A mobile technology-based tailored health promotion program for sedentary employees: development and usability study | BMC Public Health
Overview
This study was conducted in two phases. In Phase 1, the primary objective was to design and develop an evidence- and theory-based mHealth web app aimed at promoting healthy eating, increasing PA, and reducing sitting time among sedentary employees. A multidisciplinary team was assembled, including research assistants with dietitian certifications, software engineers, computer programmers, graphic designers, and experts in occupational health, nursing, and nutrition. This diverse expertise ensured the web app was both practical and tailored to the target population. Active collaboration among team members was central to aligning the web app’s features with the specific needs and preferences of sedentary employees. In Phase 2, usability testing was conducted with end users to refine the web app’s design and functionality. Participants were invited to perform a series of tasks within the app, and feedback from these sessions was incorporated into iterative improvements to ensure the app met user needs effectively.
Ethics approval
Ethics approval for this study was obtained from the China Medical University & Hospital Research Ethics Center (CRREC-106-036 [CR-1]).
Phase 1: design and development of the mHealth web app
Step 1: define purpose, target behaviors, and strategies of the app
Behavior change interventions grounded in behavioral theories are known to be effective in promoting lifestyle modifications [26]. Thus, this study incorporated principles from Bandura’s social cognitive theory [20], which emphasizes self-regulation and self-efficacy, and the ecological model [21], which considers various levels of influence on health behavior, including intrapersonal, organizational, community, and policy factors. These theoretical frameworks were used to structure key behavior change techniques (BCTs), such as goal setting, self-monitoring, and personalized feedback, within the app’s design to enhance users’ dietary habits, PA, and reduce sedentary behavior [8, 22, 25, 26, 29].
The app’s primary features included: (1) Health behavior tracking and monitoring: enables users to engage in self-observation and track their health behaviors over time. (2) Goal setting, recommendations, and performance ranking: helps users establish and assess health goals, supporting self-assessment and motivation. (3) Personalized advice: provides tailored tips and insights designed to reinforce self-regulation and self-efficacy. Advice includes elements such as performance accomplishments, opportunities to observe others’ healthy behaviors (vicarious experience), verbal encouragement, and emotional regulation techniques. (4) Integration of environmental influences: recognizes environmental factors impacting behavior and incorporates them into the advice provided, ensuring a holistic approach to health promotion.
Step 2: platform selection
After defining the app’s structure and goals, the next step was to select the platform—either a native app or a web app. Native apps run directly on the smartphone’s operating system and must be downloaded from app stores such as Google Play or the Apple App Store [30]. In contrast, web apps run on external servers and are accessed through web browsers. Turner-McGrievy et al. recommend choosing a platform based on specific app needs [31]. Native apps are suitable when features like photo capturing, location tracking, touchscreen interaction, on-device processing, or notifications are required. However, a web app is more suitable when the goals include low development cost, platform independence, quick updates, and long-term support [31]. For this intervention, a web app was selected due to its accessibility across devices without requiring installation, allowing users to access it on any device with a browser, increasing usability and reach.
Step 3: program content and tailoring algorithms
The intervention’s content and structure were based on evidence-based guidelines and theoretical models of self-regulation, self-efficacy, and environmental influences. The research team reviewed current PA and dietary guidelines [32,33,34,35,36], as well as empirical studies, to inform the app’s program content and features. The core components developed for the app included:
Health behavior goals
-
1.
Hourly stand-up goals: Encourages users to log standing frequency every hour, aiming for two instances per hour to promote light PA based on evidence from Pedisic et al. [37] and Diaz et al. [38]. Users track the number of times they stand each hour.
-
2.
Daily PA goals: Walking is promoted as a simple, accessible PA, with options to track steps via an activity tracker or self-report. Goals are set at 10,000 steps per day [39] or 150–300 min of weekly moderate PA [36]. For users with limited PA, the app starts with 5,000 steps per day, gradually increasing by 10% each week [40]. The app offers incremental targets, such as adding 1,000 steps every two weeks, encouraging self-regulation.
