Use of accelerometry to classify activity beneficial to bone in premenopausal women.

PURPOSE

The aims of this study were to quantify the relation between ground reaction force (GRF) and peak acceleration from hip- and wrist-worn accelerometers and determine peak acceleration cut-points associated with a loading rate previously demonstrated as beneficial to bone (43 body weights (BW)·s⁻¹) in premenopausal women.

METHODS

Forty-seven premenopausal women (age, 39.2 ± 5.6 yr; mass, 65.9 ± 11.0 kg; height, 1.67 ± 0.06 m) performed walking (slow, fast, and with bag), floor sweeping, running (slow and fast), jumping (low, <5 cm; high, >5 cm), and box drop (20 cm) trials. Peak accelerations were sampled at 100 Hz by GENEActiv and ActiGraph GT3X+ accelerometers (ActiGraph LLC, Pensacola, FL) worn at the hip (vertical and resultant) and the wrist (resultant). A force plate (960 Hz, AMTI) was used to assess peak vertical GRF and peak loading rate for eight steps per activity. Receiver operating characteristic curves were used to determine the optimal peak acceleration cut-points associated with a loading rate of 43 BW·s⁻¹ in 37 participants, and these cut-points were cross-validated in the remaining 10 participants.

RESULTS

For all activities combined, peak accelerations were positively and significantly (P < 0.001) correlated with peak vertical GRF (hip r > 0.8, wrist r > 0.7) and peak loading rate (hip r > 0.7, wrist r > 0.57). Irrespective of monitor type and wear site, peak acceleration discriminated between loading rates above and below 43 BW·s⁻¹ with high levels of accuracy (area under the curve >0.92, P < 0.001). Overall classification agreement was >85% for both monitors worn at either the wrist or hip in the cross-validation sample.

CONCLUSION

GENEActiv and ActiGraph GT3X+ accelerometers worn at the wrist or hip can be used as an unobtrusive tool to identify the occurrence of loading rates likely beneficial to bone in premenopausal women during their daily activity.