Skip to main content

Advertisement

Table 4 Validity results of new PAQs

From: A systematic review of reliability and objective criterion-related validity of physical activity questionnaires

Age Group Reference Criterion method Duration of validation PAQ Variables tested Criterion intensity thresholds Validity results
        Correlation coefficients Agreement
Youth Dwyer (2011)[19] Acc (ActiGraph) 4 - 5 days Pre-PAQ Level 5 min/day(Q) – VPA min/day(Acc) >5016 counts/min Pearson r = 0.17 MD (95 % LoA) = 1.9 ± 39.4 min/day
      Level 4 min/day(Q) – MPA min/day(Acc) 3560-5016 counts/min Pearson r = 0.13 MD (95 % LoA) = 48.2 ± 73.1 min/day
      Level 3 min/day(Q) – LPA min/day(Acc) 1592-3560 counts/min Pearson r = −0.07 MD (95 % LoA) = −4.8 ± 100.7 min/day
      Levels 1–2 min/day(Q) – sedentary min/day(Acc) <1592 counts/min Pearson r = 0.19 MD (95 % LoA) = −235.4 ± 147.7 min/day
Youth Economos (2010)[20] Acc (ActiGraph) 2 days BONES PAS High METs(Q) – total counts/min(Acc) – Spearman r (95 % CI) = 0.25 (−0.07;0.52) –
      High METs(Q) – VPA(Acc) 6-9 METs, 1952–5724 counts/min Spearman r (95 % CI) = 0.23 (−0.09;0.51) –
      Moderate-high METs(Q) – total counts/min(Acc) – Spearman r (95 % CI) = 0.27 (−0.05;0.54) –
Youth Martinez-Gomez (2010)[21] Acc (ActiGraph) 1 day RPAR Total MET-min(Q) – total counts(Acc) – Pearson r = 0.42 (P = 0.021) κ = 0.16
      MVPA min(Q) – MVPA counts(Acc) ≥2000 counts/min Pearson r = 0.52 (P < 0.001) MD (95 % LoA) = 2.15 ± 7.19 min
   Acc (Biotrainer) 1 day   Total MET-min(Q) – total counts(Acc) – Pearson r = 0.40 (P = 0.025) κ = 0.39
      Total MET-min(Q) – total counts/mov(Acc) – Pearson r = 0.54 (P = 0.004) κ = 0.16
Youth Philippaerts (2006)[22] Acc (ActiGraph) 7 days FPACQ Total hr/week(Q) – total counts(Acc) – Pearson r = 0.56 (P < 0.01) –
      Total hr/week(Q) – mean counts/min(Acc) – Pearson r = 0.43 (P < 0.05) –
      TEE(Q) – total counts(Acc) – Pearson r = 0.58 (P < 0.01) –
      TEE(Q) – mean counts/min(Acc) – Pearson r = 0.49 (P < 0.05) –
      Inactivity(Q) – total counts(Acc) – Pearson r = −0.13 –
      Inactivity(Q) – mean counts/min(Acc) – Pearson r = −0.06 –
Youth Ridley (2001)[23] Acc (Caltrac) 2x 1 day CDPAQ Total METs(Q) – total counts(Acc) – Pearson r = 0.41 (P < 0.05) –
      Total compendium METs(Q) – total counts(Acc) – Pearson r = 0.54 (P < 0.05) –
      Total mins(Q) – total counts(Acc) – Pearson r = 0.41 (P < 0.05) –
   HR (Polar) 2x 1 day   MVPA mins(Q) – MVPA mins(HR) ≥145 bpm Pearson r = 0.66 (P = 0.01) –
   Acc (Caltrac) 2x 1 day CDPAQ-HC Total METs(Q) – total counts(Acc) – Pearson r = 0.25 (P < 0.05) –
      Total compendium METs(Q) – total counts(Acc) – Pearson r = 0.22 (P < 0.05) –
