methodological cautions, we do not interpret centrality as evidence of causal leverage. Network comparison tests revealed substantial overall gender invariance: global network structure, global expected influence, and node centrality did not differ significantly between male and female networks. Network comparison tests revealed only two small local edge differences and one bridge-centrality difference (all small in magnitude); these local differences are reported as exploratory. These cross-sectional associative patterns generate hypotheses regarding family-care linkages that future longitudinal and experimental research should test before being translated into intervention design.