Longitudinal relationships among work-family conflict, enrichment, and work-life balance in Chinese nurses: a cross-lagged panel network analysis

When nurses’ satisfaction with balancing work and personal life declines, the consequences reach beyond the individual to threaten care quality, patient safety, and workforce retention. Achieving work-life integration is critical for nurses’ well-being and for the quality of patient care, yet longitudinal, multidimensional evidence on how work-family conflict (WFC), work-family enrichment (WFE), and work-life balance satisfaction (WLBS) influence one another over time remains limited. We recruited 1,677 clinical nurses from four Grade-A tertiary hospitals in Beijing, Xi’an, and Chengdu, China; 1,446 completed both waves of assessment across an approximately six-month interval (86.2% retention). WFC and WFE were each modeled as six sub-dimensional nodes (time-, strain-, and behavior-based conflict; developmental, affective, and capital/efficiency enrichment, in both work-to-family and family-to-work directions), and WLBS as a single node. A cross-lagged panel network (CLPN) was estimated, with accuracy and stability assessed via non-parametric bootstrapping. The thirteen-node system combined high autoregressive stability ( M = 0.39) with substantial cross-lagged connectivity: 78 of 156 possible cross-lagged edges (50%) were non-zero, with reinforcing pathways outnumbering dampening ones by more than two to one. WFC and WFE sub-dimensions were more densely connected within their own communities (weighted density 0.044–0.087) than with each other (weighted density 0.020; bootstrap difference = 0.035, 95% CI [0.025, 0.045]), and WFE sub-dimensions exhibited a positive, self-reinforcing resource spiral. WLBS did not emerge as a downstream outcome: its cross-lagged association with future strain-based work-to-family conflict (WLBS→WFC_Sb, β = −0.11) was essentially identical in magnitude to the reverse pathway (WFC_Sb→WLBS, β = −0.11), indicating a balanced, reciprocal relationship. Conflict and enrichment behaved as structurally separable systems rather than as opposite poles of a single continuum, and enrichment was not uniformly protective. Rather than functioning as a purely passive downstream outcome, WLBS was reciprocally and comparably predictive of future strain-based conflict, suggesting that satisfaction with work-life integration is better understood as an active participant in, rather than a pure cause or a pure terminus of, the work-family system. Because the CLPN is an exploratory method, these pathways require confirmatory replication; if they hold, organizational strategies should treat satisfaction with work-life integration as a variable worth monitoring and directly intervening on. Not applicable.