Validation of youth physical activity questionnaire (YPAQ) against actigraph GT3X accelerometer among preadolescents and adolescents in urban setting of Karachi, Pakistan

Date of Award


Document Type


Degree Name

Master of Science in Epidemiology & Biostatistics (MSc Epidemiology & Biostats)


Community Health Sciences


Background and purpose: Physical activity (PA) is one of the four major modifiable risks and has a fundamental role in the prevention and treatment of chronic diseases. PA pattern established during child hood is an important contributor to sound health in adulthood. Assessment of PA through objective methods is expensive and laborious to implement during surveys; thus, estimates of PA rely on subjective instruments including the Youth Physical Activity Questionnaire (YPAQ). However, these instruments have not been validated in the South Asian population. We aimed to validate a subjective tool that can provide accurate and precise measurement of PA in children to assess trends and effect of impending interventions enhancing PA in sedentary population. Objectives The objectives of this study were to assess validity and reliability of modified version of YPAQ for various intensities of PA and to determine Actigraph optimal thresholds on YPAQ for various durations of moderate to vigorous physical activity (MVPA) in children and youth. Methods PA measured for seven consecutive days through Actigraph GT3X accelerometer on aged 9 to 14 years school going children in urban settings of Karachi. YPAQ administered twice to estimate one week PA, first when Actigraph was attached and subsequently when it was retrieved. Pearson and Spearman rank coefficients were calculated to assess criterion validity and reliability of YPAQ on repeat estimates one week apart. To test the hypothesis, the 95% CIs and p values were calculated with reference to null hypothesized values of 0.5 and 0.7 for validity and reliability respectively. Bland Altman plots constructed to assess mean bias and limits of agreement for all intensities. ROC curve constructed to determine Actigraph optimal threshold on YPAQ for moderate-to-vigorous estimates of YPAQ. To process the accelerometer data Actilife version 6.0 and for all statistical analysis SPSS and Medcalc software were used. Results A total 252 children participated and 234 provided valid accelerometer data. The coefficient for criterion validity of PA on all days was 0.11(95% CIs: -0.02, 0.23) and for MVPA was 0.37, (95% CIs: 0.24, 0.49). The estimate was fair for the youth compare to younger children 0.42, (95% CI: 0.28, 0.53) vs. 0.32, (95% CIs: 0.19, 0.44). Bland Altman plots showed YPAQ over-estimated time spent in total activities and MVPA domain, however under-estimated sedentary and light activities. The sensitivity at 30 and 45 minutes of MVPA daily were 75.3 &59.7% and specificity were56.9&68.4% respectively.The areas under the curve were 0.68 and 0.65 respectively at these cutoffs.Test-re-test coefficients for overall activities was suboptimal 0.54, (95% CI: 0.42, 0.67) and for MVPA wasmoderately strong0.62, (95% CI: 0.49, 0.74). The estimates for these domains were better for youth compare to children 0.64, (95% CI: 0.51, 0.76) vs. 0.57, (95% CI: 0.45, 0.69). Conclusions: YPAQ has limited validity and reliability for overall activities, however has fairvalidity and reliability for MVPA particular for youth. The specificity was fair at 30 and 45 minutes of MVPA daily however,the area under the curve werebelow the desirable ranges. Since no subjective tool has acceptable validity and reliability, therefore YPAQ can be used at large scale complimented with other appropriate tool like activity logs/diaries to reduce inherent source of bias including recall, respondent and social desirability biases .However, a better assessment tool is required for children to get valid and precise estimates. Children spent less time in MVPA, particularly adolescent girls with high adiposity indices- a population most at need for preventive physical activity initiatives.

This document is available in the relevant AKU library