What Did the 72-Week Hydrogen-Water Trial Actually Tell Us?
This article focuses on one Parkinson’s disease study that has been quoted repeatedly as evidence against hydrogen water: the 72-week randomized, double-blind, multicenter trial published in Movement Disorders. The trial is important, and its neutral primary result should not be ignored. But I also think the details matter—especially the H₂ detected in the placebo water, how dissolved hydrogen was measured, the unexpectedly small change in UPDRS scores in both groups, and how this one result fits within the broader body of promising preclinical evidence and the earlier positive human clinical signal.
My goal is not to make a neutral study positive. My goal is to make sure we are interpreting exactly what was tested and not turning one protocol into a blanket conclusion about molecular hydrogen, hydrogen water, or Parkinson’s disease.
H2HUBB Takeaway
- The 72-week trial was neutral. It did not find a statistically significant difference in total UPDRS change between the hydrogen-water and placebo groups.
- The placebo water was not completely H₂-free. The authors reported approximately 0.16 ± 0.16 ppm H₂ in the placebo water and considered that concentration sufficiently low to function as placebo.
- Low H₂ should not automatically be assumed biologically meaningless. In a separate Parkinson’s mouse model, 0.08 ppm hydrogen water produced neuroprotection close to that seen with much higher H₂ concentrations under that animal protocol.
- That does not prove the placebo altered the human trial. It does mean the assumption that low-concentration H₂ is necessarily inert deserves scrutiny.
- H2HUBB’s independent testing raised measurement concerns with an ORP-based Trustlex meter. Our testing involved the ENH-2000, not the ENH-1000 used in the Parkinson’s trial, so it cannot prove the study’s measurement was wrong. It does show why dissolved-H₂ measurement methodology deserves attention.
- I do not use this trial as a verdict on the entire field. I use it as one important piece of evidence that should be interpreted alongside the positive 48-week trial, preclinical studies, inhalation research, and newer mechanistic data.
In This Article
What the 72-Week Parkinson’s Hydrogen-Water Trial Found
The study enrolled 178 people with Parkinson’s disease across 14 hospitals in Japan. Participants were randomized to hydrogen water or placebo water, and the study followed changes in the Unified Parkinson’s Disease Rating Scale (UPDRS) for 72 weeks.
The published result was neutral. The mean change in total UPDRS was approximately 1.6 ± 14.0 points in the H₂-water group and 0.8 ± 9.6 points in the placebo group, and the investigators reported no statistically significant difference between the groups. They also did not find significant between-group differences in the reported UPDRS subscales, Hoehn and Yahr stage, or PDQ-39.
That finding matters. If someone asks me what the study reported, I am not going to tell them the trial was positive. It was not.
Where I disagree with some of the online discussion is what comes next. A neutral result does not automatically tell us that all hydrogen water, all H₂ concentrations, all dosing patterns, and all Parkinson’s disease stages are ineffective. To decide what a study means, we have to look inside the protocol.
Read the 2018 multicenter trial ↗
The Placebo Water Contained Molecular Hydrogen
One of the details that immediately caught my attention is that the placebo water was not completely free of H₂.
I understand the logic. If the investigators believed 0.16 ppm was too low to produce meaningful biological effects in people with Parkinson’s disease, then it could function as a placebo.
The problem is that we already had Parkinson’s animal data suggesting that low concentrations of H₂ are not necessarily biologically inactive.
A Parkinson’s Model Reported Effects at Only 0.08 ppm H₂
In the 2009 MPTP mouse study, researchers compared several concentrations of hydrogen water. They reported that 0.08 ppm H₂ produced almost the same protective effect on dopaminergic neuronal loss as saturated H₂ water around 1.5 ppm under that experimental protocol.
That finding is important for interpreting the placebo issue because 0.08 ppm is lower than the average 0.16 ppm reported in the placebo water of the later human trial.
What This Does—and Does Not—Mean
It does not prove that 0.16 ppm H₂ in the placebo water produced a clinical therapeutic effect in people with Parkinson’s disease. A mouse neurotoxin model cannot establish a human therapeutic threshold.
What it does show is that I would not automatically describe low-concentration H₂ as biologically meaningless. The relationship between H₂ concentration and biological response is not always a simple linear dose-response curve.
Read the 2009 low-concentration hydrogen-water Parkinson’s study ↗
Why I Also Question the Dissolved-Hydrogen Measurement Method
The trial reported H₂ measurements using an oxidation-reduction-potential-based Trustlex ENH-1000 method. This is another area where I think we should be careful.
