A ring and a wrist device can disagree about the same night by forty-five minutes — and both can still be called accurate by the researchers who tested them. That sounds like a contradiction. It isn't.
Two instruments reading the same sleeping body from different places, through different hardware, will land in different spots on the fine-grained questions while agreeing almost perfectly on the coarse ones, and the useful thing is knowing which questions fall into which bucket. The evidence on that has gotten genuinely good. Researchers have put smart rings and wrist-worn trackers against polysomnography in sleep laboratories, against medical-grade actigraphy in people's own bedrooms, and against ambulatory electrocardiogram across a full day of ordinary life — named authors, named journals, numbers you can look up.
Deborah came at her own purchase the same way: slowly, by comparison. She met hydrogen in a genetic-testing course whose instructor sold a machine priced beyond reason, so Deborah went and looked at everything else instead. Not a quick decision. That posture is the right one for wearables too — find out what a device has been measured against before you trust what it tells you.
What the Finger Sees That the Wrist Doesn't
Almost every consumer wearable runs on photoplethysmography — PPG, an optical method that reads small changes in blood volume just under the skin. Heart rate, heart rate variability, oxygen-saturation estimates and the sleep staging that fills your morning dashboard all descend from that one signal. Alugubelli, Abuissa and Roka, reviewing the field in Sensors in 2022, set out how much of remote heart-rate and HRV monitoring now rests on PPG.
Why placement changes the signal
The finger carries a denser bed of vessels close to the surface than the wrist does, and a ring stays in contact with it whether you are still or moving. Wrist devices own the opposite advantages: a screen, a bigger battery envelope, satellite positioning, the exercise-tracking features people actually open during a workout. Form factor does most of the work there. A smart ring is minimalist by necessity — no display, nothing to swipe, discreet enough that most people forget it is on, which is the property that makes overnight wear effortless. A fitness watch is a small computer you also sleep in.
The Ring Against the Laboratory Standard
The most substantial validation to date is Svensson and colleagues in Sleep Medicine, 2024. Their team compared the Oura Ring Gen3, running its 2.0 sleep-staging algorithm, against multi-night ambulatory polysomnography in 96 participants — 421,045 thirty-second epochs of paired data, an unusually large pile of evidence here.
What 421,045 epochs showed
On the measures most people look at, the authors reported no significant difference from the laboratory reference: time in bed, total sleep time, sleep onset latency, sleep period time, wake after sleep onset, and both light and deep sleep. Sensitivity for detecting sleep ran 94.4% to 94.5%, specificity 73.0% to 74.6%, accuracy 91.7% to 91.8%. Read that again. A consumer device, measured against the standard clinicians use, holding its own.
The staging numbers, stated plainly
Two figures moved. Svensson's group reported that the ring underestimated sleep efficiency by 1.1 to 1.5 percent and REM by 4.1 to 5.6 minutes, and that stage-level accuracy ran from 75.5% for light sleep up to 90.6% for REM. Minutes, not hours — and the authors printed them alongside the agreement figures rather than behind them.
Three Devices, One Night, One Sleep Lab
Robbins and colleagues ran the comparison everyone wants. Their 2024 paper in Sensors followed 35 healthy adults through one night of inpatient polysomnography wearing an Oura Ring Gen3, a Fitbit Sense 2 and an Apple Watch Series 8 at once — same body, same reference.
Sleep versus wake: all three hold up
Sensitivity for separating sleep from wake came in at 95% or better for every device. All three. On the binary question — asleep, or awake — the hardware has arrived.
Stage by stage, the minutes diverge
Stage-level sensitivity ran from 50% to 86% — and that is where the forty-five minutes at the top of this article came from. Robbins's team reported the ring did not differ significantly from polysomnography on wake, light, deep or REM. The Fitbit Sense 2 overestimated light sleep by 18 minutes and underestimated deep sleep by 15. The Apple Watch Series 8 underestimated wake by 7 minutes and deep sleep by 43, and overestimated light sleep by 45.
Seven Days of Real Life — and a Twenty-Four-Hour Study of One Person
Laboratory nights are clean. Your Tuesday is not. Asgari Mehrabadi and colleagues took that seriously in JMIR mHealth and uHealth in 2020, running a commercially available smart ring and a smartwatch against medical-grade actigraphy on 45 healthy adults for seven days at home. They reported the ring's total sleep time correlating at r=0.86, wake after sleep onset at r=0.41 and sleep efficiency at r=0.47, against r=0.59 for the watch's total sleep time. Their stated conclusion: the ring outperformed the watch on the sleep parameters that do not involve staging.
