Table of Contents
Kids Wearables Evidence: What Health Tracking Really Does
The kids wearables evidence now includes 21 randomised trials and 3,676 children. Trackers add steps but not harder exercise. What that means at home.
The kids wearables evidence is now good enough to answer the question most parents actually ask, and the answer is specific: a tracker on a childâs wrist increases how many steps they take and does not increase how much hard exercise they do. That comes from a 2024 systematic review and meta-analysis in The Lancet Digital Health pooling 21 randomised controlled trials and 3,676 young people with a mean age of 13.7 years. More walking, same intensity. Whether that is worth the device, the subscription and the data depends on what you were hoping for.
Key Takeaways
- Au and colleagues (2024) pooled 21 randomised trials covering 3,676 young people. Daily steps rose with a standardised mean difference of 0.37 (95% CI 0.09 to 0.65, p=0.013). Moderate-to-vigorous physical activity did not change (â0.08, 95% CI â0.18 to 0.02, p=0.11).
- The authors explicitly called for more rigorously designed trials with less missing data before treating the step finding as settled.
- Sleep tracking is weaker than activity tracking. In seven consumer devices tested against laboratory polysomnography, sensitivity for detecting sleep was 0.93 or higher while specificity for detecting wakefulness ran 0.18 to 0.54.
- The problem being addressed is real: WHO reported in June 2024 that 81% of adolescents aged 11 to 17 were physically inactive, 85% of girls and 78% of boys.
- Data handling is a separate question from efficacy, and the enforcement record is not reassuring. The FTC and DOJ fined Amazon $25 million in 2023 over childrenâs voice and geolocation data retention.
The strongest study, read carefully
A systematic review and meta-analysis pools results from multiple trials to estimate an overall effect. Au, Recchia, Fong, Wong, Chan, Capio, Yu, Wong, Sit, Ip, Chen, Thompson and Siu published theirs in The Lancet Digital Health in 2024, volume 6, issue 9, pages e625 to e639.
The numbers that matter:
Twenty-one randomised controlled trials. Three thousand six hundred and seventy-six young people. Mean age 13.7 years. Objectively measured daily steps increased, with a standardised mean difference of 0.37 and a 95% confidence interval from 0.09 to 0.65 (p=0.013). Moderate-to-vigorous physical activity showed a standardised mean difference of â0.08, confidence interval â0.18 to 0.02 (p=0.11), which is a null result.
A standardised mean difference of 0.37 is a small-to-moderate effect. The confidence interval runs from 0.09, which is almost nothing, to 0.65, which is meaningful. That width tells you the trials disagreed with each other.
The authorsâ own conclusion is appropriately cautious: more rigorously designed trials that minimise missing data are warranted to validate the positive findings on steps and to explore possible long-term effects. In other words, the step finding is real but not yet durable, and nobody has shown it persists.
The split between steps and intensity is the interesting part. A step count is something a device can show you, so it is what children optimise. Moderate-to-vigorous activity requires getting out of breath, which no notification can produce.
What each marketing claim is actually supported by
| The claim | What the evidence says | Verdict |
|---|---|---|
| Increases physical activity | 21 RCTs, 3,676 young people: steps up (SMD 0.37), intensity unchanged | Partly supported, for steps only |
| Improves sleep | No comparable meta-analysis in children; device wake detection is poor | Not established |
| Spots health problems early | Consumer trackers are wellness products, not cleared diagnostic devices | No |
| Builds body awareness | No good evidence in either direction | Unknown |
| Motivates through streaks and badges | Engagement findings are mostly short-term and self-reported | Weak |
| Keeps family data private | Depends entirely on the vendor; enforcement record includes a $25M childrenâs data penalty | Vendor-specific |
The middle rows are where most purchase decisions are actually made, and they are the rows with the least behind them.
Why sleep numbers deserve more scepticism than step numbers
A step is a discrete physical event an accelerometer can count. A sleep stage is an inference from heart rate, movement and timing, and the inference is harder.
Chinoy and colleagues tested seven consumer sleep-tracking devices against polysomnography, the electrode-based laboratory standard, in a 2021 paper in Sleep (volume 44, issue 5, article zsaa291). Every device detected sleep well, with sensitivity at or above 0.93. Detecting wakefulness was much worse, with specificity from 0.18 to 0.54. Total sleep time was off by 2.6 minutes in the best case and by 43.7 and 46.8 minutes for two Garmin models. All six devices that reported stages differed significantly from the laboratory on light sleep.
