Limits...
Validation of the historical adulthood physical activity questionnaire (HAPAQ) against objective measurements of physical activity.

Besson H, Harwood CA, Ekelund U, Finucane FM, McDermott CJ, Shaw PJ, Wareham NJ - Int J Behav Nutr Phys Act (2010)

Bottom Line: The average PAEE from the 2 periods of objective measurements was compared to the self-reported data from the corresponding time periods in the questionnaire.Similarly, self-reported time spent in vigorous PA was significantly correlated with objective measurements of vigorous PA (Spearman r = 0.40; P < 0.001).HAPAQ demonstrates convergent validity for total PAEE and vigorous PA.

View Article: PubMed Central - HTML - PubMed

Affiliation: Medical Research Council Epidemiology Unit, IMS, Cambridge, UK. herve.besson@mrc-epid.cam.ac.uk.

ABSTRACT

Background: Lifetime physical activity energy expenditure (PAEE) is an important determinant of risk for many chronic diseases but remains challenging to measure. Previously reported historical physical activity (PA) questionnaires appear to be reliable, but their validity is less well established.

Methods: We sought to design and validate an historical adulthood PA questionnaire (HAPAQ) against objective PA measurements from the same individuals. We recruited from a population-based cohort in Cambridgeshire, UK, (Medical Research Council Ely Study) in whom PA measurements, using individually calibrated heart rate monitoring, had been obtained in the past, once between 1994 and 1996 and once between 2000 and 2002. 100 individuals from this cohort attended for interview. Historical PA within the domains of home, work, transport, sport and exercise was recalled using the questionnaire by asking closed questions repeated for several discrete time periods from the age of 20 years old to their current age. The average PAEE from the 2 periods of objective measurements was compared to the self-reported data from the corresponding time periods in the questionnaire.

Results: Significant correlations were observed between HAPAQ-derived and objectively measured total PAEE for both time periods (Spearman r = 0.44; P < 0.001). Similarly, self-reported time spent in vigorous PA was significantly correlated with objective measurements of vigorous PA (Spearman r = 0.40; P < 0.001).

Conclusions: HAPAQ demonstrates convergent validity for total PAEE and vigorous PA. This instrument will be useful for ranking individuals according to their past PA in studies of chronic disease aetiology, where activity may be an important underlying factor contributing to disease pathogenesis.

No MeSH data available.


Related in: MedlinePlus

Distribution of objective measurements of the percentage of time spent in vigorous physical activity (VPA) by the three levels of VPA measured using HAPAQ. Data are expressed as medians and inter-quartile ranges.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC2902409&req=5

Figure 5: Distribution of objective measurements of the percentage of time spent in vigorous physical activity (VPA) by the three levels of VPA measured using HAPAQ. Data are expressed as medians and inter-quartile ranges.

Mentions: Regarding time spent in VPA, estimations derived from HAPAQ correlated significantly with objective measurements (r = 0.40, p < 0.0001; ICC = 0.24, p = 0.008) (Table 2). HAPAQ tended to underestimate VPA (median bias -0.26, IQR = 1.59). The underestimation was very similar for periods 1 (median bias -0.21, IQR = 0.97) and period 2 (-0.16, IQR = 1.99). In figure 4, the Bland and Altman plots demonstrate that the mean bias between self-reported and objectively measured VPA is proportional to the objectively measured VPA, though the relative bias did not display the same significantly proportional error (data not shown). The distribution of objectively measured VPA by the self-reported VPA categories and the correlation between these measurements are shown in Figure 5. The weighted Kappa between categorized self-reported PA and the corresponding categorized objective measurement was very modest (0.13; p = 0.009).


Validation of the historical adulthood physical activity questionnaire (HAPAQ) against objective measurements of physical activity.

Besson H, Harwood CA, Ekelund U, Finucane FM, McDermott CJ, Shaw PJ, Wareham NJ - Int J Behav Nutr Phys Act (2010)

Distribution of objective measurements of the percentage of time spent in vigorous physical activity (VPA) by the three levels of VPA measured using HAPAQ. Data are expressed as medians and inter-quartile ranges.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC2902409&req=5

Figure 5: Distribution of objective measurements of the percentage of time spent in vigorous physical activity (VPA) by the three levels of VPA measured using HAPAQ. Data are expressed as medians and inter-quartile ranges.
Mentions: Regarding time spent in VPA, estimations derived from HAPAQ correlated significantly with objective measurements (r = 0.40, p < 0.0001; ICC = 0.24, p = 0.008) (Table 2). HAPAQ tended to underestimate VPA (median bias -0.26, IQR = 1.59). The underestimation was very similar for periods 1 (median bias -0.21, IQR = 0.97) and period 2 (-0.16, IQR = 1.99). In figure 4, the Bland and Altman plots demonstrate that the mean bias between self-reported and objectively measured VPA is proportional to the objectively measured VPA, though the relative bias did not display the same significantly proportional error (data not shown). The distribution of objectively measured VPA by the self-reported VPA categories and the correlation between these measurements are shown in Figure 5. The weighted Kappa between categorized self-reported PA and the corresponding categorized objective measurement was very modest (0.13; p = 0.009).

Bottom Line: The average PAEE from the 2 periods of objective measurements was compared to the self-reported data from the corresponding time periods in the questionnaire.Similarly, self-reported time spent in vigorous PA was significantly correlated with objective measurements of vigorous PA (Spearman r = 0.40; P < 0.001).HAPAQ demonstrates convergent validity for total PAEE and vigorous PA.

View Article: PubMed Central - HTML - PubMed

Affiliation: Medical Research Council Epidemiology Unit, IMS, Cambridge, UK. herve.besson@mrc-epid.cam.ac.uk.

ABSTRACT

Background: Lifetime physical activity energy expenditure (PAEE) is an important determinant of risk for many chronic diseases but remains challenging to measure. Previously reported historical physical activity (PA) questionnaires appear to be reliable, but their validity is less well established.

Methods: We sought to design and validate an historical adulthood PA questionnaire (HAPAQ) against objective PA measurements from the same individuals. We recruited from a population-based cohort in Cambridgeshire, UK, (Medical Research Council Ely Study) in whom PA measurements, using individually calibrated heart rate monitoring, had been obtained in the past, once between 1994 and 1996 and once between 2000 and 2002. 100 individuals from this cohort attended for interview. Historical PA within the domains of home, work, transport, sport and exercise was recalled using the questionnaire by asking closed questions repeated for several discrete time periods from the age of 20 years old to their current age. The average PAEE from the 2 periods of objective measurements was compared to the self-reported data from the corresponding time periods in the questionnaire.

Results: Significant correlations were observed between HAPAQ-derived and objectively measured total PAEE for both time periods (Spearman r = 0.44; P < 0.001). Similarly, self-reported time spent in vigorous PA was significantly correlated with objective measurements of vigorous PA (Spearman r = 0.40; P < 0.001).

Conclusions: HAPAQ demonstrates convergent validity for total PAEE and vigorous PA. This instrument will be useful for ranking individuals according to their past PA in studies of chronic disease aetiology, where activity may be an important underlying factor contributing to disease pathogenesis.

No MeSH data available.


Related in: MedlinePlus