Accelerometer-Derived Activity Phenotypes in Young Adults: a Latent Class Analysis

Int J Behav Med. 2018 Oct;25(5):558-568. doi: 10.1007/s12529-018-9721-4.

Abstract

Purpose: To identify "activity phenotypes" from accelerometer-derived activity characteristics among young adults.

Methods: Participants were young adults (n = 628, mean age, 22.1, SD 0.6) in the Raine Study in Western Australia. Sex-specific latent class analyses identified sub-groups using eight indicators derived from 7-day hip-worn Actigraph GT3X+ accelerometers: daily steps, total daily moderate-to-vigorous physical activity (MVPA), MVPA variation, MVPA intensity, MVPA bout duration, sedentary-to-light ratio, sedentary-to-light ratio variation, and sedentary bout duration.

Results: Five activity phenotypes were identified for women (n = 324) and men (n = 304). Activity phenotype 1 for both women (35%) and men (30%) represented average activity characteristics. Phenotype 2 for women (17%) and men (16%) was characterized by below average total activity and MVPA (10.6 and 16.7 min of MVPA/day, women and men respectively). Phenotype 3 for women (15%) and men (23%) was characterized by below average total physical activity, average MVPA (32.6 and 36.5 min/day), high sedentary-light ratio and long sedentary bouts. Phenotype 4 differed between women (29%) and men (18%) but both had low sedentary-to-light ratios and shorter sedentary bouts. Finally, phenotype 5 in both women (4%) and men (12%) was characterized by extreme MVPA metrics (81.3 and 96.1 min/day).

Conclusions: Five activity phenotypes were identified for each gender in this population of young adults which can help design targeted interventions to enhance or modulate activity phenotypes.

Keywords: Health outcomes; Longitudinal; Phenotypes; Physical activity; Sedentary.

MeSH terms

  • Actigraphy / instrumentation
  • Actigraphy / statistics & numerical data*
  • Exercise / psychology*
  • Female
  • Fitness Trackers / statistics & numerical data*
  • Health Status
  • Humans
  • Latent Class Analysis*
  • Male
  • Phenotype*
  • Sedentary Behavior
  • Wearable Electronic Devices
  • Western Australia
  • Young Adult