Heart rate, measured against an ECG, in a study of one
Heart rate has its own validation story, and the most-quoted single-day study of it is far smaller than readers assume. Nelson and Allen, in JMIR mHealth and uHealth in 2019, used a single-subject design — n=1 — tracking 102,740 heartbeats against ambulatory electrocardiogram across sitting, walking, running, ordinary daily activity and sleep. The Apple Watch 3 came in at a mean difference of −1.80 bpm, a mean absolute percentage error of 5.86% and 95% agreement; the Fitbit Charge 2 at −3.47 bpm, 5.96% and 91% agreement. Both stayed inside ±10% in every condition except the Apple Watch during unstructured daily activities. One person, one day, 102,740 beats — a design that says something precise about a single body and deliberately nothing about the spread across a population, which is exactly how the authors framed it, and stating that out loud is the only way the rest of these numbers stay worth anything.
Weeks tell the truth. Single days are noise. Jerath, Syam and Ahmed, reviewing smartwatch and HRV technology in Sensors in 2023, described where the field is heading: devices that hand the wearer something to act on, not just a signal to store. We took that apart in decoding wearable data.
Your Ring Flagged a Rough Week. Now What?
Here the comparison runs out of road. A ring measures, a watch measures, neither intervenes. The markers flagged after a hard training block — a suppressed HRV trend, a higher resting heart rate, a run of fragmented nights — sit downstream of physiology rather than being the physiology, and one thread researchers keep pulling on underneath them is oxidative load.
Joseph knows the "now what" question from the other end of it. He found hydrogen through a video online, spent his first stretch with tablets and single-serve pouches off a grocery shelf, and his verdict on that stretch is the most honest sentence in his story: "I'm not sure if it worked or not, you know?"
Joseph had no instrument, no reference standard, nothing at all to compare against. Anyone who has watched a dashboard and wondered whether a number moved for a reason knows the feeling.
A selective antioxidant, not a blunt one
The foundational paper is Ohsawa and colleagues in Nature Medicine in 2007. Working in cell and rat models — not a human trial, and the paper says so — they reported that molecular hydrogen appeared to reduce the hydroxyl radical selectively, while not reacting with the other reactive oxygen species that carry physiological signaling roles. That selectivity is why athletes started paying attention, since high-dose non-selective antioxidants have in some trials blunted the very adaptations athletes are chasing. We pulled the distinction apart in comparing antioxidant strategies.
What the exercise trials reported
Aoki and colleagues published a pilot study — their word, and worth repeating — in Medical Gas Research in 2012, with 10 male elite soccer players in a double-blind crossover. They reported blood lactate did not climb on hydrogen-rich water the way it did on placebo, and that peak torque fell on placebo while holding at the early phase on hydrogen-rich water; d-ROMs, BAP and creatine kinase showed no significant change.
Botek and colleagues took it to resistance work. Their 2022 randomized, double-blind, placebo-controlled crossover in the Journal of Strength and Conditioning Research — 12 men — reported lunges performed faster on hydrogen-rich water (p<0.001), lactate lower both mid-session and immediately afterward (5.3 and 5.1 versus 6.5 and 6.3 mmol/L, p≤0.008), and lower muscle-soreness scores on a visual analogue scale at 24 hours (26 ± 11 versus 41 ± 20 mm, p=0.002).
Zhou K and colleagues then ran eight days of intake in Frontiers in Physiology in 2024 with 18 trained men, reporting significantly higher total power output (50,866.7 versus 46,431.0 W, p=0.032) and more total repetitions (78.2 versus 70.3, p=0.019), with no significant difference in countermovement jump, soreness scores or perceived readiness. Zhou Q and colleagues, reviewing the literature from 1980 through April 2024 in Metabolites, lined the proposed mechanisms up across endurance, strength and recovery. Three research groups, three protocols, one direction of travel. More on the athlete side in hydrogen water for athletes and exercise recovery.
