Health and Technology

, Volume 4, Issue 3, pp 273–287

Towards tailored physical activity health intervention: Predicting dropout participants

  • Xi Long
  • Marten Pijl
  • Steffen Pauws
  • Joyca Lacroix
  • Annelies H. C. Goris
  • Ronald M. Aarts
Original Paper


Physical activity is important for people’s health. The physical activity intervention program reported here includes daily wearing of an activity monitor to provide people with insight into their activity behavior. The activity monitor consists of a triaxial accelerometer, where measured accelerations are transformed to a physical activity level (PAL). The PAL data quantifies the level of the daily physical activity and reflects the daily energy expenditure of the wearer. In the program, coaches provide e-mail based intervention to motivate participants to increase their activity step-by-step within 12 weeks. However, a significant portion of participants (∼41%) failed to complete the program. This paper examines methods to predict participants who are at risk of dropping out of the program based on a classification task. This allows for a timely delivery of tailored interventions and motivating triggers to prevent stopping of the program. In particular, this paper proposes to combine the features extracted from participants’ personal information, their behaviors during the use of the device, the observed PAL data and the features extracted from the process of predicting future PAL data to classify dropouts and non-dropouts every week. Experiment results show that a k-nearest-neighbor classifier achieved a dropout and a non-dropout prediction accuracy of 66.4 ± 13.8% and 74.1 ± 7.3%, respectively.


Physical activity level Tailored intervention Triaxial accelerometer Dropout prediction 


