chatGPT:
Below is a cleaned-up transcript based on the text you provided, followed by a summary and critique. I also checked the main studies the speaker refers to.
Tidy transcript
The Alarming 2025 Melatonin Study and Its Flaws
0:00
The internet is ablaze with an alarming study linking melatonin to a sharp increase in the risk of heart failure. Now a new study tells a completely different story, and this is why clickbait sucks: it just confuses everyone.
Luckily, when you go through the data, it actually all points in the same direction, and it gives us a clearer picture regarding whether to take melatonin supplements.
This is important because Americans now take melatonin at five times the rate they did in 1999, and a 2023 survey found that 64% of adults have taken it at some point. That is tens of millions of people. Plenty of short-term studies have shown benefits for sleep, but while melatonin supplements appear safe, long-term data has been lacking.
That is why the alarming 2025 study was so significant. It looked at five years of electronic health records to assess the effects of melatonin on a longer time horizon, and it found that long-term users had an 89% higher risk of heart failure, were about 3.5 times more likely to be hospitalized for it, and were over twice as likely to die from any cause.
That sort of data journalists love because it makes for wonderful clickbait titles. But if we take a careful look at the study, there were some big issues with the methodology.
1:14
The first had to do with the composition of the two groups the researchers compared. They were trying to separate long-term melatonin users from non-users, which is foundational if you want to answer the question of the safety of long-term melatonin use.
When we look at how the researchers established the two groups, we spot a key flaw. On one side were those with electronic health records indicating a prescription for melatonin. That was the melatonin group. The non-melatonin group was made up of people whose records did not mention melatonin.
Here is the problem: the TriNetX global research network database includes data from numerous countries. In some countries, like the UK, patients require a prescription for melatonin; in others, like the US, they do not. In this study, the melatonin group included only those whose records showed a prescription for melatonin.
So anyone taking melatonin over the counter would end up in the non-melatonin group. Obviously, that is a huge problem. How can we draw conclusions if the non-melatonin group may actually be consuming melatonin?
Understanding observational studies and causal connections
2:21
The second major issue is that this is an observational study, not a randomized controlled trial. That means it tells us about association, not causation.
We can find all sorts of associations in datasets, but not all of them reflect an underlying causal connection. Consider the relationship between ice cream sales and shark attacks: both peak in summer, but ice cream does not cause shark attacks. Or look at how closely autism diagnoses follow growth in organic food sales: correlation, yes; causal connection, surely not.
But even if we ignore those two issues from the 2025 study and assume the association between heart failure and melatonin use is real, why might that be?
Maybe people with more severe insomnia are more likely to get a prescription for melatonin. If that were true, it would make sense of the pattern, because we already have plenty of data showing that sleep disruption leads to worse heart-related health outcomes.
For instance, a 2024 meta-analysis concluded that there is a strong causal connection between insomnia and risk of heart disease. That would make sense, since we know lack of sleep ramps up inflammation, and inflammation is at the heart of many kinds of heart disease.
So it is not necessarily melatonin that is causing the increase in heart failure. It might instead be the insomnia itself. To really figure out what is going on, we need randomized controlled trials where one group takes melatonin and the other definitely does not, but instead takes a placebo.
Besides all this, what we are looking at is just an abstract of a paper presented at a conference. It has not yet gone through peer review. All of that is preliminary.
Even though this is a preliminary finding with major methodological issues, it got a truckload of press, even though we already have studies pointing in the opposite direction.
Contradictory studies on melatonin and heart health
4:22
For instance, a 2025 meta-analysis looked at melatonin use in people already diagnosed with heart failure and combined four randomized controlled trials. Melatonin improved patient scores on measures of heart failure and significantly boosted quality of life. It also increased an important metric of blood vessel health. There was even a trend toward improvement in the heart’s pumping ability, though this was not statistically significant.
This is consistent with an earlier analysis from 2022, which summarized the evidence and found that melatonin protected the heart in numerous human and animal studies.