      Total mins(Q) – total counts(Acc) – Pearson r = 0.33 (P < 0.05) –
   HR (Polar) 2x 1 day   MVPA mins(Q) – MVPA mins(HR) ≥145 bpm Pearson r = 0.48 (P = 0.05) –
Youth Ridley (2006)[24] Acc (ActiGraph) 1 day MARCA PAL(Q) – total counts(Acc) – Spearman r = 0.45 (P < 0.01) –
Youth Telford (2004)[25] Acc (ActiGraph) 8 days CLASS-parental report 5-6 yrs: total min/day(Q) – total min/day(Acc) – Spearman r = −0.04 MD (95 % LoA) = −140.7 (−164.9;-116.6) min/day
      10-12 yrs: total min/day(Q) – total min/day(Acc) – Spearman r = 0.09 MD (95 % LoA) = 11.2 (−6.9;29.4) min/day
      5-6 yrs: total min/day(Q) – total raw counts/day(Acc) – Spearman r = 0.05 –
      10-12 yrs: total min/day(Q) – total raw counts/day(Acc) – Spearman r = 0.11 –
     CLASS-self 10-12 yrs: total min/day(Q) – total min/day(Acc) – Spearman r = −0.04 MD (95 % LoA) = 1.5 (−17.2;20.3) min/day
      10-12 yrs: total min/day(Q) – total raw counts/day(Acc) – Spearman r = 0.06 –
Youth Treuth (2003)[26] Acc (ActiGraph) 4 days GAQ Yesterday: GAQ score(Q) – mean counts/min(Acc) – Pearson r = 0.27 (P < 0.05) –
      Usual: GAQ score(Q) – mean counts/min(Acc) – Pearson r = 0.29 (P < 0.05) –
      Yesterday: TV watching(Q) – mean counts/min(Acc) – Pearson r = −0.145 (P = 0.24) –
      Usual: TV watching(Q) – mean counts/min(Acc) – Pearson r = −0.004 (P = 0.98) –
      Yesterday: other sedentary(Q) – mean counts/min(Acc) – Pearson r = 0.0227 (P = 0.85) –
      Usual: other sedentary(Q) – mean counts/min(Acc) – Pearson r = −0.0916 (P = 0.46) –
     Activitygram Activitygram score(Q) – mean counts/min(Acc) – Pearson r = 0.37 (P < 0.002) –
Youth Treuth (2005)[27] Acc (Actiwatch) 6 days Fels PAQ Elementary: Fels PAQ score(Q) – mean counts/min(Acc) – Spearman r = 0.34 (P = 0.004) –
      Middle: Fels PAQ score(Q) – mean counts/min(Acc) – Spearman r = 0.11 (P = 0.31) –
      High: Fels PAQ score(Q) – mean counts/min(Acc) – Spearman r = 0.21 (P = 0.006) –
Youth Welk (2007)[28] Acc (ActiGraph) 7 days YMCLS Weekly PA bouts(Q) – weekly PA bouts(Acc) – r = 0.24 (P < 0.05) MD (95 % LoA) = −8.4 ± 28.4 min
      Previous day: total MVPA mins(Q) – total MVPA mins(Acc) 3-6 METs r = 0.53 (P < 0.05) MD (95 % LoA) = 14.5 ± 173.9 min
Youth Wong (2006)[29] Acc (ActiGraph) 7 - 9 days SHAPES VPA min/day(Q) – VPA min/day(Acc) ≥8200 counts/min Spearman r = 0.25 (P = 0.07) –
      MVPA min/day(Q) – MVPA min/day(Acc) ≥3200 counts/min Spearman r = 0.44 (P < 0.01) –
      MPA min/day(Q) – MPA min/day(Acc) 3200-8199 counts/min Spearman r = 0.31 (P = 0.02) –
Adults Ainsworth (2000)[30] Acc (Caltrac) 2x 7 days KPAS 3 point summary index(Q) – MET-min/day(Acc) – Spearman r = 0.53 (P < 0.01) –
      4 point summary index(Q) – MET-min/day(Acc) – Spearman r = 0.49 (P < 0.01) –