  1. 1.
    Penedo FJ, Dahn JR. Exercise and well-being: review of mental and physical health benefits associated with physical activity. Curr Opin Psychiatry 2005;18(2):189–93.CrossRefGoogle Scholar
  2. 2.
    Pate RR, Pratt M, Blair SN, Haskell WL, Macera CA, Bouchard C, Buchner D, Ettinger W, Heath GW, King AC. Physical activity and public health: A recommendation from the centers for disease control and prevention and the American college of sports medicine. J Am Med Assoc 1995;273(5):402–7.CrossRefGoogle Scholar
  3. 3.
    Fox KR. The influence of physical activity on mental well-being. Public Health Nutr 1999;2(3a):411–8.CrossRefGoogle Scholar
  4. 4.
    Driver HS, Taylor SR. Exercise and sleep. Sleep Med Rev 2000;4(4):387–402.CrossRefGoogle Scholar
  5. 5.
    Ware LJ, Hurling R, Bataveljic O, Fairley BW, Hurst TL, Murray P, Rennie KL, Tomkins CE, Finn A, Cobain MR, Pearson DA, Foreyt JP. Rates and determinants of uptake and use of an internet physical and weight management program in office and manufacturing work sites in England: cohort study. J Med Internet Res 2008;10(4):e56.CrossRefGoogle Scholar
  6. 6.
    Goris AHC, Holmes R. The effect of a lifestyle activity intervention program on improving physical activity behavior of employees. In Proc. 3rd Int Conf Persuasive Technol (PERSUASIVE), Oulu, Finland. 2008.Google Scholar
  7. 7.
    Lacroix J, Saini P, Holmes R. The relationship between goal difficulty and performance in the context of a physical activity intervention program. In Proc 10th Conf MobileHCI, Amsterdam, Netherlands. 2008;415–8.Google Scholar
  8. 8.
    Plasqui G, Joosen AM, Kester AD, Goris AHC, Westerterp KR. Measuring free-living energy expenditure and physical activity with triaxial accelerometry. Obes Res 2005;13(8):1363–9.CrossRefGoogle Scholar
  9. 9.
    Hurling R, Catt M, Boni MD, Fairley BW, Hurst T, Murray P, Richardson A, Sodhi JS. Using internet and mobile phone technology to deliver an automated physical activity program. Randomized controlled trial. J Med Internet Res 2008;9(2):e7.CrossRefGoogle Scholar
  10. 10.
    Clarkson BP. Life patterns: structure from wearable sensor. Ph.D Thesis: MIT Media Lab;2002.Google Scholar
  11. 11.
    Long X, Yin B, Aarts RM. Single-accelerometer-based daily physical activity classification. In Proc 31st Ann Int Conf IEEE EMBS, Minneapolis, MN, 2009;6107–10.Google Scholar
  12. 12.
    Bouten C, Westerterp K, Verduin M, Janssen J. Assessment of energy expenditure for physical activity using a triaxial accelerometer. Med Sci Sports Exerc 2003;26(12):1516–23.Google Scholar
  13. 13.
    Bonomi AG, Plasqui G, Goris AHC, Westerterp KR. Estimation of free-living energy expenditure using a novel activity monitor designed to minimize obtrusiveness. Obes 2010;18(9):1845–51.CrossRefGoogle Scholar
  14. 14.
    Leenders NY, Sherman WM, Nagaraja HN, Kien CL. Evaluation of methods to assess physical activity in free-living conditions. Med Sci Sports Exerc 2001;33(7):1233–40.CrossRefGoogle Scholar
  15. 15.
    Marcus BH, Lewis BA, Williams DM, Dunsiger S, Jakicic JM, Whiteley JA, Albrecht AE, Napolitano MA, Bock BC, Tate DF, Sciamanna CN, Parisi AF. A comparison of internet and print-based physical activity interventions. Arch Intern Med 2007;167(9):944–9.CrossRefGoogle Scholar
  16. 16.
    van den Berg MH, Schoones JW, Vliet Vlieland TP. Internet-based physical activity interventions: a systematic review of the literature. J Med Internet Res 2007;9(3):e26.CrossRefGoogle Scholar
  17. 17.
    Adams J White M. Are activity promotion interventions based on the transtheoretical model effective? A critical review. Br J Sports Med 2003;37(2):106–14.CrossRefGoogle Scholar
  18. 18.
    Marcus BH, Bock BC, Pinto BM, Forsyth LH, Robeerts MB, Traficante RM. Efficacy of an individualized, motivationally-tailored physical activity intervention. Ann Behav Med 1998;20(3):174–80.CrossRefGoogle Scholar
  19. 19.