That study used adults. Smaller wrists and higher resting heart rates make the problem harder in children, not easier, and comparable paediatric validation is thin.
What this means at a practical level: a device can tell you your child was in bed for six hours instead of nine. It cannot reliably tell you how much of that was deep sleep. The first fact is actionable. The second is a number on a screen.
The actionable comparison is already published. The CDCâs sleep recommendations, last reviewed May 15, 2024 and drawn from American Academy of Sleep Medicine guidance, give 9 to 12 hours for ages 6 to 12 and 8 to 10 hours for ages 13 to 17. A bedtime and a wake time measure compliance with that for free.
The problem that makes trackers tempting in the first place
None of the above means children are fine.
The WHO physical activity fact sheet, updated June 26, 2024, reports that 81% of adolescents aged 11 to 17 were physically inactive, with girls at 85% and boys at 78%. That is a population-level failure with serious downstream consequences.
How serious was set out by the American Heart Association. A 2016 scientific statement in Circulation led by Robert Ross, running from page e653 to e699 and titled âImportance of Assessing Cardiorespiratory Fitness in Clinical Practice: A Case for Fitness as a Clinical Vital Sign,â argued that low cardiorespiratory fitness is associated with high risk of cardiovascular disease, all-cause mortality and several cancer mortality rates, and may predict mortality better than traditional risk factors including smoking and hypertension.
So the target is right. The question is whether a $99 band with an app is the right instrument, and the meta-analysis says it moves steps and not intensity.
The data question, which is separate
Efficacy and privacy are independent. A device could work well and handle data badly, or the reverse.
The relevant precedent for children specifically is the FTC and Department of Justice action against Amazon on May 31, 2023. It produced a $25 million civil penalty over allegations that childrenâs Alexa voice recordings and geolocation data were retained indefinitely, parentsâ deletion requests were undermined, and the retained data was used to improve algorithms. Samuel Levine of the FTC: âCOPPA does not allow companies to keep childrenâs data forever for any reason, and certainly not to train their algorithms.â
Commercial pressure is also real and worth naming. Oura shelved a $2.2 billion initial public offering on September 29, 2026 despite 5.7 million paying members and projected 90% revenue growth, and its filings show memberships at roughly 20% of sales carrying an 89% gross margin. Companies under that kind of subscription pressure have incentives that do not always align with a nine-year-oldâs interests. Our analysis of what the wearables market tells families goes further into those economics.
What to do at home
Decide what you want before you buy
If the goal is more walking, a tracker has modest support and so does a dog, a bike and a walk to school. If the goal is better sleep, change the bedtime and remove the phone from the room; the device will not do it. If the goal is a clinical answer, see a clinician.
Use a tracker for a defined experiment, not indefinitely
Six weeks, one question, one measurement. âDoes walking to school raise daily steps?â Then take the device off. Open-ended wear for a child produces open-ended data collection and, in some children, open-ended self-monitoring.
Hide the scores and keep the counts
Most platforms allow turning off readiness, recovery and sleep scores while leaving steps and heart rate visible. Scores are proprietary composites with no validation in children. Counts are measurements. Our guide to what smartwatches collect and what kids should know covers the settings in detail.
Watch for the specific children for whom this goes wrong
A child already anxious about food, body or performance can convert any number into a target. For that child, the balance of the evidence â a small step effect and no intensity effect â does not justify the risk. There is no version of âdeep sleep: 41 minutesâ that helps a worried thirteen-year-old.
Ask the two data questions before setting up an account
Can I export the raw data, and can I delete the account and have the history removed? If the answers are not clearly yes, choose a different product. Our buying guide for kidsâ wearables has the full checklist.
What not to do
Do not use a tracker to monitor a childâs activity without telling them what you are looking at. A wearable is a reporting device pointed at a person, and discovering that quietly is a different experience from agreeing to it.
What to Watch For Over the Next 3 Months
- Week 4: Check whether the device is still being worn. Attrition in wearable studies is high enough that the meta-analysis authors specifically flagged missing data as a limitation. Most children stop within a month.