Professional-Strength Hydrogen Water: Concentration and Purity Together
The category argues about parts per million as though one number settled everything. Concentration matters. Purity matters at least as much — a device you drink from every single morning delivers whatever else is in the water along with the hydrogen, and the published trials used water produced under controlled, research-grade conditions. Professional-strength hydrogen water means both at once: enough dissolved hydrogen to match the protocols the researchers used, and a purity profile most of the category cannot document.
Given those two criteria, here is how the Lourdes Hydrofix Premium Edition addresses them. It runs a separate-chamber (dual-chamber) electrolysis system with a multi-layer fibriform polymer membrane and solid — not plated — high-purity titanium and platinum electrodes, the titanium certified TP270C at 99.928% purity on an independent metallurgical certificate (Certificate No. 17-MANS-0078-B). Output is approximately 120 mL/min of hydrogen gas, depending on usage conditions, with Masa International Corp., a third-party testing lab, measuring up to approximately 134.2 mL/min under test conditions (Test No. MM03-6024-01). On purity, Japan Food Research Laboratories (Certificate No. 23028707001-0201) found selected plasticizers, BPA, iron and titanium not detected. The water comes out pH neutral, within ±0.1 of the source.
You can find the Lourdes Hydrofix in our best hydrogen water system collection.
Every certificate number in this article is one you can look up on our Certifications page — publishing the documents rather than merely citing them was the editorial standard we set for ourselves, and it is the same standard that makes a validation paper worth more than a spec sheet. Every unit is tested before it leaves the factory and ships with its own certificate of authenticity.
Materials were exactly where Deborah's questions landed. Price was one; the other was the metal — she wanted to know what she would be drinking out of every day, and she wanted it shown to her. She weighed her instructor's model and a friend's machine, both priced above the Lourdes.
Then Deborah did the thing most buyers skip. She tracked down a woman who owned both machines and talked with her at length; that owner told her the Lourdes was the easier of the two to look after, which mattered, because Deborah does not enjoy complicated technology. She runs it for about an hour and a half each morning while she meditates, sometimes drinking the water with her supplements.
Joseph's turning point was the same kind of conversation. A long phone call with another owner moved him further than any specification sheet had (he had read plenty of those), and the machine being made in Japan added to his confidence. He drinks roughly a liter a day now, mostly mornings. "It's not really a moment, it's more like a lifestyle," he says, and calls himself "a happy consumer."
Two Tools, Two Jobs
A smart ring, a fitness watch and a glass of hydrogen-rich water are not competing for the same job. Svensson's 96 participants, Robbins's 35, Asgari Mehrabadi's 45 and Nelson's single subject all describe instruments — things that watch, estimate and report. The hydrogen literature describes something people drink, examined in small, tightly controlled trials whose full numbers are on the page.
So: pick the form factor that fits the finger or wrist you will actually keep it on, read the weekly trend rather than the Tuesday reading, and take the staging minutes as the estimates researchers say they are. Deborah spent a year with her decision and would make it again — "I had already decided that the Lourdes Machine is the one I would buy," she recalls of the moment the comparing stopped, and the Lourdes Hydrofix has been part of her morning ever since. Joseph, who once had no way to tell anything, now has a habit he no longer thinks about. Measurement and routine, each doing its own work.
Holy Hydrogen products, including the Lourdes Hydrofix Premium Edition, are not medical devices and are not intended to diagnose, treat, cure, or prevent any disease. All information on this site is provided for educational and general wellness purposes only and should not be considered medical advice. Always consult a qualified healthcare provider before beginning any new wellness practice, especially if you have a medical condition, are pregnant or nursing, or take prescription medications.
Further Reading
- Alugubelli N, Abuissa H, Roka A. "Wearable Devices for Remote Monitoring of Heart Rate and Heart Rate Variability—What We Know and What Is Coming." Sensors, 2022. PMID: 36433498 — a review of the optical sensor behind all of these devices.
- Jerath R, Syam M, Ahmed S. "The Future of Stress Management: Integration of Smartwatches and HRV Technology." Sensors, 2023. PMID: 37687769 — a review of where wrist-worn HRV measurement is heading.
- Svensson T, et al. "Validity and reliability of the Oura Ring Generation 3 with sleep staging algorithm 2.0 compared to multi-night ambulatory polysomnography." Sleep Medicine, 2024. PMID: 38382312 — read this one first: a hundred people, hundreds of thousands of epochs, every agreement statistic printed.