Adults Besson (2010)[31] DLW 14 days RPAQ TEE(Q) – TEE(DLW) – Spearman r = 0.67 (P < 0.0001) MD (95 % LoA) = −3451.9 ± 2025.1 kJ/day (P < 0.05)
      PAEE(Q) – PAEE(DLW) – Spearman r = 0.39 (P = 0.0004) MD (95 % LoA) = −12.9 ± 23.9 kJ/day (P < 0.05)
   Acc + HR (Actiheart) 11 days   VPA(Q) – VPA(Acc + HR) >6 METs Spearman r = 0.70 (P < 0.0001) MD (95 % LoA) = 0.2 ± 0.4 h/day
      MPA(Q) – MPA(Acc + HR) 3.6-6 METs – MD (95 % LoA) = −0.8 ± 1.0 h/day
      Light PA(Q) – light PA(Acc + HR) 2-3.5 METs – MD (95 % LoA) = −0.1 ± 2.4 h/day
      Sedentary time(Q) – sedentary time(Acc + HR) <2 METs Spearman r = 0.27 (P = 0.06) MD (95 % LoA) = 0.7 ± 2.8 h/day
Adults Chasan-Taber (2004)[32] Acc (ActiGraph) 7 days PPAQ Total activity(Q) – Swartz cut point min/day(Acc) ≥3 METs, ≥574 counts/min Spearman r = 0.32 –
      Total activity(Q) – Hendelman cut point min/day(Acc) ≥3 METs, ≥191 counts/min Spearman r = 0.43 –
      Total activity(Q) – Freedson cut point min/day(Acc) ≥3 METs, ≥1952 counts/min Spearman r = 0.08 –
      Total activity(Q) – mean counts/min(Acc) – Spearman r = 0.27 –
      Sedentary(Q) – Swartz cut point min/day(Acc) <1.5 METs Spearman r = −0.17 –
      Sedentary(Q) – Hendelman cut point min/day(Acc) <1.5 METs Spearman r = −0.34 –
      Sedentary(Q) – Freedson cut point min/day(Acc) <1.5 METs Spearman r = 0.12 –
      Sedentary(Q) – mean counts/min(Acc) – Spearman r = −0.10 –
Adults Chinapaw (2009)[33] Acc (ActiGraph) 14 days AQuAA Adolescents: AQuAA score(Q) – counts/min(Acc) ≥ 2 METs, ≥699 counts/min Spearman r = 0.13 –
      Adults: AQuAA score(Q) – counts/min(Acc) ≥ 2 METs, ≥699 counts/min Spearman r = −0.16 –
      Adolescents: sedentary(Q) – counts/min(Acc) < 2 METs, <699 counts/min Spearman r = 0.23 –
      Adults: sedentary(Q) – counts/min(Acc) < 2 METs, <699 counts/min Spearman r = 0.15 –
Adults Craig (2003)[34] Acc (ActiGraph) 7 days IPAQ Long form: total PA(Q) – total counts(Acc) – Pooled Spearman r (95 % CI) = 0.33 (0.26;0.39), range: -0.27 - 0.61 –
      Short form: total PA(Q) – total counts(Acc) – Pooled Spearman r (95 % CI) = 0.30 (0.23;0.36), range: -0.12 - 0.57 –
Adults Fjeldsoe (2009)[35] Acc (ActiGraph) 7 days AWAS Total activity(Q) – total activity(Acc) ≥100 counts/min Spearman r = 0.13 (P = 0.24) –
      HEPA(Q) – Freedson cut point min/week(Acc) – Spearman r = 0.28 (P = 0.01) –
      HEPA(Q) – Swartz cut point min/week(Acc) – Spearman r = 0.06 (P = 0.64) –
      Sitting(Q) – sitting(Acc) <100 counts/min Spearman r = 0.32 (P = 0.006) –
Adults Friedenreich (2006)[36] Acc (ActiGraph) 4x 7 days PYTPAQ Total MET-hr/week(Q) – total MET-hr/week(Acc) – Spearman r = 0.26 (P < 0.05), ICC (95 % CI) = 0.18 (0.03;0.32) –
