Skip to main content
SpringerLink
Log in

A Review of Statistical Analyses on Physical Activity Data Collected from Accelerometers

  • Published:
Statistics in Biosciences Aims and scope Submit manuscript

Abstract

Studies for the associations between physical activity and disease risk have been supported by newly developed wearable accelerometer-based devices. These devices record raw activity/movement information in real time on a second-by-second basis and the data can be converted to a variety of summary metrics, such as energy expenditure, sedentary time and moderate-vigorous intensity physical activity. Here we review some of the methods used to analyze the accelerometer data and the R packages that can generate activity related variables from raw data. We also discuss longitudinal data and functional data approaches to perform analyses for various research purposes.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Ainsworth BE, Caspersen CJ, Matthews CE, Mâsse LC, Baranowski T, Zhu W (2012) Recommendations to improve the accuracy of estimates of physical activity derived from self report. J Phys Act Health 9:S76–S84

    Article  Google Scholar 

  2. Bai J, He B, Shou H, Zipunnikov V, Glass TA, Crainiceanu CM (2014) Normalization and extraction of interpretable metrics from raw accelerometry data. Biostatistics 15:102–116

    Article  Google Scholar 

  3. Bai J, Di C, Xiao L, Evenson KR, LaCroix AZ, Crainiceanu CM, Buchner DM (2016) An activity index for raw accelerometry data and its comparison with other activity metrics. PLoS ONE 11:e0160644

    Article  Google Scholar 

  4. Bai J, Sun Y, Schrack JA, Crainiceanu CM, Wang M-C (2018) A two-stage model for wearable device data. Biometrics 74:744–752

    Article  MathSciNet  MATH  Google Scholar 

  5. Breslow NE, Clayton DG (1993) Approximate inference in generalized linear mixed models. J Am Stat Assoc 88:9–25

    MATH  Google Scholar 

  6. Brønd JC, Arvidsson D (2015) Sampling frequency affects the processing of actigraph raw acceleration data to activity counts. J Appl Physiol 120:362–369

    Article  Google Scholar 

  7. Butera NM, Li S, Evenson KR, Di C, Buchner DM, LaMonte MJ, LaCroix AZ, Herring A (2018) Hot deck multiple imputation for handling missing accelerometer data. Stat Biosci 1–27

  8. Carroll RJ, Ruppert D, Stefanski LA, Crainiceanu CM (2006) Measurement error in nonlinear models: a modern perspective, 2nd edn. Chapman and Hall, London

    Book  MATH  Google Scholar 

  9. Catellier D, Hannan P, Murray D, Addy C, Conway T, Yang S, Rice J (2005) Imputation of missing data when measuring physical activity by accelerometry. Med Sci Sports Exerc 37:S555

    Article  Google Scholar 

  10. Choi L, Liu Z, Matthews CE, Buchowski MS (2011) PhysicalActivity: process physical activity accelerometer data. R package version 0.1-1

  11. Colley RC, Garriguet D, Janssen I, Craig CL, Clarke J, Tremblay MS (2011) Physical activity of canadian adults: accelerometer results from the 2007 to 2009 canadian health measures survey. Health Rep 22:7

    Google Scholar 

  12. Csizmadi I, Neilson HK, Kopciuk KA, Khandwala F, Liu A, Friedenreich CM, Yasui Y, Rabasa-Lhoret R, Bryant HE, Lau DC, Robson PJ (2014) The sedentary time and activity reporting questionnaire (STAR-Q): reliability and validity against doubly labeled water and 7-day activity diaries. Am J Epidemiol 180:424–435

    Article  Google Scholar 

  13. de Almeida Mendes M, da Silva IC, Ramires VV, Reichert FF, Martins RC, Tomasi E (2018) Calibration of raw accelerometer data to measure physical activity: a systematic review. Gait Posture 61:98–110

    Article  Google Scholar 

  14. Domelen DRV (2015) Accelerometry: functions for processing minute-to-minute accelerometer data. R package version 2.2.5

  15. Domelen DRV, Pittard WS, Harris TB (2014) nhanesaccel: process accelerometer data from NHANES 2003–2006. R package version 2.1.1

  16. Ferrari S, Cribari-Neto F (2004) Beta regression for modelling rates and proportions. J Appl Stat 31:799–815

    Article  MathSciNet  MATH  Google Scholar 

  17. Fitzmaurice GM, Laird NM, Ware JH (2004) Applied longitudinal analysis. Wiley, Hoboken

    MATH  Google Scholar 

  18. Freedson PS, Melanson E, Sirard JR (1998) Calibration of the computer science and applications, inc. accelerometer. Med Sci Sports Exerc 30:777–781

