A semiparametric model for wearable sensor-based physical activity monitoring data with informative device wear

Jaejoon Song, Michael D. Swartz, Kelley Pettee Gabriel, Karen Basen-Engquist

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Wearable sensors provide an exceptional opportunity in collecting real-Time behavioral data in free living conditions. However, wearable sensor data from observational studies often suffer from information bias, since participants' willingness to wear the monitoring devices may be associated with the underlying behavior of interest. The aim of this study was to introduce a semiparametric statistical approach for modeling wearable sensor-based physical activity monitoring data with informative device wear. Our simulation study indicated that estimates from the generalized estimating equations showed ignorable bias when device wear patterns were independent of the participants physical activity process, but incrementally more biased when the patterns of device non-wear times were increasingly associated with the physical activity process. The estimates from the proposed semiparametric modeling approach were unbiased both when the device wear patterns were (i) independent or (ii) dependent to the underlying physical activity process. We demonstrate an application of this method using data from the 2003-2004 National Health and Nutrition Examination Survey (N=4518), to examine gender differences in physical activity measured using accelerometers. The semiparametric model can be implemented using our R package acc, free software developed for reading, processing, simulating, visualizing, and analyzing accelerometer data, publicly available at the Comprehensive R Archive Network.

Original languageEnglish (US)
Pages (from-to)287-298
Number of pages12
JournalBiostatistics
Volume20
Issue number2
DOIs
StatePublished - Apr 1 2019

Fingerprint

Semiparametric Model
Monitoring
Sensor
Accelerometer
Gender Differences
Observational Study
Generalized Estimating Equations
Nutrition
Modeling
Estimate
Biased
Health
Physical activity
Semiparametric model
Simulation Study
Real-time
Software
Dependent
Demonstrate

Keywords

  • Accelerometry
  • Augmented estimating equations
  • Information bias
  • Semiparametric regression model

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

A semiparametric model for wearable sensor-based physical activity monitoring data with informative device wear. / Song, Jaejoon; Swartz, Michael D.; Gabriel, Kelley Pettee; Basen-Engquist, Karen.

In: Biostatistics, Vol. 20, No. 2, 01.04.2019, p. 287-298.

Research output: Contribution to journalArticle

Song, Jaejoon ; Swartz, Michael D. ; Gabriel, Kelley Pettee ; Basen-Engquist, Karen. / A semiparametric model for wearable sensor-based physical activity monitoring data with informative device wear. In: Biostatistics. 2019 ; Vol. 20, No. 2. pp. 287-298.
@article{90db449af2434a909b79dd3e94f94fa6,
title = "A semiparametric model for wearable sensor-based physical activity monitoring data with informative device wear",
abstract = "Wearable sensors provide an exceptional opportunity in collecting real-Time behavioral data in free living conditions. However, wearable sensor data from observational studies often suffer from information bias, since participants' willingness to wear the monitoring devices may be associated with the underlying behavior of interest. The aim of this study was to introduce a semiparametric statistical approach for modeling wearable sensor-based physical activity monitoring data with informative device wear. Our simulation study indicated that estimates from the generalized estimating equations showed ignorable bias when device wear patterns were independent of the participants physical activity process, but incrementally more biased when the patterns of device non-wear times were increasingly associated with the physical activity process. The estimates from the proposed semiparametric modeling approach were unbiased both when the device wear patterns were (i) independent or (ii) dependent to the underlying physical activity process. We demonstrate an application of this method using data from the 2003-2004 National Health and Nutrition Examination Survey (N=4518), to examine gender differences in physical activity measured using accelerometers. The semiparametric model can be implemented using our R package acc, free software developed for reading, processing, simulating, visualizing, and analyzing accelerometer data, publicly available at the Comprehensive R Archive Network.",
keywords = "Accelerometry, Augmented estimating equations, Information bias, Semiparametric regression model",
author = "Jaejoon Song and Swartz, {Michael D.} and Gabriel, {Kelley Pettee} and Karen Basen-Engquist",
year = "2019",
month = "4",
day = "1",
doi = "10.1093/biostatistics/kxx073",
language = "English (US)",
volume = "20",
pages = "287--298",
journal = "Biostatistics",
issn = "1465-4644",
publisher = "Oxford University Press",
number = "2",

}

TY - JOUR

T1 - A semiparametric model for wearable sensor-based physical activity monitoring data with informative device wear

AU - Song, Jaejoon

AU - Swartz, Michael D.

AU - Gabriel, Kelley Pettee

AU - Basen-Engquist, Karen

PY - 2019/4/1

Y1 - 2019/4/1

N2 - Wearable sensors provide an exceptional opportunity in collecting real-Time behavioral data in free living conditions. However, wearable sensor data from observational studies often suffer from information bias, since participants' willingness to wear the monitoring devices may be associated with the underlying behavior of interest. The aim of this study was to introduce a semiparametric statistical approach for modeling wearable sensor-based physical activity monitoring data with informative device wear. Our simulation study indicated that estimates from the generalized estimating equations showed ignorable bias when device wear patterns were independent of the participants physical activity process, but incrementally more biased when the patterns of device non-wear times were increasingly associated with the physical activity process. The estimates from the proposed semiparametric modeling approach were unbiased both when the device wear patterns were (i) independent or (ii) dependent to the underlying physical activity process. We demonstrate an application of this method using data from the 2003-2004 National Health and Nutrition Examination Survey (N=4518), to examine gender differences in physical activity measured using accelerometers. The semiparametric model can be implemented using our R package acc, free software developed for reading, processing, simulating, visualizing, and analyzing accelerometer data, publicly available at the Comprehensive R Archive Network.

AB - Wearable sensors provide an exceptional opportunity in collecting real-Time behavioral data in free living conditions. However, wearable sensor data from observational studies often suffer from information bias, since participants' willingness to wear the monitoring devices may be associated with the underlying behavior of interest. The aim of this study was to introduce a semiparametric statistical approach for modeling wearable sensor-based physical activity monitoring data with informative device wear. Our simulation study indicated that estimates from the generalized estimating equations showed ignorable bias when device wear patterns were independent of the participants physical activity process, but incrementally more biased when the patterns of device non-wear times were increasingly associated with the physical activity process. The estimates from the proposed semiparametric modeling approach were unbiased both when the device wear patterns were (i) independent or (ii) dependent to the underlying physical activity process. We demonstrate an application of this method using data from the 2003-2004 National Health and Nutrition Examination Survey (N=4518), to examine gender differences in physical activity measured using accelerometers. The semiparametric model can be implemented using our R package acc, free software developed for reading, processing, simulating, visualizing, and analyzing accelerometer data, publicly available at the Comprehensive R Archive Network.

KW - Accelerometry

KW - Augmented estimating equations

KW - Information bias

KW - Semiparametric regression model

UR - http://www.scopus.com/inward/record.url?scp=85060093562&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85060093562&partnerID=8YFLogxK

U2 - 10.1093/biostatistics/kxx073

DO - 10.1093/biostatistics/kxx073

M3 - Article

C2 - 29415194

AN - SCOPUS:85060093562

VL - 20

SP - 287

EP - 298

JO - Biostatistics

JF - Biostatistics

SN - 1465-4644

IS - 2

ER -