H2HUBB later independently tested a Trustlex ENH-2000, an ORP-based dissolved-hydrogen meter from the same product family. In our testing, changing pH and TDS produced a much wider range in meter readings than the comparison H2Blue titration method. Across the test conditions, the ORP-based meter produced about a 70% measurement range, while the H2Blue titration results varied by about 8%.
This is consistent with the concern I raised years ago: an ORP-derived hydrogen estimate can be influenced by water chemistry, which makes me cautious about using one meter reading as absolute proof of dissolved H₂ concentration.
For me, this is especially important because H₂ escapes water rapidly. If the active and placebo water were prepared, stored, transported, and consumed over many months, I want to know how often H₂ concentrations were independently verified, how stable the products were, and whether the measurement method was sensitive to the chemistry of the water.
Both Groups Changed Surprisingly Little on the UPDRS
The authors themselves noted another unusual feature: the mean total UPDRS change remained within roughly three points in both groups over the study period.
They compared that with larger deterioration seen in placebo groups from other Parkinson’s trials, including DATATOP and QE3. That does not establish that hydrogen slowed progression in both groups. There are many reasons different Parkinson’s trials can progress differently, including patient selection, disease stage, medications, follow-up, and measurement variability.
Still, I think it is a legitimate observation. If the placebo group received measurable H₂ and both groups deteriorated less than expected based on selected historical comparisons, it is reasonable to ask whether the comparator was as biologically inactive as assumed. It is a hypothesis—not a conclusion.
The Earlier 48-Week Trial Produced a Different Human Result
Five years earlier, the same general research group published a much smaller randomized, double-blind pilot study. Participants drank 1,000 mL/day of hydrogen water for 48 weeks.
In that study, the H₂-water group improved in total UPDRS while the placebo group worsened, and the difference between groups reached statistical significance. The sample was very small—only 17 participants were included in the reported analysis—so I would never treat it as definitive.
| Study | Participants | Duration | Reported Result | Main Interpretation |
|---|---|---|---|---|
| 2013 Pilot | Small single-center trial; reported H₂ n=9 and placebo n=8 | 48 weeks | H₂ group improved in total UPDRS while placebo worsened; significant between-group difference. | Promising signal, but too small to be definitive. |
| 2018 Multicenter Trial | 178 participants across 14 hospitals | 72 weeks | No significant between-group difference in total UPDRS. | Stronger sample size and multicenter design, but the placebo-H₂ and protocol details deserve consideration. |
Read the 2013 randomized pilot trial ↗
What the Later Parkinson’s H₂ Research Adds
When I originally discussed the 72-week study, the Parkinson’s hydrogen literature was much smaller. We now have additional evidence that helps put the trial into a broader context.
Short Low-Dose H₂ Inhalation
A randomized crossover trial used an effective inhaled H₂ concentration around 1.2–1.4% for 10 minutes twice daily over four weeks. Clinical PD measures did not significantly improve, but urinary 8-OHdG increased about 16%, which the authors discussed in relation to hormetic cellular adaptation.
16-Week Pilot Trial
Participants inhaled H₂ for one hour twice daily. No significant UPDRS benefit was detected, but no adverse events were reported. Five of 20 participants were excluded for inadequate inhalation time, and the authors stated that the final study was underpowered.
Stage-Dependent Biological Effects
Hydrogen-saturated saline improved several outcomes in early and medium stages, reduced ROS and alpha-synuclein, and influenced autophagy-related signaling. The same benefit was not evident in the late-stage animals.
2% H₂ Reduced Dyskinesia
In rats, a one-hour 2% H₂ inhalation reduced abnormal involuntary movements without impairing L-DOPA’s locomotor benefit and was associated with reduced microglial activation and inflammatory signaling.
2021 randomized inhalation pilot ↗
2021 rotenone Parkinson’s model ↗
2023 L-DOPA dyskinesia study ↗
My Current Interpretation of the 72-Week Trial
I think we should report the study exactly as it was published: the 72-week multicenter trial did not find a statistically significant clinical advantage for its hydrogen-water group over its placebo group.
I also think it would be a mistake to jump from that statement to “hydrogen water has no therapeutic potential for Parkinson’s disease.”