Adults Kurtze (2007)[37] Acc (ActiReg) 7 days HUNT2 Hard activity(Q) – EE(Acc) – Spearman r = 0.11 –
      Hard activity(Q) – PAL(Acc) – Spearman r = 0.16 –
      Light activity(Q) – EE(Acc) – Spearman r = 0.21 (P < 0.05) –
      Light activity(Q) – PAL(Acc) – Spearman r = 0.08 –
      Occupational activity(Q) – EE(Acc) – Spearman r = 0.39 (P < 0.01) –
      Occupational activity(Q) – PAL(Acc) – Spearman r = 0.38 (P < 0.01) –
Adults Kurtze (2008)[38] Acc (ActiReg) 7 days HUNT1 Summary index(Q) – EE(Acc) – Spearman r = 0.03 –
      Summary index(Q) – PAL(Acc) – Spearman r = 0.07 –
      Summary index(Q) – MET-min/day(Acc) – Spearman r = 0.07 –
Adults Lowther (1999)[39] Acc (Caltrac) 4 days SPAQ Total mins(Q) – total kcal(Acc) – r = 0.1294, corrected for confounding: r = 0.52 (P < 0.05) –
Adults Mäder (2006)[40] Acc (ActiGraph) 7 days SHS97 Sweat episodes/week(Q) – total counts/min(Acc) – Spearman r = 0.23 –
     HEPA99 – – – –
     IPAQ Total MET-min/week(Q) – total counts/min(Acc) – Spearman r = 0.39 (P < 0.05) –
      Sitting(Q) – sitting(Acc) <100 counts/min Spearman r = 0.22 –
     OIMQ Total MET-min/week(Q) – total counts/min(Acc) – Spearman r = 0.44 (P < 0.05) –
Adults Meriwether (2006)[41] Acc (MTI) 14 days PAAT VPA min/week(Q) – VPA min/week(Acc) ≥5 METs, ≥5725 counts/min Spearman r = 0.380 (P < 0.01) –
      MVPA min/week(Q) – MVPA min/week(Acc) ≥5 METs, ≥1952 counts/min Spearman r = 0.392 (P < 0.01) –
      MPA min/week(Q) – MPA min/week(Acc) 3-4.9 METs, 1952–5724 counts/min Spearman r = 0.392 (P < 0.01) –
Adults Reis (2005)[42] Acc (ActiGraph) 7 days OPAQ Total hr/week(Q) – VPA(Acc) ≥5725 counts/min Spearman r = −0.02 –
      Total hr/week(Q) – MPA(Acc) 1952-5724 counts/min Spearman r = 0.12 –
      Total hr/week(Q) – light activity(Acc) <1952 counts/min Spearman r = 0.22 –
      Sedentary(Q) – light activity(Acc) <1952 counts/min Spearman r = −0.20 –
Adults Rosenberg (2010)[43] Acc (ActiGraph) 7 days SBQ Female: total sedentary hr/week(Q) – total sedentary counts(Acc) <100 counts/min Partial r = 0.10 (P = 0.07) –
      Male: total sedentary hr/week(Q) – total sedentary counts(Acc) <100 counts/min Partial r = −0.01 (P = 0.81) –
Adults Sobngwi (2001)[44] Acc (Caltrac) 1 day SSAAQ Female: total METs(Q) – total METs(Acc) – r = 0.74 (P < 0.01) –
      Male: total METs(Q) – total METs(Acc) – r = 0.60 (P < 0.01) –
   HR (Polar) 1 day   Urban female: total METs(Q) – total activity(HR) – r = 0.63 (P < 0.01) –
      Rural female: total METs(Q) – total activity(HR) – r = 0.41 (P < 0.05) –
      Urban male: total METs(Q) – total activity(HR) – r = 0.54 (P < 0.05) –
      Rural male: total METs(Q) – total activity(HR) – r = 0.59 (P < 0.01) –
