Objectives: We assessed the test–retest reliability, criterion validity, and agreement of the Global Physical Activity Questionnaire (GPAQ) against accelerometry for estimating physical activity (PA) and sedentary behaviour (SB) among Indian adults at high risk of type 2 diabetes. Methods: We validated the original GPAQ in a subsample of 196 participants from the Kerala Diabetes Prevention Program. Participants were identified as being at high risk based on an Indian Diabetes Risk Score (IDRS) ≥ 60 and the absence of diabetes on an oral glucose tolerance test. To assess test–retest reliability, participants completed the GPAQ on two occasions, one week apart. Criterion validity and agreement were evaluated by comparing GPAQ responses with seven days of concurrent ActiGraph GT3X accelerometer data. Statistical analyses included intraclass correlation coefficients (ICCs) for reliability, Spearman’s correlation coefficients (ρ) for criterion validity, and Bland–Altman plots for agreement. Results: The mean (SD) age of participants was 47.0 (7.4) years, and 51% were female. ICCs for test–retest reliability ranged from 0.37 for moderate-intensity PA to 0.61 for vigorous-intensity PA. Spearman’s ρ for criterion validity ranged from 0.14 for SB to 0.38 for overall moderate-to-vigorous PA. Bland–Altman plots showed that the GPAQ consistently overestimated PA across all intensity levels while underestimating SB. Conclusions: The GPAQ demonstrated poor-to-moderate test–retest reliability, low criterion validity, and poor agreement with accelerometry for assessing PA and SB among Indian adults at high risk of type 2 diabetes.
Thirunavukkarasu et al. (2026) studied this question.