Interactive Physical Activity Apps: Do the ABACUS and the MARS Measure Up? A Descriptive Analysis of Behaviour Change Taxonomies

This descriptive analysis of 17 free physical activity apps reveals that while they are generally well-rated and frequently employ basic behavior change techniques like goal setting and feedback, they often lack comprehensive evidence-based strategies, suggesting a need for developers to incorporate more robust behavior change methods to better improve user health outcomes.

The Gist

Imagine the world of fitness apps as a massive, bustling digital supermarket. There are thousands of aisles filled with apps promising to get you moving, lose weight, or build muscle. But here's the catch: most of the "good stuff" (the really effective tools) is locked behind a paywall, or it requires you to buy expensive gadgets like smartwatches to work properly.

This study is like a team of detectives who decided to walk through that supermarket and only pick up the items that are completely free and interactive (meaning they talk back to you, not just show you a static list of exercises). They wanted to answer two big questions:

1. Do these free apps actually know how to change your behavior, or are they just pretty pictures?
2. Do the apps that are rated highly by regular users actually have the "secret sauce" of behavior science?

Here is the breakdown of their findings, served with a side of metaphors.

1. The Great Filter: Finding the "Free" Gems

The researchers started with 600 apps (imagine a giant pile of 600 different boxes). They had to throw most of them away because:

• They weren't free (they were like "sample boxes" that only let you taste a little before asking for your credit card).
• They weren't interactive (they were just digital brochures).
• They required expensive hardware (like needing a $300 watch to count your steps).

After the great filter, only 17 apps remained. These were the "free samples" that actually let you cook the meal without buying the kitchen.

2. The Behavior Change Checklist (ABACUS)

To see if these apps were actually good at changing habits, the researchers used a checklist called ABACUS. Think of ABACUS as a 21-point "Behavior Change Report Card." It asks things like:

• Does the app ask what your goals are?
• Does it give you feedback when you do well?
• Does it help you plan for when you want to quit?
• Does it explain the consequences of staying sedentary?

The Score:
The apps were like students who studied hard but missed the final exam.

• The Good News: Every single app got an "A" on the basics. They all collected your info, gave you workout instructions, and cheered you on when you finished a set.
• The Bad News: No app got a perfect score. The average score was about 60%.
• The Missing Pieces: The apps were terrible at the "deep" stuff. Almost none of them explained why you should keep going (consequences of quitting), and very few helped you plan for obstacles (like "What if it rains and I can't run?"). It's like a gym coach who tells you how to lift a weight but never tells you why it matters or how to keep going when you're tired.

3. The Quality Check (MARS)

Next, they used a tool called MARS to judge the app's "vibe" and usability. Think of this as judging a restaurant on its food, decor, and service.

• Functionality (The Kitchen): The apps were excellent here. They worked well, didn't crash, and were easy to navigate.
• Subjective Quality (The Chef's Recommendation): This is where the researchers (the experts) gave the apps a lower score. They felt the apps were good tools, but not great health interventions. They were like a fast-food burger: tasty and easy to eat, but maybe not the healthiest meal you could have.

4. The User Rating Mystery

Here is the twist in the story. The researchers looked at the App Store star ratings (what regular people think).

• The Reality: The apps had high star ratings (mostly 4 to 5 stars). People loved them!
• The Disconnect: The researchers found that the apps with the highest behavior-change scores (the ones that actually knew how to change habits) did NOT necessarily have the highest user ratings.

The Analogy:
Imagine a strict personal trainer (high behavior science, high effectiveness) vs. a fun, encouraging cheerleader (low behavior science, high fun).

• The apps in this study were mostly cheerleaders. They were fun, easy to use, and gave you a high-five. Users loved them and gave them 5 stars.
• However, they weren't necessarily the strict trainers who would force you to do the hard work needed for long-term health.
• The study found that the only thing that predicted a high user rating was the app's overall "smoothness" and how it felt to use. The deep, scientific stuff didn't seem to matter to the average user when they were clicking "rate this app."

5. The "Honeymoon Phase" Problem

The researchers suspect that users are rating these apps while they are in the "Honeymoon Phase."

• When you first download a new app, you are excited. You set a goal, you do a workout, and you feel great. You give it 5 stars.
• But the study suggests that once the "honeymoon" wears off (usually after 3 months), people stop using the apps because the apps didn't have the right tools to help them stick with it when things got boring or hard.

The Bottom Line

The Verdict:
Free, interactive fitness apps are a great starting point, especially for people who can't afford expensive gym memberships or smartwatches. They are like a free trial membership to the world of fitness.

The Warning:
However, they are currently incomplete tools. They are great at getting you to start (uptake), but they are missing the "secret sauce" to help you stay (adherence). They are like a map that shows you the path to the mountain peak but doesn't give you a compass for when the fog rolls in.

The Takeaway for You:
If you use a free fitness app, enjoy the free workouts and the encouragement! But be aware that the app might not be giving you the deep, psychological support needed to make exercise a lifelong habit. You might need to bring your own "coach" (a friend, a habit tracker, or a bit of willpower) to fill in the gaps that the app left behind.

Technical Summary

1. Problem Statement

Despite the proven health benefits of physical activity (PA) and the proliferation of digital health tools, a significant gap exists in the accessibility and efficacy of cost-free, interactive physical activity apps.

