A latent profile and network analysis of sleep inertia in shift nurses

To identify latent profiles of sleep inertia among shift nurses, examine factors associated with latent profile membership, and compare the central associated factors across the networks of different profiles. A cross-sectional study was conducted in January 2026. With a combination of convenience and purposive sampling, 519 clinical shift nurses from Lishui, Zhejiang Province, China, were selected. Data were collected using a demographic questionnaire, the Sleep Inertia Questionnaire, the Patient Health Questionnaire, the Generalized Anxiety Disorder Scale, the Fatigue Scale, and the Ten-item Personality Inventory in China. Descriptive statistics and network analysis were performed using R 4.6.0; latent profile analysis (LPA) and R3STEP-adjusted multinomial logistic regression were conducted in Mplus 8.3. Network analysis assessed the relative centrality of nodes using standardized strength, with expected influence used for sensitivity analysis. Network comparison tests with Holm correction were used to compare network characteristics across latent profile subgroups. Sleep inertia among the 519 shift nurses was classified into 3 profiles: the low inertia–adaptive group (18.11%), the moderate inertia–sluggish response group (48.75%), and the high inertia–cognitive reactive dysregulation group (33.14%). R3STEP regression analysis indicated that average daily sleep duration, anxiety, and fatigue were significantly associated with latent profile membership ( P < 0.05). In the overall sample, the central associated factors were depression and cognitive inertia; the central associated factors in the low inertia–adaptive group were emotional stability and conscientiousness, whereas in both the moderate inertia–sluggish response group and the high inertia–cognitive reactive dysregulation group, the central associated factors were depression and anxiety. After Holm correction, no statistically significant differences in network structure or overall network strength were found in pairwise comparisons among the three profiles. Sleep inertia among shift nurses exhibits heterogeneity, and identifying central associated factors within the networks of different latent profiles may facilitate a better understanding of variable associations across profiles and offer directions for subsequent research. Not applicable.