Domain-specific adult sedentary behaviour questionnaire (ASBQ) and the GPAQ single-item question: A reliability and validity study in an Asian population

Abstract

This study examined the validity and reliability of a domain-specific Adult Sedentary Behaviour Questionnaire (ASBQ) and the Global Physical Activity Questionnaire (GPAQ) single-item sitting question using self- and interviewer-administered modes of administration against the triaxial ActiGraph wGT3X-BT accelerometer. The ASBQ and the GPAQ were administered twice, seven days apart. Participants were asked to put on the waist-worn accelerometer for seven days. Convergent validity was assessed using Spearman’s rho, mean absolute error (MAE), and Bland-Altman analysis (n = 78). Reliability was assessed using the Spearman’s rho and intraclass correlation coefficient (ICC) (n = 84). Participants were adults aged 20–65 years and identifying as Chinese, Malay, or Indian. Only the self-administered GPAQ was significantly correlated with accelerometry-based measures (rho: 0.46), but not the interviewer-administered version (rho: 0.12). MAE for GPAQ was 207.5–218.3 min/day in relation to the accelerometer and for ASBQ was 154.7–174.6 min/day. Bland-Altman plots demonstrated large limits of agreement between questionnaire and accelerometry-based measures. While the self-administered GPAQ demonstrated a moderate correlation with accelerometry, the mean bias and the limits of agreement were large. The GPAQ (rho: 0.68–0.79; ICC: 0.68–0.78) and the ASBQ (rho: 0.53–0.64; ICC: 0.66–0.74) showed moderate-to-good reliability for total sedentary time using either self- or interviewer-administration. Future research should incorporate accelerometers to generate useful sedentary behaviour measures.