Melatonin and heart failure What the 90% study actually found

1. What the study found

Researchers at SUNY Downstate in Brooklyn searched a large database of anonymised medical records. They pulled out 130,828 adults with chronic insomnia. Average age was 55.7. Around 61% were women.

Half of them had melatonin recorded for at least a year. The other half did not. The two groups were matched on age, sex and other health conditions, then followed for five years. These were the headline numbers:

• New heart failure: 4.6% of melatonin users against 2.7% of the others, so about 90% higher
• Hospital admission for heart failure: 19.0% against 6.6%, so roughly three and a half times as many
• Death from any cause: 7.8% against 4.3%, close to double

Those are real differences in a large group of people. They are worth taking seriously. They are also not the same thing as melatonin causing harm, and the researchers said so themselves.

2. Why 90% is not what it sounds like

Three things sit between this result and a conclusion.

It has not been peer reviewed.This is a conference abstract, which is a short summary of around 300 words. The American Heart Association is blunt about what that means. Its own release says findings from these abstracts “are considered preliminary until published as full manuscripts in a peer-reviewed scientific journal”.

The two groups were never really alike. This is the big one. Ask why one group was given melatonin and the other was not. The likeliest answer is that their sleep was worse. Bad sleep is itself tied to heart disease. So the drug and the risk may both be downstream of the same thing.

“Perhaps those prescribed the drug suffered more severe or disruptive sleep problems, symptoms that might reflect underlying health issues, including heart problems.”

— Heba Ghazal, Senior Lecturer in Pharmacy, Kingston University, November 2025

The comparison group may not be melatonin-free. Much of the data comes from the United States, where melatonin sits on a supermarket shelf. Shop-bought pills rarely reach a medical record. Some people counted as non-users were probably taking it.

The lead author, Dr Ekenedilichukwu Nnadi, framed it as a signal rather than a verdict. His line was that melatonin “may not be as harmless as commonly assumed”, and that the finding would need confirming before it changed how doctors advise anyone.

3. Melatonin is a different question in the UK

Most of the coverage you have read is American, and it does not transfer cleanly. In the UK you cannot buy melatonin over the counter. The NHS is unambiguous:

“Melatonin is available on prescription only.”

— NHS, About melatonin

The NHS also says it is mainly used for short-term sleep problems in people aged 55 and over, and that a course usually runs up to 13 weeks. Some people need longer. That matters here, because the study looked at people taking it for more than a year.

So if you are in the UK and taking melatonin, someone prescribed it. That person is the right person to ask about this news. Take the question to them.

Do not stop a prescribed medicine because of a headline. That is not us being cautious for the sake of it. An unreviewed abstract is thin ground for changing anything, and rebound insomnia is its own problem.

4. The heart risk you can actually measure

Here is the useful move. This study cannot tell you where your heart stands. A blood test can tell you a lot more, and it works whether or not the melatonin story holds up.

Four markers carry most of the signal:

• ApoB counts the harmful particles in your blood, rather than the fat inside them
• Lp(a) is a lipoprotein set almost entirely by your genes, usually worth measuring once
• hs-CRP is a sensitive marker of low-grade inflammation, the kind you cannot feel
• HbA1c and the full lipid panel give your average blood sugar over three months, plus the cholesterol breakdown that makes the rest readable

Our Advanced Heart Health panel (£159) measures all of them from a home finger-prick sample, analysed by UKAS-accredited UK laboratories. Blood markers describe risk. They cannot see plaque inside an artery, so a clean result is a reassuring picture rather than an all-clear.

There is a second reason to test. Poor sleep and raised inflammation travel together, and the same is true of blood sugar. If your sleep has been bad for years, those markers are worth seeing. Our guide to sleep and blood tests covers the wider set, and the heart health hub explains how the numbers fit together.

Frequently asked questions