    Segerstahl K, Oinas-Kukkonen H. Distributed user experience in persuasive technology environments. In Proc. 2nd Int Conf Persuasive Technol (PERSUASIVE), Palo Alto, CA. 2007.Google Scholar
  20. 20.
    Xu W, Zhang M, Sawchuk AA, Sarrafzadeh M. Robust Human activity and sensor location co-recognition via sparse signal representation. IEEE Trans Biomed Eng 2012;59(11):3169–76.CrossRefGoogle Scholar
  21. 21.
    Khan AM, Lee YK, Kim TS. A triaxial accelerometer-based physical-activity recognition via augmented-signal features and a hierarchical recognizer. IEEE Trans Inf Technol Biomed 2010;14(5):1166–72.CrossRefGoogle Scholar
  22. 22.
    Zhu C, Sheng W. Realtime recognition of complex human daily activities using human motion and location data. IEEE Trans Biomed Eng 2012;59(9):2422–30.CrossRefGoogle Scholar
  23. 23.
    Mazilu S, Hardegger M, Zhu Z, Roggen D. Online detection of freezing of gait with smartphones and machine learning techniques. In Proc 6th Int Conf PervasiveHealth 2012:123–30.Google Scholar
  24. 24.
    Lustrek M, Kaluza B. Fall detection and activity recognition with machine learning. Informatica 2009;33:205–12.Google Scholar
  25. 25.
    Jin A, Yin B, Morren G, Duric H, Aarts RM. Performance evaluation of a tri-axial accelerometry-based respiration monitoring for ambient assisted living. In Proc. 31st Ann Int Conf IEEE EMBS, Minneapolis, MN, 2009;5677–80.Google Scholar
  26. 26.
    Nijsen TME, Aarts RM, Cluitmans PJM, Griep PAM. Time-frequency analysis of accelerometry data for detection of myoclonic seizures. IEEE Trans Inf Technol Biomed 2010;14(5):1197–203.CrossRefGoogle Scholar
  27. 27.
    Pentland A, Liu A. Modeling and prediction of human behavior. Neur Comp 1999;11:229–42.CrossRefGoogle Scholar
  28. 28.
    Yang M, Wong SCP, Coid J. The efficacy of violence prediction: A meta-analytic comparison of nine risk assessment tools. Psych Bullet 2010;136(5):740–67.CrossRefGoogle Scholar
  29. 29.
    Cheng S, Tom K, Thomas L, Pecht M. Wireless sensor system for prognostics and health management. IEEE Sensors J 2010;10(4):856–62.CrossRefGoogle Scholar
  30. 30.
    Long X, Pauws S, Pijl M, Lacroix J, Goris AHC, Aarts RM. Predicting daily physical activity in a lifestyle intervention program. In: Gottfried B, Aghajan H, editors. Behaviour Monitoring and Interpretation - Well-Being. The Netherlands: IOS Press; 2011, pp. 131–46.Google Scholar
  31. 31.
    Si XS, Hu CH, Yang JB, Zhou ZJ. A new prediction model based on belief rule base for system’s behavior prediction. IEEE Trans Fuzzy Syst 2011;19(4):636–51.CrossRefGoogle Scholar
  32. 32.
    Schwabacher M, Goebel K. A survey of artificial intelligence for prognostics. In Proc AAAI Fall Sym, Arlington, VA. 2007;107–14.Google Scholar
  33. 33.
    Niu G, Yang BS. Dempster-Shafer regression for multi-step-ahead time-series prediction towards data-driven machinery prognosis. Mech Syst Sig Process 2009;23:740–51.CrossRefGoogle Scholar
  34. 34.
    Japkowicz N, Stephen S. The class imbalance problem: A systematic study. Intell Data Anal 2002;6(5):429–49.MATHGoogle Scholar
  35. 35.
    Ling CX, Li C. Data mining for direct marketing: problems and solutions. In Proc 4th ACM SIGKDD Int Conf Knowl Disc Data Min 1998:73–9.Google Scholar
  36. 36.
    Zhang J, Mani I. kNN approach to unbalanced data distributions: a case study involving information extraction. In Proc ICML Workshop Learning from Imbalanced Datasets. Washington, DC. 2003.Google Scholar
  37. 37.
    Bradley AP. The use of area under the ROC curve in the evaluation of machine learning algorithms. Patt Recogn 1997;30(7):1145–59.CrossRefGoogle Scholar
  38. 38.
    Duda R, Hart P, Stork D. Pattern Classification, 2nd ed: Wiley;2001.Google Scholar
  39. 39.
    Cover T, Hart P. Nearest neighbor pattern classification. IEEE Trans Inf Theory 1967;13(1):21–7.CrossRefMATHGoogle Scholar
  40. 40.
    Hand DJ, Yu K.Idiot’s Bayes – not so stupid after all. Int Stat Review 2001;69(3):385–98.MATHGoogle Scholar
  41. 41.
    Bhatia N. Vandana. Survey of nearest neighbor techniques. Int J Comput Sci Inf Secur 2010;8(2):302–5.Google Scholar
