A latent profile analysis of the pain-fatigue-sleep disturbance symptom clusters in postoperative cervical cancer patients: a multicenter cross-sectional study.

OBJECTIVE: To investigate the potential subtypes and influencing factors of the pain-fatigue-sleep disturbance (PFS) symptom cluster in postoperative cervical cancer patients, thereby providing a basis for developing targeted intervention measures. METHODS: A cross-sectional study design was adopted. Using convenience sampling, 374 postoperative cervical cancer patients admitted to three Grade A Level 3 hospitals in Liaoning Province between April 2025 and September 2025 were selected as study subjects. Questionnaire surveys were conducted using a demographic information form, the Brief Pain Scale, the Fatigue Severity Scale, the Pittsburgh Sleep Quality Index, the Hospital Depression and Anxiety Scale, and the Perceived Social Support Scale. Model fit indices for Latent Profile Analysis (LPA) included the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), sample-corrected Bayesian Information Criterion (aBIC), entropy, the Lore-Mundel-Rubin-Tobin (LMRT) test, and the bootstrap-based likelihood ratio test (BLRT). The AIC, BIC, and aBIC values of the four-profile model were markedly lower, indicating a significant improvement in model fit. Both the LMRT and BLRT test results showed p < 0.001, confirming that the four-category classification did not merely improve model fit by increasing the number of categories; rather, there are objectively four distinct patterns of pain-fatigue-sleep disturbance symptom combinations. The entropy value of 0.848 indicates excellent discriminatory ability among the subtypes, with nearly all patients accurately classified into their corresponding symptom subtypes. The study also identified influencing factors through univariate and multivariate logistic regression analyses. RESULTS: A total of 374 valid questionnaires were collected, with a response rate of 95.9%. The PFS of postoperative cervical cancer patients can be classified into four potential categories: Low Symptom Expression Type (29.679%), Pain-Dominated-Sleep-Impaired Type (26.738%), High Pain and Fatigue-Sleep Compensation Type (26.471%), and Multisymptom Coexistence-Sleep Disorder Type (17.112%). Logistic regression analysis revealed that perceived social support and laparoscopic surgery were common protective factors; hospital-related anxiety and depression were common risk factors; monthly household income, age, place of residence, postoperative complications, and occupation were specific risk factors. CONCLUSION: There is population heterogeneity in PFS among postoperative cervical cancer patients. Healthcare providers should implement targeted interventions based on protective and risk factors specific to different latent profiles to promote effective management of this symptom cluster.