Should Your Fitness App Track MVPA Instead of Calories?

Your fitness app can count every step, minute, and calorie estimate it can scrape from your wrist. That does not mean each number tells you equally useful information. Physical-activity guidelines focus on moderate-to-vigorous physical activity (MVPA): activity intense enough to raise breathing and heart rate, not simply time spent moving or a calorie-burn total.

A 2025 meta-analysis by Lee et al. was designed to answer a useful question for adults with overweight or obesity: do digital interventions work differently when trials measure intensity-qualified outcomes, such as MVPA minutes and threshold-based steps, versus energy-expenditure outcomes, such as MET-minutes or kilocalories per week?

The honest answer is that we cannot yet say what that review found on this point. The accessible publication text describes the question and methods, but does not report the effect estimates, confidence intervals, heterogeneity statistics, or age-subgroup results needed to verify the conclusion. The idea that MVPA metrics show a larger response is a plausible hypothesis, not an established result of this review.

Why MVPA is a more useful target than calorie burn

For health guidance, intensity matters. The usual benchmark is at least 150 minutes a week of moderate-intensity activity, with vigorous activity counting more heavily. A brisk walk, cycling at a purposeful pace, or a workout that makes conversation possible but less leisurely can qualify; the exact threshold depends on the person and activity.

Calories are a noisier coaching target. Wearables estimate energy expenditure from movement, heart rate, body size, and proprietary algorithms. Two people can earn different calorie estimates during the same walk, and the number can invite a familiar trap: treating exercise as a receipt for food. It also does not tell you whether the activity reached a guideline-relevant intensity.

That does not make steps or calorie estimates useless. Step counts can make activity visible and can help people build a walking habit. The stronger claim is narrower: if an app asks you to choose one health-oriented activity goal, a goal tied to MVPA or a personalized intensity threshold better matches public-health guidance than a generic calorie target.

What the primary review actually set out to test

Lee et al. conducted a PRISMA-compliant systematic review and meta-analysis of 15 randomized controlled trials published from 2010 through 2024. The trials enrolled adults aged 18 to 65 with a BMI of at least 25 and tested websites, mobile apps, or text-message programs intended to increase physical activity. Control groups received usual care or were placed on waitlists.

The review treated physical activity as the outcome, rather than folding it into a weight-management score. Its prespecified outcomes included MVPA minutes, step counts, MET-minutes per week, and kilocalories per week. It also planned age-band analyses for adults aged 18–34, 35–49, and 50–64.

That design matters because these outcomes do not measure the same thing. MVPA and threshold-based steps ask whether someone is doing activity at a meaningful intensity. MET-minutes and calorie totals aggregate estimated energy expenditure. A digital prompt that gets someone to turn a slow stroll into a brisk one could improve MVPA without producing a dramatic shift in weekly calorie estimates.

Still, a sensible mechanism is not a result. Without the review's reported subgroup estimates, no one should claim that an MVPA-focused app has proved superior to a calorie-focused app for this population.

Digital programs can help, but their average effect is modest

The broader evidence supports a more restrained conclusion: digital interventions can increase physical activity in people with overweight or obesity, but effects vary and should not be sold as a substitute for an exercise environment, clinician, coach, or social support system.

A 2024 Cochrane review of smartphone interventions in adolescents and adults with overweight or obesity found modest but detectable improvements in physical activity. Cochrane's assessment also places limits on the certainty of this literature: evidence quality commonly falls in the low-to-moderate range. A statistically detectable change may still be too small, too short-lived, or too dependent on the measurement method to transform health outcomes.

A 2021 meta-analysis of long-term mHealth randomized trials likewise found that digital physical-activity interventions can work, while raising concern that effects may attenuate with extended follow-up. That is the recurring problem with notifications: they can prompt action, but they do not automatically create a durable routine after novelty wears off.

Longer trials give the same caution. The Evident 3 randomized trial followed adults with overweight or obesity for 18 months using a smartphone app plus a smart band. The authors characterized long-term effectiveness as modest. Hardware can make measurement easier; it cannot make adherence inevitable.

Age may change the response, but the key estimates remain unknown

Lee et al. specifically tested whether effects differed across three working-age bands. This is a reasonable question. Comfort with apps, baseline activity, motivation, physical limitations, and preferences for support can differ with age. But plausibility does not tell us the size or direction of a subgroup effect.

Evidence outside Lee et al.'s 18–65 range shows that digital programs do not stop working at 65. A 2025 systematic review and meta-analysis in adults aged 65 and older found that mHealth interventions improved physical activity and body-composition outcomes. Its population was broader than adults with overweight or obesity, so it cannot validate an age effect in Lee et al.'s target group.

Similarly, a 2021 meta-analysis in healthy adults over 55 provides relevant context but not a direct comparison. A nonrandomized 12-week study of a peer-supported app in community-dwelling adults 65 and older found the approach feasible and reported improvements in activity and physical function, but participants chose their group. That design cannot separate the app's effect from differences between people who elected to use it and those who did not.

Features matter more than a prettier dashboard

Many digital programs combine tracking with reminders, goal setting, feedback, social features, or gamification. A 2025 randomized trial of a gamified digital physical-activity program for obesity management examined this design against usual care, but the available summary does not provide extractable physical-activity effect estimates. Gamification may help engagement; it has not earned a blanket claim of superior activity outcomes here.

Human contact may be more consequential. The Cochrane evidence and the design rationale for the Spark digital weight-loss trial both support modest expectations for fully automated programs. Spark is a protocol, not outcome evidence, and its cited 2% to 4% weight-loss range applies to fully digital weight-loss interventions, not directly to physical-activity change. Its practical lesson is still fair: a tracker is usually a starting tool, not a personal trainer in a rectangle.

How to use this evidence without overselling it

• Set a weekly MVPA goal, or use an intensity-based target your clinician or qualified trainer has tailored to your health status.
• Use steps as a behavior cue when they get you walking more, but do not mistake a step total for a full measure of exercise intensity.
• Treat calorie-burn estimates as rough estimates, not as a precise budget.
• Choose an app that offers feedback, planning, and support you will actually use. If you repeatedly disengage, adding a coach, class, walking partner, or clinician may matter more than switching dashboards.
• Judge progress over months. A short-term rise in app-recorded activity does not establish durable behavior change or a clinical benefit.

The evidence supports digital tools as modest aids to physical activity. It does not yet prove that tracking MVPA beats tracking calories or MET-minutes in adults with overweight or obesity.

The unanswered metric question is worth studying because it could improve how apps define success. For now, choose an intensity-aware goal because it matches activity guidelines, not because a 2025 meta-analysis has already settled the contest.