# mHealth Intervention to Promote Nonexercise Physical Activity in Patients With Type 2 Diabetes: Secondary Analysis and Implementation Study

## TL;DR

This study explores how mHealth tools can help patients with type 2 diabetes increase nonexercise physical activity, focusing on behavior change needs and implementation challenges.

## Contribution

The study identifies specific behavior change needs and evaluates the implementation of an mHealth intervention for nonexercise physical activity in type 2 diabetes patients.

## Key findings

- Participants identified key behavior change needs in capability and motivation.
- The intervention showed high fidelity and acceptability, with some technological challenges.
- Pretesting technology-based approaches is emphasized for better implementation in clinical practice.

## Abstract

Physical activity (PA) has an important role in the prevention and treatment of type 2 diabetes (T2D). Interventions with mobile-based technology (mobile health [mHealth]) seem promising in PA promotion, but their behavioral framework is often vague, and the implementation is seldom reported.

This paper examines perceived behavior change needs and implementation of an mHealth approach in increasing nonexercise PA in patients with T2D.

A 3-arm mHealth intervention was conducted in primary care. Information on perceived behavior change needs was collected with a modified capability, opportunity, motivation—behavior (COM-B) questionnaire before the intervention from a separate sample of patients with T2D (n=25) and at the intervention baseline (n=119). Implementation evaluation focused on the fidelity and acceptability of the main arm of the intervention (n=39), which included 24-hour accelerometer use, a smartphone app with personal feedback, a PA leaflet, a YouTube video on walking, and individual counseling with 3 face-to-face sessions and 4 telephone contacts. Data on fidelity were accumulated during the intervention through counseling cards and cloud computing. Data on acceptability were collected with a questionnaire at the end of the intervention (Likert scale from 1 to 5). Data analysis was mainly descriptive.

The participants’ responses revealed 3 items in capability and 2 in motivation, which stood out as perceived behavior change needs. Moreover, the main intervention arm showed good fidelity (eg, face-to-face sessions: 112/117, 96% and telephone contacts completed: 145/156, 93%; mean weekly accelerometer use 54%; ranging from 80% to 17% during the intervention) and acceptability (mean score ranging from 3.8 to 4.8), although some challenges were also experienced, especially in cloud-computed feedback and accelerometer-app use.

The findings on behavior change needs call for additional research since no comparable studies were found. In addition, the explanatory value of the COM-B model and the psychometric properties of the COM-B questionnaire deserve further attention. The main intervention arm seemed applicable to clinical practice. However, the challenges discovered underscore the importance of pretesting technology-based approaches in patients with T2D.