    Article  Google Scholar 

  19. Freedson PS, Bowles HR, Troiano R, Haskell W (2012) Assessment of physical activity using wearable monitors: recommendations for monitor calibration and use in the field. Med Sci Sports Exerc 44:S1–S4

    Article  Google Scholar 

  20. Geraci M (2014) pawacc: physical activity with accelerometers. R package version 1.2.1

  21. Goldsmith J, Zipunnikov V, Schrack J (2015) Generalized multilevel function-on-scalar regression and principal component analysis. Biometrics 71:344–353

    Article  MathSciNet  MATH  Google Scholar 

  22. Goldsmith J, Liu X, Jacobson J, Rundle A (2016) New insights into activity patterns in children, found using functional data analyses. Med Sci Sports Exerc 48:1723–1729

    Article  Google Scholar 

  23. Goldstein H, Rasbash J (1996) Improved approximations for multilevel models with binary responses. J R Stat Soc 159:505–513

    Article  MathSciNet  MATH  Google Scholar 

  24. Gruen ME, Alfaro-Córdoba M, Thomson AE, Worth AC, Staicu A-M, Lascelles BDX (2017) The use of functional data analysis to evaluate activity in a spontaneous model of degenerative joint disease associated pain in cats. PloS ONE 12:e0169576

    Article  Google Scholar 

  25. Harrington DM, Welk GJ, Donnelly AE (2011) Validation of met estimates and step measurement using the activpal physical activity logger. J Sports Sci 29:627–633

    Article  Google Scholar 

  26. Howe CA, Staudenmayer JW, Freedson PS (2009) Accelerometer prediction of energy expenditure: vector magnitude versus vertical axis. Med Sci Sports Exerc 41:2199–206

    Article  Google Scholar 

  27. Keadle SK, Sampson J, Li H, Lyden K, Matthews CE, Carroll RJ (2017) An evaluation of accelerometer-derived metrics to assess daily behavioral patterns. Med Sci Sports Exerc 49:54–63

    Article  Google Scholar 

  28. Kozey-Keadle S, Libertine A, Lyden K, Staudenmayer J, Freedson PS (2014) Changes in sedentary time and spontaneous physical activity in response to an exercise training and/or lifestyle intervention. J Phys Act Health 11:1324–1333

    Article  Google Scholar 

  29. Kozey-Keadle S, Lyden K, Staudenmayer J, Hickey A, Viskochil R, Braun B, Freedson PS (2014) The independent and combined effects of exercise training and reducing sedentary behavior on cardiometabolic risk factors. Appl Physiol Nutr Metab 39:770–780

    Article  Google Scholar 

  30. Li H, Staudenmayer J, Carroll RJ (2014) Hierarchical functional data with mixed continuous and binary measurements. Biometrics 70:802–811

    Article  MathSciNet  MATH  Google Scholar 

  31. Li H, Kozey-Keadle S, Staudenmayer J, Assaad H, Huang J, Carroll RJ (2015) Methods to assess an exercise intervention trial based on three-level functional data. Biostatistics 16:754–771

    Article  MathSciNet  Google Scholar 

  32. Li H, Zhang Y, Carroll RJ, Keadle SK, Sampson JN, Matthews CE (2017) A joint modeling and estimation method for multivariate longitudinal data with mixed types of responses to analyze physical activity data generated by accelerometers. Stat Med 36:4028–4040

    Article  MathSciNet  Google Scholar 

  33. Li H, Staudenmayer J, Wang T, Keadle SK, Carroll RJ (2018) Three-part joint modeling methods for complex functional data mixed with zero-and-one-inflated proportions and zero-inflated continuous outcomes with skewness. Stat Med 37:611–626

    Article  MathSciNet  Google Scholar 

  34. Little RJA (1995) Modeling the drop-out mechanism in repeated-measures studies. J Am Stat Assoc 90:1112–1121

    Article  MathSciNet  MATH  Google Scholar 

  35. Liu C, Rubin DB (1994) The ECME algorithm: a simple extension of EM and ECM with faster monotone convergence. Biometrika 81:633–648

    Article  MathSciNet  MATH  Google Scholar 

  36. Matthews CE, Keadle SK, Moore SC, Schoeller DS, Carroll RJ, Troiano RP, Sampson JN (2018) Measurement of active and sedentary behavior in context of large epidemiologic studies. Med Sci Sports Exerc 50:266–276