• Accessibility Barrier: Many popular PA apps require paid subscriptions or expensive wearable hardware, limiting access for low socioeconomic status (SES) populations.
• Efficacy Gap: While digital interventions show promise, it is unclear how publicly available, free apps adhere to evidence-based behaviour change techniques (BCTs).
• Research Gap: No prior studies have examined the relationship between app-store user ratings, app quality (measured by the Mobile App Rating Scale), and the incorporation of specific behaviour change taxonomies (measured by the App Behaviour Change Scale) in free, interactive apps.

2. Methodology

The study employed a digital environmental scan using a descriptive content analysis approach.

• Data Collection:
  • Sources: Google Play (Android) and Apple iTunes (iOS) stores.
  • Search Terms: "Physical activity," "fitness," and "exercise."
  • Initial Sample: 600 apps (300 per store).
  • Inclusion Criteria: English language, PA as primary target, adult population (18+), interactive/two-way interface, and completely cost-free for interactive components.
  • Exclusion Criteria: Paid subscriptions, non-interactive repositories, apps requiring wearables (e.g., smartwatches) or external equipment, and non-PA focused apps.
• Screening Process:
  • Two independent reviewers screened titles/descriptions.
  • Final sample: 17 apps (after exclusions for paid features, inactivity, or lack of interactivity).
  • Apps were downloaded and monitored for 48 hours to capture notifications and pop-ups.
• Measurement Instruments:
  • ABACUS (App Behaviour Change Scale): A 21-item scale assessing BCTs across four domains: Knowledge/Information, Goals/Planning, Feedback/Monitoring, and Actions.
  • MARS (Mobile App Rating Scale): Assesses app quality across four objective domains (Engagement, Functionality, Aesthetics, Information) and one subjective quality domain.
• Data Analysis:
  • Descriptive statistics (means, frequencies).
  • Pearson correlations to explore relationships between ratings.
  • Linear Regressions: To determine if ABACUS scores or MARS domains predict app store user ratings and subjective quality.

3. Key Contributions

• First Study of its Kind: This is the first study to exclusively analyze free, interactive PA apps using both ABACUS and MARS frameworks simultaneously.
• Commercial vs. Health Focus: It highlights the tension between commercial motivations (brand marketing) and evidence-based health outcomes in free apps.
• Validation of Tools: It demonstrates the utility of ABACUS and MARS in evaluating the "real-world" landscape of free digital health interventions.

4. Key Results

A. App Characteristics

• Sample: 17 apps (13 Android, 4 iOS).
• Affiliation: 100% commercially affiliated (often linked to large brands like Nike, Decathlon).
• User Ratings: Generally high ($\mu = 4.5 \pm 0.4$ out of 5), despite varying quality in behaviour change features.
• Updates: 52.9% updated within the last year.

B. ABACUS Findings (Behaviour Change Techniques)

• Overall Score: No app addressed all 21 indicators. Scores ranged from 8 to 18 (Mean: $12.8 \pm 2.4$).
• Most Common Features:
  1. Collection of baseline information (100%).
  2. Instructional PA content (100%).
  3. Ability to give user feedback (100%).
• Least Common Features:
  1. Ability to export data (0%).
  2. Consequences for PA dis/continuance (5.9%).
  3. Planning for barriers (11.8%).
• Gap: Only 11.7% of apps were consistent with national PA guidelines or demonstrated expertise in PA domains.

C. MARS Findings (App Quality)

• Focus Areas: 100% focused on behaviour change; 94.1% included goal setting; 58.8% addressed physical health; 41.2% included mindfulness.
• Strategies: All apps used feedback, information/education, monitoring, and advice. None used evidence-based therapeutic techniques like Cognitive Behavioral Therapy (CBT) or Acceptance Commitment Therapy (ACT).
• Quality Scores:
  • Functionality: Highest mean score ($\mu = 4.2$).
  • Subjective Quality: Lowest mean score ($\mu = 3.5$).
  • Overall Quality: $\mu = 4.0$.

D. Regression Analysis

• User Ratings vs. MARS: Aggregated MARS domains accounted for 54% of the variance in app store user ratings ($p=0.02$). Individual domains did not significantly predict ratings.
• Subjective Quality vs. ABACUS: Significant positive correlations were found between MARS Subjective Quality and:
  • ABACUS Total Score ($R^2 = 0.60$).
  • ABACUS Feedback & Monitoring ($R^2 = 0.53$).
  • ABACUS Actions ($R^2 = 0.54$).
• Implication: Apps with more robust behaviour change techniques were rated higher by expert reviewers (subjective quality), but this did not always translate linearly to public user ratings, which remained generally high across the board.

5. Significance and Conclusion

• Equity vs. Quality: Free apps provide a vital, equitable entry point for PA, particularly for low-SES groups, but they often lack the depth of evidence-based behaviour change techniques required for long-term adherence.
• Commercial Influence: The high prevalence of commercial affiliations suggests these apps may prioritize brand engagement over clinical efficacy. For example, the Nike Training Club app had a low ABACUS score (8/21) and lower user ratings compared to others, suggesting users may detect a lack of genuine support for behaviour change.
• Design Recommendations: Developers should incorporate more advanced BCTs, specifically:
  • Planning for barriers.
  • Discussing consequences of inactivity.
  • Evidence-based therapeutic strategies (e.g., CBT).
  • Data export capabilities.
• Future Directions: Future research must address the "drop-off" after the initial "honeymoon phase" of app use. While current apps are good at uptake (getting users to start), they lack features to support long-term adherence.

Final Verdict: While free interactive apps are accessible and generally well-rated by users, they currently underutilize evidence-based behaviour change science. To truly improve public health outcomes, developers must move beyond basic tracking and instruction to incorporate comprehensive, evidence-informed behaviour change strategies.