- Month 2 red flags: Your child checking the score before an activity rather than after it. Negotiating steps rather than taking them. Comparing numbers with friends. Any of these means the device has become the goal.
- Month 3 self-check: Compare one concrete behaviour from before and after: walks per week, bedtime on school nights, minutes of sport. If none changed, the device measured the status quo, which is not worth a subscription.
Frequently Asked Questions
Do fitness trackers actually make children more active?
Partly. Pooling 21 randomised trials with 3,676 young people, daily steps increased with a standardised mean difference of 0.37, while moderate-to-vigorous physical activity showed no significant change. More walking, not harder exercise, and the authors asked for better trials before treating even that as settled.
At what age is a wearable appropriate?
The trial evidence centres on a mean age of 13.7 years, so there is essentially no efficacy evidence for younger children. For under-tens, treat a step counter as a toy rather than a health device, and skip anything that produces a health score.
Can a smartwatch tell me if my child is sleeping badly?
It can tell you roughly how long they were asleep, which is the useful part. Sleep stages and wakefulness are measured poorly: across seven consumer devices tested against laboratory polysomnography, specificity for detecting wake ranged from 0.18 to 0.54.
Is the data on these devices safe?
It varies by company and the record includes real failures, including a $25 million FTC and DOJ penalty against Amazon in 2023 over childrenâs voice and geolocation data. Before creating an account, confirm you can export the raw data and delete the history.
What works better than a tracker?
The interventions with the best support are structural rather than technological: active travel to school, a sport the child chose, a consistent bedtime, and phones charging outside bedrooms. WHOâs finding that 81% of 11-to-17-year-olds are inactive is a problem of opportunity and time, and a wrist device does not change either.
About the author
Ricky Flores is the founder of HiWave Makers and an electrical engineer with 15+ years of experience building consumer technology at Apple, Samsung, and Texas Instruments. He writes about how kids learn to build, think, and create in a tech-saturated world. Read more at hiwavemakers.com.
Sources
- Au, W. W., Recchia, F., Fong, D. Y., Wong, S. H. S., Chan, D. K. C., Capio, C. M., Yu, C. C. W., Wong, S. W. S., Sit, C. H. P., Ip, P., Chen, Y. J., Thompson, W. R., & Siu, P. M. (2024). âEffect of wearable activity trackers on physical activity in children and adolescents: a systematic review and meta-analysis.â The Lancet Digital Health, 6(9), e625âe639. https://doi.org/10.1016/S2589-7500(24)00139-0
- Chinoy, E. D., Cuellar, J. A., Huwa, K. E., Jameson, J. T., Watson, C. H., Bessman, S. C., Hirsch, D. A., Cooper, A. D., Drummond, S. P. A., & Markwald, R. R. (2021). âPerformance of seven consumer sleep-tracking devices compared with polysomnography.â Sleep, 44(5), zsaa291. https://doi.org/10.1093/sleep/zsaa291
- World Health Organization. (2024, June 26). âPhysical activity.â WHO Fact Sheet. https://www.who.int/news-room/fact-sheets/detail/physical-activity
- Ross, R., et al. (2016). âImportance of Assessing Cardiorespiratory Fitness in Clinical Practice: A Case for Fitness as a Clinical Vital Sign: A Scientific Statement From the American Heart Association.â Circulation, 134, e653âe699. https://doi.org/10.1161/CIR.0000000000000461
- Centers for Disease Control and Prevention. (2024, May 15). âAbout Sleep.â CDC. https://www.cdc.gov/sleep/about/index.html
- Federal Trade Commission & U.S. Department of Justice. (2023, May 31). âFTC and DOJ Charge Amazon with Violating Childrenâs Privacy Law by Keeping Kidsâ Alexa Voice Recordings Forever and Undermining Parentsâ Deletion Requests.â FTC Press Release. https://www.ftc.gov/news-events/news/press-releases/2023/05/ftc-doj-charge-amazon-violating-childrens-privacy-law-keeping-kids-alexa-voice-recordings-forever
- TechCrunch. (2026, September 29). âOura shelves its $2.2B IPO, citing âuncertaintyâ in the market.â TechCrunch. https://techcrunch.com/2026/09/29/oura-shelves-its-2-2b-ipo-citing-uncertainty-in-the-market/