PTSystematic ReviewJournal of neuroengineering and rehabilitation2022

Criterion validity of ActiGraph monitoring devices for step counting and distance measurement in adults and older adults: a systematic review.

Armelle-Myriane Ngueleu, Corentin Barthod, Krista Lynn Best and 3 others

PMID 36253787

WHAT IT FOUND

ActiGraph step counts were most accurate with GT3X+/wGT3X+ or wGT3X-BT on the hip at self-selected comfort walking speeds.

Slow walking below 0.9 m/s was poor, though a low frequency extension filter helped GT3X+. Wrist wear and older devices gave poor counts. Distance evidence was limited.

Key findings

01Across 21 studies of 637 adults and older adults, GT3X+/wGT3X+ and wGT3X-BT matched reference step counts better than 7164, GT1M, and wGTX+.

02Step counting was not valid below 0.9 m/s, and hip-worn wGT3X-BT was less valid at 0.5 m/s.

03Wrist-worn ActiGraph step counts differed significantly from reference counts, while hip-worn devices generally showed better validity.

STILL TO COME

How it was doneWhat they foundWhat it means for PTs

Read the rest of this summary

You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.

Already have one?

What it does not show

The review included only healthy adults and older adults. It does not show how valid the devices are for people with stroke or other walking impairments. Most studies were short indoor tests, from 2 to 15 min, and only two studies were outdoor. Small errors over short tests may become larger over a day. Study samples were small, so results cannot be generalized. Many studies did not report signal processing, sensitivity, dynamic range, or analysis algorithm. Device comparisons are difficult. Only one study assessed distance, so conclusions about distance are limited. Quality was mixed: on COSMIN, 11 articles were high-quality, 2 moderate-quality, and 8 low-quality.

The easy way to misread this

Do not conclude that ActiGraph step counts are accurate for all walking assessments. The review found poor default validity below 0.9 m/s and for wrist-worn devices, and only one study tested distance.

Read it on PubMed →