- Robbins R, et al. "Accuracy of Three Commercial Wearable Devices for Sleep Tracking in Healthy Adults." Sensors, 2024. PMC11511193 — the rare head-to-head, with per-stage minute gaps laid out device by device.
- Nelson BW, Allen NB. "Accuracy of Consumer Wearable Heart Rate Measurement During an Ecologically Valid 24-Hour Period." JMIR mHealth and uHealth, 2019. PMC6431828 — one volunteer, one day, a hundred thousand beats against a medical monitor.
- Zhou Q, et al. "Hydrogen-Rich Water to Enhance Exercise Performance: A Review of Effects and Mechanisms." Metabolites, 2024. PMC11509640 — a review collecting four decades of exercise work and the proposed mechanisms.
References
Svensson T, Madhawa K, Nt H, Chung UI, Svensson AK. "Validity and reliability of the Oura Ring Generation 3 (Gen3) with Oura sleep staging algorithm 2.0 (OSSA 2.0) when compared to multi-night ambulatory polysomnography." Sleep Medicine. 2024;115:251-263. PMID: 38382312; DOI: 10.1016/j.sleep.2024.01.020
Robbins R, Weaver MD, Sullivan JP, et al. "Accuracy of Three Commercial Wearable Devices for Sleep Tracking in Healthy Adults." Sensors (Basel). 2024;24(20):6532. PMID: 39460013; PMC11511193; DOI: 10.3390/s24206532
Asgari Mehrabadi M, Azimi I, Sarhaddi F, et al. "Sleep Tracking of a Commercially Available Smart Ring and Smartwatch Against Medical-Grade Actigraphy in Everyday Settings." JMIR mHealth and uHealth. 2020;8(10):e20465. PMID: 33038869; PMC7669442; DOI: 10.2196/20465
Nelson BW, Allen NB. "Accuracy of Consumer Wearable Heart Rate Measurement During an Ecologically Valid 24-Hour Period: Intraindividual Validation Study." JMIR mHealth and uHealth. 2019;7(3):e10828. PMID: 30855232; PMC6431828; DOI: 10.2196/10828
Alugubelli N, Abuissa H, Roka A. "Wearable Devices for Remote Monitoring of Heart Rate and Heart Rate Variability—What We Know and What Is Coming." Sensors (Basel). 2022;22(22):8903. PMID: 36433498; PMC9695982; DOI: 10.3390/s22228903
Jerath R, Syam M, Ahmed S. "The Future of Stress Management: Integration of Smartwatches and HRV Technology." Sensors (Basel). 2023;23(17):7314. PMID: 37687769; PMC10490434; DOI: 10.3390/s23177314
Ohsawa I, Ishikawa M, Takahashi K, et al. "Hydrogen acts as a therapeutic antioxidant by selectively reducing cytotoxic oxygen radicals." Nature Medicine. 2007;13(6):688-694. PMID: 17486089; DOI: 10.1038/nm1577
Aoki K, Nakao A, Adachi T, Matsui Y, Miyakawa S. "Pilot study: Effects of drinking hydrogen-rich water on muscle fatigue caused by acute exercise in elite athletes." Medical Gas Research. 2012;2:12. PMID: 22520831; PMC3395574; DOI: 10.1186/2045-9912-2-12
Botek M, Krejčí J, McKune A, Valenta M, Sládečková B. "Hydrogen Rich Water Consumption Positively Affects Muscle Performance, Lactate Response, and Alleviates Delayed Onset of Muscle Soreness After Resistance Training." Journal of Strength and Conditioning Research. 2022;36(10):2792-2799. PMID: 33555824; DOI: 10.1519/JSC.0000000000003979
Zhou K, Yuan C, Shang Z, Jiao W, Wang Y. "Effects of 8 days intake of hydrogen-rich water on muscular endurance performance and fatigue recovery during resistance training." Frontiers in Physiology. 2024;15:1458882. PMID: 39434721; PMC11491356; DOI: 10.3389/fphys.2024.1458882
Zhou Q, Li H, Zhang Y, Zhao Y, Wang C, Liu C. "Hydrogen-Rich Water to Enhance Exercise Performance: A Review of Effects and Mechanisms." Metabolites. 2024;14(10):537. PMID: 39452918; PMC11509640; DOI: 10.3390/metabo14100537