RESUMO
This study evaluates the agreement between different methods to assess moderate to vigorous physical activity (MVPA) in a large sample of Finnish adults. Methods were classified and examined pairwise (accelerometer vs diary; questionnaire vs interview). Proportion of participants meeting the aerobic health-enhancing physical activity (HEPA) recommendation was compared pairwise between all four methods. The present study of 1916 adults aged 18-75 years (mean age 50 years, 57% women) is a sub-sample of population-based Health 2011 Study conducted by the National Institute of Health and Welfare in Finland. Participants used accelerometer for 7 days and completed physical activity (PA) diary during the same period. PA questionnaire and interview were completed retrospectively to assess typical weekly PA over the past year. Agreement between the methods was analyzed with paired samples t-test and Bland-Altman plot. Kappa-test was used to compare the prevalence of meeting the HEPA recommendation. The accelerometer resulted in 13 minutes (P < 0.001) higher weekly total amount of MVPA compared to diary. According to Bland-Altman plot, the 95% limit of agreement was from +273 to -247 in weekly minutes. The comparison between questionnaire and interview showed non-significant mean difference of 3 minutes (P = 0.60) in weekly MVPA, but the Bland-Altman plot showing the 95% limit of agreement from +432 to -427 in weekly minutes. Agreement of meeting HEPA recommendation was moderate between questionnaire and interview (κ = 0.43) but poor (κ = 0.20-0.38) between other comparisons. The inter-method differences were large especially at the individual level. Thus the assessment of PA is strongly method-dependent and not interchangeable.