Efficacy of Tailored Messages for 28-Week Exercise Sustainability in People with HIV

This study demonstrates that adding automated, theory-based tailored text messages to motivational interviewing effectively sustains moderate-to-vigorous physical activity and improves self-efficacy in people with HIV over a 28-week period, outperforming non-tailored educational messages.

The Gist

Imagine your body is a car. For people living with HIV, this car has a special engine that tends to wear out a bit faster than usual, leading to rust and breakdowns (like heart disease) sooner than in other cars. The best way to keep this engine running smoothly is exercise—specifically, getting your heart rate up and moving vigorously.

But here's the problem: Starting the car is easy; keeping it running for years is hard. Most people can get excited about a new gym routine for a few weeks, but then life gets in the way, motivation fades, and the car sits in the garage.

This study asked a simple question: Can we build a better "GPS" to help people with HIV keep driving their exercise routine for the long haul?

The Experiment: Two Types of GPS

The researchers took 92 people with HIV who had already successfully completed a 16-week "boot camp" exercise program. They were all fit enough to keep going, but the researchers wanted to see how to help them stay that way.

They split these people into two groups and gave them a digital "GPS" (text messages) for the next 12 weeks. Both groups had to fill out a quick daily survey about what was stopping them from exercising (e.g., "I'm too tired," "It's raining," "I forgot").

Group A: The "Generic Map" (Educational Control)
This group got standard, one-size-fits-all advice.

• The Message: "Keep moving! Exercise is good for you."
• The Analogy: It's like getting a generic weather report that says, "It might rain today, so maybe bring an umbrella." It's true, but it doesn't tell you which umbrella to grab or where to go.

Group B: The "Smart, Personal GPS" (Tailored Messages)
This group got messages based on the Two Minds Theory. The researchers believed that while our "thinking brain" (Narrative Mind) knows exercise is good, our "feeling brain" (Intuitive Mind) is what actually makes us move.

• The Message: If you said, "I'm too tired," the system didn't just say "Don't be tired." It might send a message like, "Remember how good you felt after your last walk? Just 10 minutes will recharge you," or "Put your shoes by the door right now so you don't have to think about it later."
• The Analogy: This is like a smart GPS that knows you're stuck in traffic, sees you're running late, and says, "Hey, I know you're stressed. Let's take this shortcut through the park instead. It's faster and you'll feel calmer." It addresses your specific feelings and barriers in real-time.

The Secret Sauce: Both groups also had a "coach" who called them a few times to chat and encourage them (a technique called Motivational Interviewing), but the text messages were the main test.

The Results: Who Kept Driving?

After 28 weeks, the researchers checked the odometers (using both self-reports and motion sensors on their hips).

• The Generic Map Group: They started strong, but their exercise levels slowly drifted downward. They were like drivers who got distracted and eventually took the scenic route back to the couch.
• The Smart GPS Group: They managed to keep their speed up. Even though they faced the same tiredness and busy schedules, the personalized messages helped them push through. By the end, they were exercising significantly more than the other group.

The "Aha!" Moments:

1. It wasn't just about the words: The messages that worked best were the ones that felt like a conversation. When the system asked, "How are you feeling?" and actually listened, people felt more confident in their ability to exercise (self-efficacy).
2. The "Time" Barrier: People in the Smart GPS group started to feel like they had more time to exercise, whereas the other group still felt like they were too busy.
3. The Coach Helped, but the GPS was Key: Even though the coach was helpful, the "Smart GPS" messages seemed to do the heavy lifting. It suggests that a little bit of automated, personalized care can do a lot of the work that usually requires a human coach.

The Bottom Line

Think of this study as proving that personalized care works better than generic advice.

If you tell a person with HIV to "just exercise," they might stop. But if you have a system that says, "I know you're tired today, but remember how you felt last week? Let's just do 10 minutes," that small, tailored nudge can keep the engine running for years.

This is a big deal because it offers a low-cost, automated way to help people stay healthy as they age, turning a difficult habit into a sustainable lifestyle. It's not about forcing people to run a marathon; it's about giving them the right map to keep walking every single day.

Technical Summary

1. Problem Statement

People with HIV (PWH) face accelerated aging and an increased risk of cardiovascular disease and other comorbidities. While moderate-to-vigorous physical activity (MVPA) mitigates these risks, sustaining MVPA over the long term is a significant challenge, particularly for PWH who often have lower baseline activity levels. Existing interventions often focus on initiating behavior change (the "action phase") but fail to address the specific mechanisms required for long-term maintenance (the "maintenance phase"). Furthermore, traditional counseling methods like Motivational Interviewing (MI) are effective but resource-intensive and difficult to scale. There is a need for scalable, theory-driven interventions that specifically target the barriers to sustained physical activity in this population.

2. Methodology

Study Design

• Type: Two-phase randomized controlled trial (RCT).
• Phase 1: 142 PWH (>50 years old) completed a 16-week supervised exercise program (High-Intensity Interval Training or Continuous Moderate Exercise).
• Phase 2 (Focus of this paper): 92 participants who completed Phase 1 were re-randomized (1:1) to one of two groups for a 12-week maintenance period (Weeks 16–28):
  1. Tailored-Messaging Group: Received daily text messages tailored to their specific exercise barriers, based on Two Minds Theory (TMT), plus Motivational Interviewing (MI) coaching (at one site).
  2. Educational-Control Group: Received daily non-tailored, factual educational messages plus MI coaching (at one site) or check-in calls without MI (at the other site).

