2021
DOI: 10.2196/27779
|View full text |Cite
|
Sign up to set email alerts
|

Text Message Analysis Using Machine Learning to Assess Predictors of Engagement With Mobile Health Chronic Disease Prevention Programs: Content Analysis

Abstract: Background SMS text messages as a form of mobile health are increasingly being used to support individuals with chronic diseases in novel ways that leverage the mobility and capabilities of mobile phones. However, there are knowledge gaps in mobile health, including how to maximize engagement. Objective This study aims to categorize program SMS text messages and participant replies using machine learning (ML) and to examine whether message characteristi… Show more

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
5
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
6

Relationship

1
5

Authors

Journals

citations
Cited by 8 publications
(5 citation statements)
references
References 32 publications
0
5
0
Order By: Relevance
“…Separate analyses indicated that supportive and informational messages in particular elicited more message replies. 31 There are several limitations to our study that need to be considered. The study intervention, although evaluated in a randomized design, was not able to be blinded.…”
Section: Discussionmentioning
confidence: 98%
See 1 more Smart Citation
“…Separate analyses indicated that supportive and informational messages in particular elicited more message replies. 31 There are several limitations to our study that need to be considered. The study intervention, although evaluated in a randomized design, was not able to be blinded.…”
Section: Discussionmentioning
confidence: 98%
“…Separate analyses indicated that supportive and informational messages in particular elicited more message replies. 31…”
Section: Discussionmentioning
confidence: 99%
“…Less research has compared specific types of content and has rarely tested different types within the same study. An exception is a recent study by Klimis et al [ 15 ], wherein they used machine learning to demonstrate that text messages with informative (providing health facts or education) and instructional (providing tips or recommendations) message intents were associated with increased engagement, while notification messages that addressed noneducational matters (eg, welcome and exit messages) were associated with reduced engagement [ 15 ]. Our study targeted a two-way message and varied the levels of reflection and response burdens in that message.…”
Section: Discussionmentioning
confidence: 99%
“…Across the mHealth literature, engagement tends to be highly variable [10,11], which has spurred a whole body of research that aims to understand predictors of engagement, including user characteristics and intervention features (eg, intervention duration and frequency of sending content) [10][11][12][13][14]. However, very little research has attended to the type of mHealth content that users are expected to engage with [15] and, more specifically, how the content may be requesting more or less cognitive reflection. The primary goal in having users engage with mHealth content is health behavior change.…”
Section: Introductionmentioning
confidence: 99%
“…19 The following metrics were defined a priori based on prior studies and analyzed quantitatively using proportions. 20,21 They also served as our primary and secondary outcomes.…”
Section: Discussionmentioning
confidence: 99%