JMIR Publications

ABSTRACT

Background:

Wearable activity monitors (WAMs) provide insights into physical activity (PA) and are widely used in behavioral interventions and cancer survivorship research. However, validation studies of wearable devices in cancer populations are scarce, and existing studies using activity monitors in cancer patients lack standardization. This gap is particularly significant in lung cancer (LC) patients, who often experience unique mobility challenges and gait impairments that may affect device accuracy. This study addresses this gap by validating the Fitbit Charge 6, ActiGraph Leap, and activPAL 3micro in LC patients in both laboratory and free-living conditions, and by developing a standardized framework for assessing wearable devices in cancer populations with impaired mobility.

Objective:

The aim of this study is to validate and compare the accuracy of consumer-grade (Fitbit Charge 6) and research-grade (activPAL 3micro and ActiGraph Leap) WAMs in LC patients under both laboratory and free-living conditions. Additionally, this protocol aims to establish standardized procedures that can be adapted for validating both current and future generations of wearable devices, while accounting for disease-specific factors that may impact measurement accuracy.

Methods:

Fifteen adults diagnosed with LC (stage I-IV) will participate in laboratory and free-living protocols, wearing a Fitbit Charge 6, activPAL 3 micro, and ActiGraph Leap simultaneously. The laboratory protocol will consist of a series of structured activities including variable-time walking trials, sitting/standing tests, posture changes, and gait speed assessments. Activities will be video recorded for validation. In the free-living protocol, participants will wear the devices continuously for seven days except during water-based activities. WAM-based outcome measures will include step count, time spent at light, moderate, and vigorous PA intensity levels, and posture/posture changes (activPAL only). Validated survey instruments will be administered both prior to and after activity monitor data collection to control for potential confounding factors that may influence movement patterns and device accuracy. Lab-based validity measures will compare WAM data to video-recorded observations; sensitivity, specificity, positive predictive value, and agreement will be determined. Free-living agreement between devices will be assessed using Bland-Altman plots, intraclass correlation analysis, and ninety-five percent limits of agreement.

Results:

Data collection is ongoing, with 11 participants enrolled and 7 having completed both in-lab and free-living protocols. The average age of enrolled participants is 63.0 years (range 50.0-73.0, SD 7.8 years), with 8 out of 11 participants being women. Participant enrollment is expected to conclude in mid-2025, and initial findings are expected to be disseminated by end of 2025.

Conclusions:

This is the first study that validates WAM accuracy for LC populations while providing comprehensive recommendations for future validation studies. This study will provide critical insights into the accuracy and reliability of WAMs for assessing PA in LC survivors, which are essential for interpreting clinical research and informing future interventions.