Development and calibration of a novel social relationship item bank to measure health-related quality of life (HRQoL) in Singapore.

Background: Social relationships (SR) is an important domain of health-related quality of life. We developed and calibrated a novel item bank to measure SR in Singapore, a multi-ethnic city in Southeast Asia.
Methods: We developed an initial candidate pool of 51 items from focus groups, individual in-depth interviews and existing instruments that had been developed and/or validated for use in Singapore. We administered all items in English to a multi-stage sample of subjects, stratified for age and gender, with and without medical conditions, recruited from community and hospital settings. We calibrated their responses using Samejima's Graded Response Model (SGRM). We evaluated a final 30-item bank with respect to Item Response Theory (IRT) model assumptions, model fit, differential item functioning (DIF), and concurrent and known-groups validity.
Results: Among 503 participants (47.7% male, 41.4% above 50 years old, 34.0% Chinese, 33.6% Malay and 32.4% Indian), bi-factor model analyses supported essential unidimensionality: explained common variance of the general factor was 0.805 and omega hierarchical was 0.98. Local independence was deemed acceptable: the average absolute residual correlations were < 0.06 and 1.8% of the total item-pair residuals were flagged for local dependence. The overall SGRM model fit was adequate (p = 0.146). Five items exhibited DIF with respect to age, ethnicity and education, but were retained without modification of scores because they measured important aspects of SR. The SR scores correlated in the hypothesized direction with a self-reported measure of global health (Spearman's rho = - 0.28, p < 0.001).
Conclusion: The 30-item SR item bank has shown acceptable psychometric properties. Future studies to evaluate the validity of SR scores when items are administered adaptively are needed.