Psychometric performance of the Chinese version of Experimental EQ-TIPS-5L against PedsQL in young children with functional constipation: response distributions and construct validity

Health-related quality of life (HRQoL) measurement in early childhood is challenging, and evidence for preference-based instruments in infants and toddlers remains limited. The EuroQol Toddler and Infant Populations instrument (EQ-TIPS) was recently expanded to a five-level version (Experimental EQ-TIPS-5L) with seven dimensions, but its psychometric performance has not yet been evaluated in Chinese clinical populations. This study aimed to assess the response distribution, known-group validity, and convergent validity of Experimental EQ-TIPS-5 Lin children with functional constipation (FC). A total of 228 caregiver–child pairs were enrolled from a tertiary children’s hospital in China, with children aged 0–4 years diagnosed with FC. Caregivers completed the ExperimentalEQ-TIPS-5L and age-appropriate PedsQL versions (Infant Scales for 0–2 years, Generic Core Module for 2–4 years). Analyses were stratified by age group due to clinical and instrument differences. Known-group validity was evaluated by comparing Experimental EQ-TIPS-5L LSS, EQ VAS, and PedsQL total scores across subgroups defined by FC clinical characteristics using independent-samples t-tests and one-way ANOVA, with effect sizes estimated by Cohen’s d. Convergent validity was assessed using Spearman’s rank correlations at scales and item levels. The proportion of children in the best health state (1111111) was 28.95%. The Experimental EQ-TIPS-5L Level Sum Score (LSS) discriminated between subgroups defined by FC characteristics, with the largest effects differing between age groups: bowel movement frequency predominated among children aged 0–2 years (d = 0.95), whereas comorbidity was the strongest discriminator in children aged 2–4 years (d = 0.69). Behavioral or emotional problems showed consistent moderate effects across both age groups. PedsQL total scores largely mirrored these patterns. Experimental EQ-TIPS-5L LSS correlated with PedsQL total scores at r = − 0.43 (0–2 years) and r = − 0.63 (2–4 years). Item-level analyses revealed strong correlations between theoretically linked pairs (e.g., Pain and “hurts or aches”). This study provides the first evaluation of psychometric performance of the Experimental EQ-TIPS-5L in a Chinese clinical pediatric population. Among children with FC, Experimental EQ-TIPS-5L shows promising initial evidence of construct validity, with support for known-group and convergent validity. However, reliability, responsiveness, feasibility, a healthy control comparison, and a head-to-head three-level comparator were not assessed and remain priorities for future work.