-
3.
Daily healthy eating goals: The app calculates daily calorie targets (ranging from 1,300 to 2,000 kcal) based on Taiwanese adult height averages [41] and various PA levels, aligning with the guideline [42]. The app also follows MyPlate guidelines [35] for balanced diet proportions, categorizing food into whole grains, proteins, vegetables, and fruits. Nutrient intake follows Taiwan’s dietary guidelines, suggesting 10–20% of calories from protein, 20–30% from fat, and 50–60% from carbohydrates [33]. The app recommends six servings of vegetables for users on the 1,300 and 1,500 kcal plans and nine servings for those on the 1,800 and 2,000 kcal plans, with a consistent recommendation of three servings of fruit across all calorie levels. Table 1 provides the recommended daily servings for each food group across the four calorie plans.
Setting reminders
Users can set personalized reminders, sent via LINE@, at scheduled times to encourage behavior tracking. Reminders such as “stand up and move” and “record my health behaviors” prompt regular engagement, and users receive notifications every 14 days to reset goals. LINE@ was chosen as it is widely used in Asia and allows multimedia communication, supporting text, images, and video. To maximize convenience, the web app is integrated with participants’ personal LINE accounts, linked to the program’s official LINE@ account. This integration allows users to receive timely reminders and personalized updates directly through LINE@, keeping them engaged in the program and providing essential notifications without requiring users to manually access the web app.
Daily logging of behaviors
-
1.
Stand-up activity: Users specify whether it is a workday, logging standing frequency for work or during morning, afternoon, and evening periods on non-workdays. Research shows employees often engage in prolonged sedentary behavior on both workdays and weekends [43].
-
2.
Physical activity: The interface adjusts for easy logging based on the user’s selected goal. A guide helps users assess PA intensity, following the Taiwan Health Promotion Administration’s standards [34].
-
3.
Healthy eating: The app provides tools for logging daily dietary intake using detailed nutritional information, portion sizes, and volume charts [33, 44,45,46,47,48,49,50,51]. Foods are grouped into subcategories based on similarities in appearance, preparation, and portion sizes. When the charts lack specific items, the app allows users to weigh portions using a standard research bowl (250 ml, 10 cm diameter, 5 cm depth) or cup (215 ml). Foods are categorized into four main groups: whole grains, proteins, vegetables, and fruits. To better reflect Taiwan’s dietary habits and enhance tracking, the whole grain category is divided into six subcategories and the protein category into five, with clear serving sizes for each item. The app encourages balanced, healthy eating by dividing meals into 3 main categories: breakfast, lunch, and dinner. Users can also log additional meals or snacks to accommodate individual schedules or eating patterns (e.g., intermittent fasting). This feature enables comprehensive tracking of daily food intake, supporting mindful and balanced eating habits.
Personalized advice
An algorithmic decision-making system was developed to deliver tailored advice based on behavioral performance and goal-setting status (i.e., goal achievement rate). These personalized messages were designed to enhance self-efficacy (via observational learning, scenario simulation, verbal persuasion, and emotional regulation) and self-regulation (via self-monitoring, self-assessment, goal setting, plan development, and positive reinforcement). The advice also highlights the benefits of adopting healthy behaviors while addressing potential barriers users may face. Depending on the user’s progress, the system provides different types of advice, including (1) encouragement to increase a behavior if the target has not been met; (2) reinforcement to maintain the behavior if performance is within the acceptable range [52]; and (3) suggestions to adjust or reduce a behavior if it exceeds the target. Table 2 provides an overview of the tailored advice for each behavior, categorized by achievement rate.
Health classroom
-
1.
Educational booklets: The app offers educational booklets that provide practical guidance on increasing PA, maintaining a balanced diet, and reducing sedentary behavior. Six booklets were developed, covering the benefits of PA, tips for staying active, healthy eating basics, practical nutrition advice, safety tips, and simple healthy recipes. Two booklets are released monthly through the app, offering participants easy-to-follow steps to support these health goals.