Still, scary studies that grab headlines can leave people feeling confused and doubtful. So who is right here? Is melatonin dangerous for the heart or not?
New meta-analysis study on melatonin
5:07
A new study makes a decisive contribution to this discussion. It is a meta-analysis including 63 randomized controlled trials involving over 3,000 participants. This is a much more compelling kind of evidence than the study that raised the alarm last fall.
Not only are these randomized controlled trials instead of an observational study, but this is a meta-analysis combining data from many trials. It includes the best evidence available through October 2025, just weeks before that heart failure study made headlines.
The focus of the study is not heart failure specifically, so it does not speak to that directly. What it does provide is a wide-angle view of factors highly relevant to heart health, as well as chronic conditions like type 2 diabetes and high blood pressure.
Here is what they found: melatonin supplements significantly reduced waist circumference, blood pressure, fasting blood glucose, LDL cholesterol, and total cholesterol. They also found significant drops in several key markers of inflammation and oxidative stress, while melatonin boosted antioxidant capacity in the blood.
This is significant because inflammation and oxidative stress have major impacts on heart disease, obesity, and insulin resistance. The size of the effects was generally modest but meaningful. Crucially, taken together they point decisively toward an effect that supports heart health.
An important caution, though: the trials in this meta-analysis were mostly short-term, many lasting only a few weeks. The longest lasted just over a year. So this does not settle the question about the long-term effects of melatonin use. We still need long-term randomized controlled trials.
But there is no question that the evidence trends in a positive direction. The most up-to-date study we have today is encouraging. We do not see anything here that supports the alarming headlines from that heart failure study.
Key takeaways
7:05
There is another way to approach this issue, which is to look at what melatonin actually does. Does it make sense that it would promote heart health or get in the way of it?
The first thing to understand is that melatonin is much more than just a sleep hormone. It is produced by the pineal gland in the brain, but also throughout the body, including the gut and the skin. It is a potent antioxidant and anti-inflammatory agent, and it may also enhance the functioning of our blood vessels.
Since blood vessel dysfunction and inflammation are central to heart disease, it makes sense that we can see these cardioprotective signals in the meta-analysis. Sleep itself is linked to heart health, as mentioned earlier, so melatonin’s ability to improve sleep metrics likely has a multifaceted beneficial effect.
So the benefits of melatonin supplements make sense in theory, and we have some compelling evidence for their effects on sleep, along with suggestive evidence for heart health. But there are still things to watch out for when moving from studies to actual supplementation.
First is dosing. Melatonin supplements often contain very high doses, even over 5 mg per day, and we do not have long-term safety data for this. To put that into perspective, the body produces between 10 and 80 micrograms of melatonin at night, and about 15% of melatonin in a supplement is absorbed by the body.
So think about a 5 mg dose. That is 5,000 micrograms. If 15% gets absorbed, that is about 750 micrograms, which is around ten times or more than the amount our body naturally produces.
Personally, I would not take more than 1 mg of melatonin at night. In fact, I do take melatonin regularly as part of my sleep supplement, but it includes just 0.3 mg, or 300 micrograms. Given typical bioavailability, that sits in the middle of the range normally produced by the body. It is a mile away from the doses seen in some supplements.
But just because I take a supplement does not mean you should.
Second, we want to be careful about product quality. A 2023 analysis of melatonin gummy products found that the actual quantity of melatonin ranged from 74% to 347% of the amount specified on the label. One product did not contain any detectable melatonin at all. So it is a bit like playing gummy roulette.
If you are shopping for a melatonin supplement, look for a good manufacturer and ideally one that provides certificates of analysis.
Finally, timing is critical. Melatonin is not a sedative that simply knocks you out. It is a hormone that plays a central role in regulating the body’s natural sleep-wake cycle. It works as a chronobiotic agent that can shift that cycle.
Ideally, we want to be taking melatonin around 2 to 4 hours before we want to fall asleep.
When it comes to sleep supplements, melatonin gets all the attention, but there is also compelling evidence for another simple supplement that can have a positive impact on sleep. Check out the next video to find out more about that recent study.
