An Internet-Based, Multicenter, Large-Sample Survey: Knowledge, Attitudes, and Practices and Pathway Mechanisms Among Caregivers of Children with Central Precocious Puberty

Central precocious puberty (CPP) requires long-term treatment and follow-up, during which caregivers play a central role in treatment adherence and family decision-making. This multicenter, cross-sectional, internet-based survey aimed to characterize knowledge, attitudes, and practices (KAP) among caregivers of children with CPP and to examine determinants and pathway relationships among these domains. Caregivers of children aged 2-16 years with CPP were recruited from eight pediatric endocrinology centers in China. An anonymous online questionnaire based on the disease-specific PRO-CaregiverCPP scale was administered. Knowledge scores, positive attitudes, and good practices were assessed using predefined criteria. Internal consistency was evaluated using Cronbach's alpha. Multivariable linear and logistic regression models were used to identify factors associated with knowledge scores, positive attitudes, and good practices. Observed-variable path analysis was performed to estimate direct and indirect associations among KAP domains. A total of 1,471 caregivers were included. The mean knowledge score was 6.31 ± 2.25; 72.5% and 58.8% of caregivers demonstrated positive attitudes and good practices, respectively. Higher knowledge scores were independently associated with greater odds of positive attitudes and good practices. Among caregivers of treated children, three-monthly injections were associated with higher knowledge scores and higher proportions of positive attitudes and good practices than monthly injections. Path analysis showed that knowledge was directly associated with practices and indirectly associated with practices through positive attitudes. These findings indicate a structured KAP pattern in CPP caregiving and support physician-led education and follow-up strategies to improve caregiver engagement and long-term disease management.