-
2.
Motivational booklets. To enhance motivation and commitment, six motivational booklets were designed to address goal setting, overcoming obstacles, avoiding temptations, preventing relapse, maintaining motivation, and sustaining healthy habits [53]. One booklet is provided biweekly through the app, encouraging participants to recognize challenges, stay focused, and continue their progress toward a healthier lifestyle.
-
3.
Recommended articles and links: To further enrich the participant experience, the app curates articles and videos from reputable health sources. These resources cover reducing sedentary behavior, increasing PA, and adopting healthier eating habits. Additionally, links to trusted websites are provided, allowing users to explore these topics in greater depth and supplement their learning with valuable, credible information.
Health rankings
-
1.
My performance: Users can track key health metrics—hourly standing frequency, daily step count or minutes of PA, and daily servings of four major food groups—alongside personalized goals. The app calculates an achievement rate by comparing actual performance to target values, supporting users in self-assessing progress. Performance charts reflecting overall PA, workday, and non-workday behavior help users understand patterns, addressing findings that office workers experience prolonged sedentary periods [43].
-
2.
Personal and team rankings: Users can view personal and team-based health performance, fostering individual accountability and group engagement. Rankings are based on three core behaviors: sitting less, moving more, and healthy eating. For each behavior, scores are calculated relative to user-specific goals and aggregated into an overall score. Team scores are derived from the total of individual scores divided by the team size, allowing comparisons across teams.
-
(1)
“Sitting less score” reflects the frequency of hourly standing, with higher scores for users who stand up more frequently.
-
(2)
Physical activity levels are converted into a “moving more score” to allow fair comparisons between users with and without activity trackers. Research indicates that 30 min of moderate-intensity PA (approximately 3,800–4,000 steps) [54] aligns with standard daily levels of 6,000–7,000 steps [55]. Based on these estimates, 10,000 daily steps equate to 30 min of moderate-intensity PA. Users earn 100 points by achieving 10,000 steps, 30 min of light or moderate PA, or 20 min of vigorous PA.
-
(3)
Based on the Healthy Eating Index (HEI) [56], a “healthy eating score” evaluates daily intake of whole grains, proteins, vegetables, and fruits. Each category contributes up to 25 points for a total possible score of 100. Full points are awarded for intake within 90–110% of the target. Vegetables must meet 90% of the goal due to common dietary deficiencies, while fruits earn total points for 90–184.9% of the target; exceeding 185% results in no points, following the DASH diet’s recommendations of 5.2 servings at 2,100 kcal [57]. Food intake is evaluated daily rather than per meal, allowing flexibility. An allowable deviation range is set to encourage consistent healthy eating behaviors [52].
-
(4)
The “overall score,” 100 points, allocates 34 points for sitting less, 33 for moving more, and 33 for healthy eating, providing a balanced assessment of all behaviors.
-
(1)
Support partners
This feature includes “my team,” “join team,” and a “message board” to help users build a personalized support network within the app. Users can invite colleagues to form teams to collaborate on health goals, share progress, celebrate achievements, and offer mutual encouragement and motivation. This feature fosters interaction and peer support among individuals with similar health objectives. The message board further enhances communication by enabling discussions, sharing challenges, celebrating successes, and creating an engaging, supportive community within the app.
Step 4: web app development and database management
A software development company built the Simple Health web app under the research team’s guidance. The company was responsible for the technical development and app maintenance. To ensure the app’s content and functionality met users’ needs, it was validated by an expert panel and the target population. Figures 1, 2 and 3 display the app’s structure and homepage screenshots in computer and smartphone versions. Flowcharts and screenshots are provided in Additional File 1.
