What do self-reported questionnaires on physical activity really measure in adolescents? Reliability and validity of two different questionnaires

Aim: To create and find accurate and reliable measurements of physical activity has been a challenge in epidemiological studies. This study has investigated the reliability and validity of two physical activity questionnaires in adolescents aged 13-18 years; 1. WHO, Health Behaviour in School-aged Children (HBSC) survey; 2. International Physical Activity Questionnaire (IPAQ). Methods: The questionnaires were administered twice for evaluating the reliability. The validity was assessed by comparing the answers from the questionnaires with the results from cardiorespiratory fitness test (VO2peak) and seven days activity monitoring with ActiReg (physical activity level and energy expenditure). Results: Intraclass correlation coefficients for reliability for WHO HBSC were 0.71 for frequency and 0.73 for duration. There was a significant difference between genders, 0.87 for girls and 0.59 for boys. Spearman correlation coefficients for WHO HBSC and IPAQ recoded in three categories measured against VO2peak were acceptable, i.e. the range 0.29 ? 0.39. There was low correlation on both questionnaires measured against ActiReg. Conclusion: These data indicate that the WHO HBSC and IPAQ (recoded in three categories) questionnaire was an acceptable instrument to measure cardiorespiratory fitness, but none of the questionnaires were valid in the measurement of physical activity among adolescents 13-18 years of age.