Adults Timperio (2003)[45] Acc (ActiGraph) 7 days 1WPAR Men: total min/day(Q) – total min/day(Acc) ≥3 METs, ≥1952 counts/min Spearman r = 0.29 (P < 0.05) –
      Women: total min/day(Q) – total min/day(Acc) ≥3 METs, ≥1952 counts/min Spearman r = 0.25 (P < 0.05) –
Adults Wareham (2002)[46] HR (Polar) 4x 4 days EPAQ2 Total MET-hr/week(Q) – EE(HR) – Pearson partial r = 0.28 (P < 0.001) –
      TV time(Q) – EE(HR) – Pearson partial r = −0.07 –
Adults Wareham (2003)[47] HR (Polar) 4x 4 days EPAQ-s Physical activity index(Q) – DayPAR(HR) – P for trend = 0.003 –
      Total hr/week(Q) – DayPAR(HR) – r = 0.04 (P = 0.59) –
Adults Yore (2007)[48] Acc (ActiGraph) 7 days BRFSS PAQ VPA min/week(Q1) – VPA min/week(Acc) ≥5999 counts/min Pearson r = 0.52 –
      VPA min/week(Q2) – VPA min/week(Acc) ≥5999 counts/min Pearson r = 0.54 –
      VPA min/week(Q3) – VPA min/week(Acc) ≥5999 counts/min Pearson r = 0.63 –
      MPA min/week(Q1) – MPA min/week(Acc) 2020-5998 counts/min Pearson r = 0.27 –
      MPA min/week(Q2) – MPA min/week(Acc) 2020-5998 counts/min Pearson r = 0.20 –
      MPA min/week(Q3) – MPA min/week(Acc) 2020-5998 counts/min Pearson r = 0.16 –
Elderly Yasunaga (2007)[49] Acc (Kenz Lifecorder) 1 month PAQ-EJ PAQ-EJ score(Q) – MET-min/day(Acc) – Spearman r = 0.41 (P < 0.05) –
        Median Spearman r = 0.25 (youth: 0.22, adults: 0.27, elderly: 0.41)  
        Median Pearson r = 0.41 (youth: 0.41, adults: 0.28, elderly: –)  
  1. Q1 = first completed questionnaire, Q2 = second completed questionnaire, Q3 = third completed questionnaire, r = correlation coefficient (rho), CI = Confidence Interval (lower;upper), κ = kappa (i.e. Cohen weighted kappa unless specified otherwise), LoA = Limits of Agreement, MD = Mean Difference, – = not stated.
  2. Acc = Accelerometry [NB: ActiGraph (Model 7164) is successor of preceding accelerometer by MTI, formerly CSA]. Accelerometer names as used in the respective papers.
  3. Ainsworth (2000): 3 point summary index = 3 domains: sports/exercise, occupation, active living habits. 4 point summary index = all 4 domains: sports/exercise, occupation, active living habits, housework/caregiving.
  4. Craig (2003): Pooled Spearman = pooled results from data of 22 studies examining the IPAQ long form and 23 studies examining the short form.
  5. Dwyer (2011): Levels 1–2 = stationary, level 3 = moving slowly, level 4 = moving at a medium or moderate pace, level 5 = moving at a fast pace.
  6. Economos (2010): Moderate-high METs = 3–6 METs. High METs = ≥6 METs.
  7. Fjeldsoe (2009): Total activity includes light-, moderate-, and vigorous-intensity activities. HEPA = Health Enhancing Physical Activity: brisk walking and moderate- and vigorous activities from the planned activity and transport domains.