Screenshot of the homepage (computer version)
Screenshot of the homepage (smartphone version)
The app was built using the Hypertext Preprocessor (PHP) programming language to facilitate database communication and generate user interfaces. Responsive Web Design (RWD) was implemented using HyperText Markup Language (HTML), Cascading Style Sheets (CSS), JavaScript, JQuery, and Bootstrap, ensuring compatibility across various devices. For data security, MD5 encryption was applied to protect user passwords. Data from the relational database were regularly exported in comma-separated values (CSV) format, allowing easy integration with statistical software for analysis. Several visual assets, including images, were sourced from Shutterstock to avoid copyright issues, maintaining both a professional appearance and legal compliance.
Step 5: post-production quality control
The initial version of the web app underwent multiple rounds of alpha testing to ensure quality control, focusing on identifying and resolving technical and logical issues. Research team members and students tested the app, offering valuable feedback on content accuracy, functionality, editorial consistency, and overall usability. Feedback from external volunteers provided fresh perspectives, highlighting areas where instructions were unclear or specific program features were confusing. Based on this comprehensive feedback, revisions were implemented, and the app underwent an additional round of quality control. Additionally, the program was tested with eight sedentary employees before its full implementation, providing further insights and confirming that the app was user-friendly, effective, and ready for broader use.
Phase 2: usability testing
The usability and acceptability of the prototype Simple Health web app were evaluated through data verification and user assessment. To ensure accuracy, the principal investigator and research assistant cross-checked app data against the relational database by randomly logging dietary and activity behaviors and verifying correct data processing and display.
User perceptions of usability and acceptability were assessed using a Chinese-translated version of the MAUQ. The MAUQ consists of 17 questions covering three domains: ease of use, interface satisfaction, and usefulness [27]. Responses were recorded on a 7-point Likert scale (1 = strongly agree to 7 = strongly disagree), where lower average scores indicate higher usability. The MAUQ has demonstrated strong internal consistency (Cronbach’s alpha = 0.914), with the three subscales showing good to acceptable internal consistency (Cronbach’s alpha range: 0.72–0.91) [27]. Additionally, three open-ended questions gathered qualitative feedback on (1) the most frequently used app features, (2) the time required for daily entries, and (3) suggestions for usability improvements.
The total mean score of the MAUQ was calculated without weighting individual items, as all responses were collected on a 7-point Likert scale. This approach aligns with the validated MAUQ methodology, ensuring consistency with prior studies and preserving the comparability of results. Given the strong internal consistency of the instrument and its uniform response format, weighting was not necessary to maintain interpretability and methodological rigor.
A purposive sample of eight sedentary workers was recruited from an aerospace company in Taichung, Taiwan. Initially, we aimed to recruit five participants, as prior research indicates that usability testing with as few as five users can uncover a substantial proportion of interface issues [58]. However, during recruitment, additional employees expressed interest in participating. To ensure gender balance, we included four males and four females, ultimately expanding the sample to eight participants. While still a small sample, this slight increase allowed us to capture a broader range of user experiences and gather more diverse usability feedback.
Participants were recruited through internal workplace communication channels, with eight individuals voluntarily agreeing to participate. Once a balanced gender distribution was achieved, no further recruitment efforts were made. Eligibility criteria included being a full-time sedentary employee aged 20 or older, having no physical limitations that would hinder participation in PA, and owning a smartphone with Internet access and the LINE app installed. Informed consent was obtained prior to testing.
A “sedentary employee” was defined as an individual who spends more than six hours per workday sitting. This definition aligns with occupational health research [59], which suggests that sitting for approximately 6.4 h or more during the workday represents a substantial portion of work hours spent in a sedentary posture. This threshold is commonly used in workplace health studies to identify individuals at risk of prolonged sedentary behavior and its associated health risks.
Participants engaged with the Simple Health web app while wearing a Fitbit Alta HR activity tracker to monitor their daily step count. Iterative modifications were made following each testing round until users successfully completed 90% of assigned tasks. As an acknowledgment of their time and contribution, each participant received a US$16 incentive.
link
