Abstract
BACKGROUND: Traditional approaches to assessing sleep quality in clinical nurses often overlook population heterogeneity and the complex interplay of influencing factors. This study employs Latent Profile Analysis (LPA) and Association Rule Mining (ARM) to identify distinct sleep quality subgroups and uncover key factor combinations, thereby informing targeted intervention strategies. METHODS: A total of 1,686 nurses from 123 hospitals in Shandong Province were recruited through multistage stratified sampling. LPA was used to classify participants based on seven sleep dimensions from the Pittsburgh Sleep Quality Index (PSQI), while ARM was applied to identify frequent itemsets of sleep disorder triggers. Key influencing factors were further examined using univariate analysis and multivariate logistic regression. RESULTS: Three latent sleep profiles were identified: high (63.11%), moderate (34.10%), and low (2.79%) sleep quality. The low-sleep subgroup was characterized by higher proportions of being unmarried/divorced (42.55%), low monthly income (≤ 3,000 CNY, 42.55%), non-permanent employment (76.60%), and severe psychological distress (44.68%). In contrast, the high-sleep subgroup featured higher rates of being married (85.62%), moderate income (3,001–7,000 CNY, 73.03%), and low psychological distress (51.32%). Key determinants included marital status (OR = 2.153/2.252), income (OR = 9.098), employment type (OR = 1.475), and psychological state (OR = 0.060–0.555). ARM revealed distinct risk combinations: “low income + non-permanent employment” (lift = 3.895) for the low-sleep group; “married + moderate income + non-permanent employment + patient conflict” for the moderate group; and “high income + low psychological distress” buffering night-shift effects in the high-sleep group. CONCLUSION: By integrating LPA and ARM, this study reveals the multidimensional heterogeneity and interactive mechanisms underlying clinical nurses’ sleep quality. The findings support a stratified intervention framework combining institutional safeguards with precision strategies to enhance sleep health management in nursing populations.