Reliability And Validity Of Physical Activity Questionnaire In Anhui Province

Objective: Physical inactivity is associated with a variety of adverse health outcomes.Accurately measuring physical activity has become an important issue in epidemiological surveys.Due to the high cost of objective measurement methods,physical activity questionnaire(PAQ)with the advantages of low cost and easy acceptance are still the main means of assessing physical activity in large-scale epidemiological surveys.However,physical activity varies according to geographical and cultural factors,and it is difficult to accurately estimate the level of physical activity in different regions using a standardized questionnaire.Currently,most domestic studies use adapted versions of foreign questionnaires or construct new PAQ.To assess the level of physical activity in Anhui Province,we referred to domestic and foreign studies and combined local conditions to construct a PAQ for Anhui Province.The questionnaire included physical activities such as farm work,square dancing,and housework that were less or not investigated in previous questionnaires.However,as a subjective measurement method,PAQ can only be used to accurately measure physical activity levels after being validated for reliability and validity.Therefore,the purpose of this study was to verify the reliability and validity of the PAQ in Anhui Province and to further investigate the factors influencing the validity.Methods: This study recruited 620 volunteers in Anhui Province and completed the first PAQ survey(PAQ1)in July-August 2021 and the second PAQ survey(PAQ2)in October 2022.Different reference methods were used to validate the validity of PAQ.Considering the effects of weekday and season,482 people completed a one-year 24-hour physical activity recall(24PAR)survey,which was conducted continuously for three days in each season(2 workdays and 1 non-workday),totaling 12 times.In addition,138 people completed two 7-day physical activity diaries separated by one month.The investigated physical activities were assigned the corresponding metabolism equivalent(MET),and physical activities were classified as sedentary,light physical activity,moderate physical activity and vigorous physical activity according to 1.0-1.5 MET,1.6-2.9 MET,3-5.9 MET and ≥6 MET,respectively.The moderate-tovigorous physical activity was the sum of moderate physical activity and vigorous physical activity.Total physical activity included all intensities of physical activity.The intraclass correlation coefficient(ICC)was used to evaluate the reliability of the PAQ based on the consistency of the PAQ between the two repeated measures.Spearman’s correlation coefficient was used to evaluate PAQ validity based on the agreement between the single-measure PAQ(PAQ1 or PAQ2)and 12 24 PARs or two 7-day physical activity diaries.To adjust for random within-person variation arising from multiple 24 PARs and physical activity diaries,we corrected the Spearman correlation coefficient to obtain the deattenuated Spearman correlation coefficient.When the correlation coefficients for reliability and validity were 0-0.19,0.20-0.39,0.40-0.59,0.60-0.79,and 0.80-1.00,they represented poor,fair,moderate,good,or very good,respectively.In addition,Bland-Altman analysis was used to evaluate the consistency between PAQs and 24 PAR,and a multiple linear regression model was used to further explore the factors affecting validity.Results: The results of the reliability analysis showed that the ICC and 95% CI of the PAQ for total physical activity,sedentary,light physical activity,moderate physical activity,moderate to vigorous physical activity,and vigorous physical activity were 0.64(0.58,0.70),0.85(0.83,0.88),0.23(0.09,0.35),0.71(0.66,0.75),0.72(0.67,0.76),0.67(0.61,0.72),respectively.For the validity analysis,the deattenuated Spearman correlation coefficients for PAQ2 compared to 24 PAR as the reference standard(n = 300)were 0.54,0.74,0.35,0.64,0.65,and 0.39 for total physical activity,sedentary,light physical activity,moderate physical activity,moderate to vigorous physical activity,and vigorous physical activity,respectively.The deattenuated Spearman correlation coefficients between PAQ2 and 7-day physical activity diary as the reference standard(n = 126)for the above six categories were 0.54,0.80,0.41,0.65,0.65,and 0.61.The Bland-Altman analysis showed that compared with 24 PAR,PAQ2 overestimated total physical activity by 36.10 MET-h/day(95% LOA:-146.70,218.90),P < 0.05.The results of multiple linear regression analysis showed for every one-year increase in age,the difference in moderate physical activity measured by 24 PAR and PAQ2 decreased by 0.82 MET-h/day,while the difference in vigorous physical activity increased by 0.86 MET-h/day.In females,the difference in moderate physical activity measured by 24 PAR and PAQ2 was 27.71 MET-h/day lower than that in males,but the difference in vigorous physical activity was 17.82 MET-h/day higher than that in males.In the obese population,the difference in sedentary behavior measured by 24 PAR and PAQ2 was 7.81 MET-h/day lower than that in the normal population,while in the overweight population the difference in light physical activity measurement was 6.15 MET-h/day lower than that in the normal population.Finally,in the population with a junior high school education level,the difference in sedentary behavior measured by 24 PAR and PAQ2 was 12.84 MET-h/day lower than that in the population with a primary school education level or below,but the difference in vigorous physical activity measurement was 27.86 MET-h/day higher than that in the population with a primary school education level or below.In the population with a high school education level or above,the difference in vigorous physical activity measurement was 20.40 MET-h/day higher than that in the population with a primary school education level or below.All P < 0.05 for the above results.Conclusion: The PAQ constructed in this study was validated in the Anhui population and achieved good overall reliability as well as moderate validity,which can better reflect the physical activity level of the Anhui population.The PAQ overestimates the level of physical activity compared to the 24 PAR.Age,sex,BMI and education were the factors influencing the validity of the questionnaire.