Comparison of the ActiGraph accelerometer and Bouchard diary to estimate energy expenditure in Spanish adolescents.

INTRODUCTION

The aim of this study was to evaluate the agreement between the ActiGraph accelerometer and the Bouchard diary to estimate energy expenditure (EE) in a Spanish adolescent population.

METHODS

Sixty-one Spanish adolescents, aged 12-16 years, were recruited for this study. The Bouchard diary and the ActiGraph were administered for 3 consecutive days. EE estimated by the Bouchard diary was calculated using 2 different MET category values (BD-Bouchard and BD-Bratteby). EE estimated by the ActiGraph was calculated using 3 predictive equations (AC-Trost, AC-Freedson, and AC-Ekelund). Participants with complete 24 h data were also analyzed to control the possible loss of accuracy. Agreement was examined by Pearson and concordance correlations, paired t-test, and Bland-Altman method.

RESULTS

Thirty-seven adolescents were included in the analyses with complete 72 h. Relationships between EE calculated by the Bouchard diary and the ActiGraph were high (ranged: r = 0.61-0.78). Concordance correlations were moderate (r c = 0.60) by BD-Bratteby and AC-Trost, and low using BD-Bratteby and AC-Ekelund (r c = 0.15). EE estimated by BD-Bratteby and AC-Trost also showed no significant differences (P> 0.05) and absolutely agree 0.0 +/- 5.0 MJ/d (95% confidence interval: +/- 0.90 MJ/d), but with wide limits of agreement (+/- 9.80 MJ/d). Relationships were higher and differences were smaller in the 72 h sample than in the 24 h sample.

CONCLUSIONS

The Bouchard diary and the ActiGraph showed high relationships, moderate concordance, and large differences to estimate EE in Spanish adolescents. Advantages, disadvantages, and agreements between both instruments must be taking into consideration for health-related research.