  42. 42.
    Weiss GM, Learning with rare cases and small disjuncts. In Proc 12th Int Conf Mach Learn Lake Tahoe CA. 1995;558–65.Google Scholar
  43. 43.
    Kubat M, Matwin S. Addressing the curse of imbalanced datasets: One-sided sampling. In Proc 14th Int Conf Mach Learn 1997:179–86.Google Scholar
  44. 44.
    Pazzani M, Merz C, Murphy P, Ali K, Hume T, Brunk C. Reducing Misclassification Costs. In Proc 11th Int Conf Mach Learn 1994:217–25.Google Scholar
  45. 45.
    Yang J, Honavar V. Feature subset selection using a genetic algorithm. IEEE Trans Intell Syst App 1998;13(2):44–9.CrossRefGoogle Scholar
  46. 46.
    Bonomi AG, Plasqui G, Goris AHC, Westerterp KR. Improving assessment of daily energy expenditure by identifying types of physical activity with a single accelerometer. J Appl Physiol 2009;107(3):655–61.CrossRefGoogle Scholar
  47. 47.
    Shetty PS, Henry CJ, Black AE, Prentice AM. Energy requirements of adults: an update on basal metabolic rates (BMRs) and physical activity levels (PALs). Eur J Clin Nutr 1996;50(1):S1—23.Google Scholar
  48. 48.
    World Health Organization. Human energy requirements. Rome: Report of a Joint FAO/WHO/UNI Expert Consultation;2011.Google Scholar
  49. 49.
    Clements MP, Hendry DF. Forecasting economic time series. Cambridge: Cambridge University Press;1998.CrossRefGoogle Scholar
  50. 50.
    Helfenstein U. Box-Jenkins modelling in medical research. Stat Meth Med Res 1996;5(1):3–22.CrossRefGoogle Scholar
  51. 51.
    Peng JY, Aston JAD. The state space models toolbox for MATLAB. J Stat soft 2011;41(6):1–26.MATHGoogle Scholar
  52. 52.
    Box GEP, Jenkins GM. Time series analysis forecasting and control. San Francisco: Holden-Day;1976.MATHGoogle Scholar
  53. 53.
    Bowerman BL, O’Connell RT. Forecasting and time series: an applied approach. Belmont: Duxbury Press;1993.MATHGoogle Scholar
  54. 54.
    Harvey AC. Forecasting, structural time series models and the Kalman filter. Cambridge: Cambridge University Press;1989.Google Scholar
  55. 55.
    Schwarz G. Estimating the dimension of a model. Ann Stat 1978;6(2):416–64.CrossRefGoogle Scholar
  56. 56.
    Koza J. Genetic programming: on the programming of computers by means of natural selection: MIT Press;1992.Google Scholar
  57. 57.
    Abdi H, Williams LJ. Principal component analysis. WIREs: Comp Stat 2010;2(4):433–59.Google Scholar
  58. 58.
    Parzen E. On estimation of a probability density function and mode. Ann Math Stat 1962;333(3):1065–76.MathSciNetCrossRefGoogle Scholar
  59. 59.
    Hellinger E. Neue Begründung der Theorie quadratischer formen von unendlichvielen veränderlichen. J für die Reine und Angew Math 1909;136:210–71.MATHGoogle Scholar
  60. 60.
    Provost F, Fawcett T. Analysis and visualization of classifier performance: comparison under imprecise class and cost distribution. In Proc 3rd ACM SIGKDD 1997:43–48.Google Scholar
  61. 61.
    van Rijsbergen CJ. Information retrieval. London: Butterworths;1979.Google Scholar
  62. 62.
    Xie Y, Li X, Ngai EWT, Ying W. Customer churn prediction using improved balanced random forests. Exp Syst Appl 2009;36(3):5445–9.CrossRefGoogle Scholar
  63. 63.
    Joachims T. A support vector method for multivariate performance measures. In Proc 22nd Int Conf Mach Learn 2005:377–84.Google Scholar
  64. 64.
    King AC, Rejeski WJ, Buchner DM. Physical activity interventions targeting older adults: a critical review and recommendations. Am J Prev Med 1998;15(4):316–33.CrossRefGoogle Scholar

Copyright information

© IUPESM and Springer-Verlag Berlin Heidelberg 2014

Authors and Affiliations

  • Xi Long
    • 1
    • 2
  • Marten Pijl
    • 2
  • Steffen Pauws
    • 2
  • Joyca Lacroix
    • 2
  • Annelies H. C. Goris
    • 3
  • Ronald M. Aarts
    • 1
    • 2
  1. 1.Department of Electrical EngineeringEindhoven University of TechnologyEindhovenThe Netherlands
  2. 2.Philips ResearchEindhovenThe Netherlands
  3. 3.Philips DirectLifeAmsterdamThe Netherlands

Personalised recommendations