The placebo contained measurable H₂. A Parkinson’s animal model had previously reported biological effects at an even lower H₂ concentration. The H₂ measurement method raises legitimate technical questions. Both clinical groups changed less on total UPDRS than some historical placebo comparisons. The earlier 48-week trial was positive. Later inhalation trials did not demonstrate the same clinical benefit under their specific protocols, while later animal studies continued to report promising biological effects and a newer human inhalation case report described symptomatic improvement.
When I put all of that together, I do not see a field that has been disproven. I see evidence that molecular hydrogen has shown therapeutic potential for Parkinson’s disease, with the strongest consistency currently coming from the preclinical literature and an encouraging earlier human trial. What remains unresolved is the extent of that potential in people and the protocol needed to demonstrate it reliably.
A 2025 review in Frontiers in Neuroscience reached a similar overall position: molecular hydrogen holds potential as a neuroprotective agent in Parkinson’s disease, while larger, well-designed clinical trials are still needed to establish efficacy and optimize delivery.
That is why I think dose, route, timing, disease stage, measurement method, adherence, and biological endpoints still need to be worked out.
What I Think the Next Trial Should Do Better
- Use a comparator verified to be H₂-free with a validated dissolved-H₂ method.
- Measure H₂ repeatedly throughout the study rather than relying on an initial preparation value.
- Report the actual total H₂ dose delivered per day, not only the volume of water consumed.
- Stratify participants by disease stage and treatment status.
- Track adherence carefully.
- Include mechanistic biomarkers alongside UPDRS and quality-of-life scores.
- Consider whether hydrogen water, inhalation, or a combined route should be tested depending on the biological target.
That is how I think we get closer to answering the real question—not whether one bottle of water in one trial “worked,” but how molecular hydrogen should be administered if we want to test its therapeutic potential properly.
Read Part 1 for the Full Parkinson’s Evidence
Part 1 covers the broader molecular-hydrogen evidence, including mechanisms, animal models, human hydrogen-water studies, hydrogen inhalation, dyskinesia research, and my current H2HUBB position.
Hydrogen Water and Parkinson’s Disease Trial: Common Questions
Did the 72-week trial show that hydrogen water improved Parkinson’s disease?
No. The study did not find a statistically significant difference between the hydrogen-water and placebo groups on total UPDRS or the reported secondary clinical measures.
Did the placebo water contain hydrogen?
Yes. The authors reported approximately 0.16 ± 0.16 ppm H₂ in the placebo water and considered that level sufficiently low to function as placebo.
Could that low amount of H₂ have mattered?
We cannot say that it affected the human outcome. However, a separate MPTP Parkinson’s mouse study reported neuroprotective effects at 0.08 ppm H₂, so I do not think low H₂ concentrations should automatically be assumed biologically inert across all models.
Does H2HUBB’s Trustlex testing prove the trial’s hydrogen measurement was wrong?
No. H2HUBB tested an ENH-2000 under different water conditions, while the trial used an ENH-1000. Our testing showed that ORP-based meter readings could vary substantially with water chemistry, which supports asking for stronger measurement validation, but it cannot retrospectively prove the trial’s reported value was incorrect.
What is H2HUBB’s position on hydrogen water for Parkinson’s disease?
I think molecular hydrogen has shown therapeutic potential for Parkinson’s disease based on the preclinical evidence and earlier clinical evidence. Further human research is needed to determine the extent of that potential and the dose, route, duration, disease stage, and biological endpoints most likely to demonstrate it.
Sources & Further Reading
- Yoritaka A, et al. Randomized, double-blind, multicenter trial of hydrogen water for Parkinson’s disease.
- Yoritaka A, et al. Pilot study of H₂ therapy in Parkinson’s disease.
- Fujita K, et al. Hydrogen in drinking water reduces dopaminergic neuronal loss in the MPTP mouse model.
- Hirayama M, et al. Inhalation of hydrogen gas elevates urinary 8-OHdG in Parkinson’s disease.
- Yoritaka A, et al. Randomized double-blind placebo-controlled trial of hydrogen inhalation for Parkinson’s disease.
- Hydrogen-saturated saline and autophagy in a rotenone Parkinson’s model.
- Nascimento GC, et al. Effects of hydrogen gas inhalation on L-DOPA-induced dyskinesia.
- Ichikawa Y, et al. Mechanism and case report of molecular hydrogen inhalation for Parkinson’s disease.
- Wang F, et al. Role and mechanism of molecular hydrogen in the treatment of Parkinson’s diseases.
- H2HUBB Part 1: Molecular Hydrogen and Parkinson’s Disease.