  8. Kurtze (2007): EE = Energy Expenditure in MJ/day. PAL = total EE divided by basal metabolic rate (BMR). Light activity = no sweating or being out of breath. Hard activity = sweating/out of breath.
  9. Kurtze (2008): EE = Energy Expenditure in MJ/day. PAL = total EE divided by basal metabolic rate (BMR).
  10. Lowther (1999): Initial r = 0.1294, but after correction for less reliable high data (occupational walking data, extreme data for 4 participants) the correlation improved to 0.52.
  11. Mäder (2006): IPAQ - Total MET-min/week = MET-min/week for total activity excluding sitting. OIMQ - Total MET-min/week = MET-min/week for total activity, i.e. moderate and vigorous activities.
  12. Martinez-Gomez (2010): Counts/mov = counts adjusted by movement time over the recess time. MD = mean difference between the mean times spent at MVPA by the two instruments. Kappa = agreement between the two instruments among tertiles of total PA.
  13. Reis (2005): ActiGraph only worn during occupational hours. Sedentary = sitting or standing activities.
  14. Ridley (2001): CDPAQ-HC = hard copy of CDPAQ. MVPA = Moderate-to-Vigorous Physical Activity. Total compendium METs = compendium values to derive total METs due to reported problems associated with children's perception of intensity (Compendium of physical activities: classification of energy costs of human physical activities. Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ, Sallis JF, Paffenbarger RS Jr. Med Sci Sports Exerc. 1993 Jan;25(1):71–80).
  15. Rosenberg (2010): Partial r = partial correlation, adjusted for age, marital status, white or nonwhite ethnicity, number of children, and highest level of education.
  16. Sobngwi (2001): Total activity by Heart Rate monitoring is defined as variability in heart rate measured as area under the minute-to-minute heart rate curve and above individual resting heart rate.
  17. Telford (2004): Validity results for total PA minutes for 5 to 6 or 10 to 12 year old children in parental proxy-reports or self-administered questionnaires.
  18. Timperio (2003): Total activity in min/day is specified as ≥3 METs.
  19. Treuth (2003): GAQ score = MET weighted mean score of 18 more reliable, and more frequently performed, activities. Activitygram score = average intensity/min over 3 day period. Other sedentary = sedentary activities excluding TV watching. The scores are an average of the two days administrations.
  20. Treuth (2005): Fels PAQ score = mean Fels PAQ score (total activity) of both administrations of the PAQ. Counts/min = mean counts/min. Elementary = elementary school. Middle = middle school. High = high school.
  21. Wareham (2002): Subject wore the HR monitor 4x four days across one year. EE = Energy Expenditure in kJ/hr. TV time = hours per week watching television and videos. Partial correlation coefficient is adjusted for age and sex.
  22. Wareham (2003): Subject wore the HR monitor 4x four days across one year. Physical activity index = combined index for the four-level classification of self-reported occupational activity and four-level categorisation of time spent in cycling and other physical exercise. DayPAR = Physical Activity Ratio calculated as the ratio of daytime energy expenditure to resting energy expenditure. P for trend = P for positive trend of the association between DayPAR (measured by calibrated HR data) over four categories of physical activity (i.e. inactive, moderately inactive, moderately active, active) estimated from the EPAQ.
  23. Welk (2007): PA bouts = number of sessions of physical activity performed during the week. Total MVPA mins = total minutes in moderate to vigorous physical activity performed during the previous day. Cut point used is Freedson age-based cut point, calculated as METs = 2.757 + (0.0015*counts per minute) - (0.0896*age[yr]) - (0.000038*counts per minute*age[yr]). Correlation = group-level correlation. No correlation coefficient specified.
  24. Yasunaga (2007): PAQ-EJ score = MET score in MET-hr/week, calculated as number of days*time*intensity weight.
Back to article page