JMIR Publications

ABSTRACT

Background:

Digital phenotyping is a promising methodology for capturing moment-to-moment data that can inform individually adapted and timely interventions for youth with chronic pain.

Objective:

The objective of the current study is to investigate adolescent and parent endorsement, perceived utility, and concerns related to smartphone sensor data collection for digital phenotyping for clinical and research purposes in youth with chronic pain.

Methods:

Adolescent and parent participants were recruited for study participation at primarily at clinic visits for pediatric chronic pain treatment in the Northeast, US. Participants completed an online survey consisting of multiple-choice and open-response questions probing their endorsement, perceived utility, and concerns related to passive data collection from the following nine embedded smartphone sensors: Accelerometer, Applications, Bluetooth, Text/Call Log, Keyboard, Voice, Light, Screen, and GPS. We conducted binary logistic regression analyses, controlling for demographic variables (i.e., age, gender, and parental education), to test whether perceived utility predicted endorsement for each sensor. Open-ended responses were coded and analyzed thematically.

Results:

A total of 103 adolescents (M=15.64, SD=1.59) and 99 parents (M=47.79, SD=6.34) completed study measures. Participants generally endorsed the use of sensors, though adolescent endorsement varied by sensor, ranging from 35.0% (Keyboard and Voice) to 75.7% (Accelerometer). A majority (>50%) of adolescents endorsed use of six of nine sensors, and only Applications (47.6%), Keyboard (35.0%), and Voice (35.0%) received less than 50% endorsement. Parent endorsement of sensor use similarly varied by sensor and ranged from 53.5% (i.e., Voice) to 81.8% (i.e., Accelerometer and Screen). A majority (>50%) of parents endorsed use of each of the nine sensors. Logistic regression results also indicated adolescent perceived utility of each sensor significantly predicted the likelihood of endorsement of its use in both clinical practice (AORs ranging from 2.90 to 4.95) and research (AORs ranging from 2.18 to 3.80). The same pattern of significance emerged that parent perceived utility of each sensor significantly predicted the likelihood of endorsement of its use in both clinical practice (AORs ranging from 2.68 to 6.91) and research (AORs ranging from 1.86 to 7.47). Thematic analysis of participant concerns related to each sensor revealed five overarching themes of Privacy Concerns, Data Concerns, Absence of Personal Voice, Iatrogenic Effects, and Other. The most common themes of concerns that emerged across all sensors were Privacy Concerns and Data Concerns.

Conclusions:

Findings indicate general endorsement of digital phenotyping for youth with chronic pain and that perceived utility predicts this endorsement. Understanding adolescent and parent perspectives in this population can help refine this methodology toward an acceptable, feasible, and ethical implementation of real-time symptom monitoring for assessment and intervention in youth with chronic pain.