Summary
The video argues that the widely reported 2025 study linking long-term melatonin use to heart failure was overinterpreted. The speaker says the study had two main weaknesses: it was observational rather than randomized, and it may have misclassified many over-the-counter melatonin users as “non-users,” especially in countries where melatonin is sold without prescription.
He then contrasts that study with more favorable evidence, especially a 2025 meta-analysis of randomized trials and an additional heart-failure-focused meta-analysis. His overall conclusion is that the best current evidence does not show that melatonin is clearly dangerous for the heart, and may instead point toward modest cardiometabolic benefit.
The practical takeaway is cautious optimism: melatonin may help sleep and may have beneficial effects on blood pressure, glucose, lipids, inflammation, and oxidative stress, but long-term safety data remain limited. Because of that, the speaker advises low doses, good product quality, and careful timing.
Critique
What the video gets right
The central caution about the alarming 2025 heart-failure study is fair. The American Heart Association’s own write-up describes it as a preliminary conference study and explicitly says the association cannot prove cause and effect. (American Heart Association)
The speaker is also right that randomized evidence deserves more weight than a conference abstract from an observational database study. A 2025 meta-analysis did include 63 RCTs up to October 2025 and found significant reductions in systolic blood pressure, fasting glucose, LDL cholesterol, total cholesterol, CRP, malondialdehyde, TNF-α, and IL-6, with an increase in total antioxidant capacity. (MDPI)
He is also right that the “heart failure benefit” literature exists, but it is still small. The 2025 heart-failure meta-analysis he cites was based on only 4 articles, so it is supportive but not definitive. (PMC)
The practical warning about product quality is well supported. The 2023 JAMA analysis found melatonin gummies ranging from 74% to 347% of labeled content, with one product containing no detectable melatonin. (JAMA Network)
What the video overstates or simplifies
The biggest factual slip is that the speaker says the 63-trial meta-analysis found a significant reduction in waist circumference. It did not. The paper reported a significant reduction in hip circumference, while waist circumference was not significantly changed. (MDPI)
The phrase “points decisively” is also stronger than the evidence warrants. The 63-trial meta-analysis is encouraging, but many outcomes were modest, several were null, and the trials were mostly short. The paper itself notes no significant effect on body weight, BMI, fasting insulin, HOMA-IR, HbA1c, triglycerides, diastolic blood pressure, AST, or GGT. (MDPI)
The mechanistic discussion is plausible but still partly inferential. Melatonin has antioxidant and anti-inflammatory roles, but mechanism does not settle the clinical question. Human outcomes remain the key issue, especially for chronic use.
His dosing advice is sensible but partly opinion-based. The figures he gives for endogenous nightly production and roughly 15% oral bioavailability are grounded in literature, but translating that directly into an “ideal” supplement dose is not straightforward because bioavailability is variable and depends on formulation and individual metabolism. (PMC)
Likewise, the “2 to 4 hours before sleep” point is reasonable for circadian phase-shifting, but that timing is more applicable to chronobiotic use than to every person taking melatonin for simple sleep initiation. Evidence on timing depends on the goal, dose, and formulation. (PMC)
Bottom-line assessment
This is a good corrective video, but not a perfect one.
Its strongest contribution is methodological: it properly warns viewers not to treat a preliminary observational signal as proof that melatonin causes heart failure. That is an important and valid point. (American Heart Association)
Its weakest point is that it occasionally slides from “the evidence is more reassuring than the headlines” into “the evidence is clearly positive.” That goes a bit too far. A more careful conclusion would be:
Current evidence does not justify alarmist claims that melatonin clearly harms the heart, and short-term randomized evidence is mildly reassuring on several cardiometabolic markers, but long-term cardiovascular safety still remains incompletely resolved. (American Heart Association)
If you want, I can also turn this into:
- a table with columns for claim / evidence / critique, or
- a shorter publication-style summary with novelty and limitations.