Reliability And Validation Of A Physical Activity Diary In Assessing And Evaluating Middle-aged People’s Physical Activity

Objectives:The purpose of this study was to assess the reliability and validity of a new developed physical activity diary for middle-aged people in china. The accelerometers were employed as criterion measures.Methods:A sample of 105 people (aged 52.9 years) was recruited. Everyone was told to wear an accelerometer and record an physical activity diary simultaneously for 7 days. Energy expenditure (EE), sedentary behavior (SB), light intensity physical activity (LPA), moderate-to-vigorous intensity physical activity (total and bouts) (MVPA, CMPA), MET rate were assessed. The statistical analysis of assessing reliability consisted of ICC, Spearman-Brown prophecy formula. Pearson’s rho(r) correlation was used to evaluate the relevance and the Bland-Altman analysis was used to evaluate the consistency between self-reported diaries and accelerometers. We also used Kappa rho (Kappa) and ROC curve to analyze the ability of diagnosing insufficient physical activity.Results:(1) The ICC of random two days in physical activity diaries was good. At least 3 days were needed to meet a ratio of 0.7 if you recorded at least 9 hours each day.2. EE, SB, LPA, MVPA, CMVPA, MET rate were all significantly between diary records and accelerometer, which were 0.29,0.32,0.40,0.40,0.47,0.23 separately. The correlation of LPA was good and the others were fair.3. Females were more liked to overestimate their diary’s MVPA and fatter people were more liked to overestimate their diary’s LPA. Females were more liked to overestimate their diary’s MVPA and fatter people were more liked to overestimate their diary’s LPA.4. The Kappa between the two methods was 0.213 in accordance with IPAQ’s criterion, which was not satisfied even though it was significant. After we analyzed using ROC curve, we thought the person who did not reach 56.2 minutes of MVPA on average day was "under-active"Conclusions:The physical activity we made was a reliable and valid self-report method to assess physical activity in epidemiological research, but less acurrate in detail value.