    Article  Google Scholar 

  37. Molenberghs G, Verbeke G (2005) Models for discrete longitudinal data. Springer, New York

    MATH  Google Scholar 

  38. Morris JS, Carroll RJ (2006) Wavelet-based functional mixed models. J R Stat Soc Ser B 68:179–199

    Article  MathSciNet  MATH  Google Scholar 

  39. Neilson HK, Ullman R, Robson PJ, Friedenreich CM, Csizmadi I (2013) Cognitive testing of the STAR-Q: insights in activity and sedentary time reporting. J Phys Act Health 10:379–389

    Article  Google Scholar 

  40. Physical Activities Guidelines Advisory Committee and others (2008) Physical activity guidelines advisory committee report. US Department of Health and Human Services, Washington, DC

  41. R Core Team (2018). R: A language and environment for statistical computing. R foundation for statistical computing, Vienna, Austria

  42. Ramsay JO, Silverman BW (2005) Functional data analysis. Springer, New York

    Book  MATH  Google Scholar 

  43. Ruppert D, Wand MP, Carroll RJ (2003) Semiparametric regression. Cambridge University Press, Cambridge

    Book  MATH  Google Scholar 

  44. Schafer JL (1998). Some improved procedures for linear mixed models. Technical report, The Methodological Center, The Pennsylvania State University

  45. Schrack JA, Zipunnikov V, Goldsmith J, Bai J, Simonsick EM, Crainiceanu C, Ferrucci L (2014) Assessing the “physical cliff”: detailed quantification of age-related differences in daily patterns of physical activity. J Gerontol 69:973–979

    Article  Google Scholar 

  46. Shou H, Zipunnikov V, Crainiceanu CM, Greven S (2015) Structured functional principal component analysis. Biometrics 71:247–257

    Article  MathSciNet  MATH  Google Scholar 

  47. Staudenmayer J, Zhu W, Catellier DJ (2012) Statistical considerations in the analysis of accelerometry-based activity monitor data. Med Sci Sports Exerc 44:S61–S67

    Article  Google Scholar 

  48. Troiano R, Berrigan D, Dodd K, Mâsse L, Tilert T, McDowell M (2008) Physical activity in the united states measured by accelerometer. Med Sci Sports Exerc 40:181–188

    Article  Google Scholar 

  49. van Hees VT, Fang Z, Zhao JH, Sabia S (2016) GGIR: Raw accelerometer data analysis. R package version 1.2.2

  50. Xiao L, Huang L, Schrack J, Ferrucci L, Zipunnikov V, Crainiceanu C (2015) Quantifying the lifetime circadian rhythm of physical activity: a covariate-dependent functional approach. Biostatistics 16:352–367

    Article  MathSciNet  Google Scholar 

  51. Zhang Y, Li H, Kozey-Keadle S, Matthews CE, Carroll RJ (2015) PAactivPAL: summarize daily physical activity from ’activPAL’ accelerometer data. R package version 1

  52. Zhang Y, Yavari M, Haennel B, Li H (2016) PASenseWear: summarize daily physical activity from ’SenseWear’ accelerometer data. R package version 1

Download references

Acknowledgements

Zhang and Li were supported by the Natural Sciences and Engineering Research Council of Canada (RGPIN-2015-04409). Keadle was supported by a National Cancer Institute grant (R01-CA121005). Carroll was supported by a grant from the National Cancer Institute (U01-CA057030).

Author information

Authors and Affiliations

  1. Department of Mathematics and Statistics, University of Calgary, Calgary, AB, T2N 1N4, Canada

    Yukun Zhang & Haocheng Li

  2. Kinesiology Department, California Polytechnic State University, San Luis Obispo, CA, 93407, USA

    Sarah Kozey Keadle

  3. National Cancer Institute, Bethesda, MD, 20892, USA

    Charles E. Matthews

  4. Department of Statistics, Texas A&M University, College Station, TX, 77843, USA

    Raymond J. Carroll

  5. Department of Mathematics and Statistics, University of Technology Sydney, Ultimo, NSW, 2007, Australia

    Raymond J. Carroll

Authors
  1. Yukun Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Haocheng Li
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Sarah Kozey Keadle
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Charles E. Matthews
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Raymond J. Carroll
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Haocheng Li.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest to report.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, Y., Li, H., Keadle, S.K. et al. A Review of Statistical Analyses on Physical Activity Data Collected from Accelerometers. Stat Biosci 11, 465–476 (2019). https://doi.org/10.1007/s12561-019-09250-6

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12561-019-09250-6

Keywords

Search

Navigation