Participants

• Population: PWH aged >50, sedentary at baseline, on antiretroviral therapy with viral load <200 copies/mL.
• Sample Size: 92 participants re-randomized for Phase 2 (22 in tailored group and 25 in control group completed the full 28 weeks with data).
• Setting: Two sites (Aurora, CO and Seattle, WA).

Intervention Details

• Theoretical Framework: Two Minds Theory (TMT), which posits that behavior is driven by an "Intuitive Mind" (non-conscious, emotional, social) and a "Narrative Mind" (conscious, rational). Messages were designed to address barriers using strategies targeting both minds.
• Tailored Messaging: An algorithm selected one of 400 messages daily based on the participant's response to a daily survey identifying their primary barrier (out of 10 options). Messages included psychological behavior change strategies (e.g., social imagery, problem-solving, cues/reminders).
• Control Messaging: 25 generic educational messages designed to be informative but not tailored to individual barriers or psychological principles.
• Coaching: MI was delivered by health coaches at Weeks 16, 20, and 24. Note: There was a protocol deviation where one site delivered MI to both groups, while the other site delivered MI only to the tailored group.

Measures

• Primary Outcome: Minutes of MVPA per day, measured via Actigraphy (ActiGraph GT3X worn for 7 days/month at weeks 16, 20, 24, 28).
• Secondary Outcome: Self-reported MVPA minutes from daily surveys.
• Exploratory Outcomes: Exercise self-efficacy, perceived health (SF-36), fatigue, mood, pain, and specific barriers (e.g., lack of time).
• Analysis: Linear mixed models with group-by-time interaction effects, using Intent-to-Treat (ITT) principles (treating missing data as zero activity) and completer analyses.

3. Key Contributions

• Theoretical Application: This is the first study to test an intervention specifically designed using Two Minds Theory (TMT) to sustain physical activity. It operationalizes the distinction between conscious (Narrative) and non-conscious (Intuitive) drivers of behavior in a digital health context.
• Scalable Maintenance Strategy: Demonstrates that automated, theory-based tailored messaging can effectively sustain MVPA, offering a scalable alternative or supplement to resource-heavy human coaching.
• Disentangling Mechanisms: Through a unique site-level variation in MI delivery, the study provides exploratory evidence that tailored messaging has an effect independent of, and additive to, Motivational Interviewing.
• Barrier-Specific Tailoring: Validates the efficacy of matching intervention content to daily, self-identified barriers rather than using static educational content.

4. Results

Primary Outcome (Actigraphy)

• Sustainability: Participants in the tailored-messaging group maintained their MVPA levels, whereas the educational-control group showed a decline.
• Interaction Effect: A significant group-by-time interaction was found ($p = .04$ for ITT; $p = .01$ for completers).
• End-of-Study Data (Week 28):
  • Tailored Group: Mean = 48.8 min/day (SD = 45.8).
  • Control Group: Mean = 40.7 min/day (SD = 24.6).
  • Note: While the control group started slightly higher, the tailored group surpassed them by the end of the study.

Secondary Outcome (Self-Report)

• Results mirrored the actigraphy data. The tailored group showed a significant increase in reported MVPA relative to the control group over time ($p = .002$ for ITT interaction).

Exploratory Outcomes

• Self-Efficacy & Health: The tailored group reported significantly higher exercise self-efficacy and better perceived health over time compared to the control group.
• Barriers: The tailored group became significantly less likely to report "lack of time" as a barrier.
• Message Characteristics: Messages featuring free-text responses, social imagery, and mindfulness strategies were associated with higher same-day MVPA.

Adverse Events

• 10 adverse events were reported (8 participants), mostly unrelated to the study. No significant safety concerns were identified.

5. Significance and Conclusion

Clinical and Public Health Impact
The study confirms that automated, theory-based tailored messaging is an effective strategy for sustaining physical activity in older adults with HIV. Given the high burden of cardiovascular disease in this population, interventions that successfully maintain MVPA are critical for healthy aging. The intervention is low-cost and scalable, making it a viable candidate for widespread implementation in HIV care settings.

Theoretical Implications
The findings support the utility of Two Minds Theory in digital health interventions. By addressing both conscious barriers (via the Narrative Mind) and non-conscious habits/emotions (via the Intuitive Mind), the tailored messages successfully countered the natural decay of exercise habits over time.

Limitations and Future Directions

• Confounding: The separation of MI effects from messaging effects was confounded by study site, requiring future research with a pure "messaging-only" arm.
• Initiation vs. Maintenance: The study only tested maintenance in participants who had already exercised for 16 weeks. It remains unknown if this approach works for initiating exercise in sedentary individuals.
• Attrition: While ITT analyses confirmed robustness, missing data remains a challenge in digital health studies.

Conclusion
Tailored messages, grounded in behavioral theory and delivered via mHealth, are superior to non-tailored educational messages for sustaining long-term physical activity in people with HIV. This approach offers a promising, cost-effective pathway to improve long-term health outcomes in an aging HIV-positive population.