Methylphenidate and the heart: what the absolute risk means

Methylphenidate is the medicine I discuss most often with families of children with ADHD (attention deficit hyperactivity disorder), many of whom are also autistic. The question about the heart comes up regularly. A large French study now gives a clearer long-term picture. The headline figure sounds large. The absolute numbers matter more, and they point to something practical.

What the study did

Researchers in Lyon used the French National Health Data System. They identified everyone who started methylphenidate between 2014 and 2023. Each new user was matched with three non-users of the same sex and the same month and year of birth. People with existing heart disease or relevant risk conditions were excluded.

That gave 212,216 new users and 636,648 non-users. The mean age was 13.7 years, and 72% were male. Follow-up was about four years.

The main outcome was cardiovascular disease. This was defined by diagnostic codes or by starting a medicine for high blood pressure.

What they found

Over about four years, there were 1,960 events in users, or 9.2 per 1,000. There were 1,898 events in non-users, or 3 per 1,000.

After adjustment for other factors, users had 2.37 times the risk of non-users.

The team then checked for residual confounding. They used negative control outcomes, which are outcomes the medicine should not affect. This corrected estimate was lower, at 1.85 times the risk. It remained statistically significant.

Relative and absolute risk

A hazard ratio of nearly two sounds large. A hazard ratio compares how often new events happened over time in users and in non-users. It is a ratio. It tells you nothing on its own about how many children are affected.

The absolute figures matter more. In this study, about 991 in every 1,000 children who started methylphenidate had no recorded cardiovascular event over four years. The raw gap between the groups was about 6 extra events per 1,000. That raw gap is before adjustment, and the corrected estimate suggests part of it is not due to the medicine.

Note also what was counted. Starting a blood pressure medicine was included as an event. So this outcome is likely to capture blood pressure problems, not only serious heart disease.

Why monitoring matters

The authors conclude that the findings support cardiovascular monitoring during treatment. I agree.

Stimulants can raise pulse and blood pressure. A rise is easy to measure and easy to act on. The point of monitoring is to catch the small group in whom it matters.

The NICE (National Institute for Health and Care Excellence) guideline on ADHD, NG87, is clear. Measure pulse and blood pressure before starting. Repeat before and after each dose change, and every six months. Compare readings with the normal range for age, using a correctly sized cuff. Readings that are high on two occasions (above the 95th centile), a sustained resting pulse over 120, or an arrhythmia, should lead to a dose reduction and referral.

In practice, the six-monthly checks are the ones that slip. Shared care, missed appointments and long waits all play a part.

Take-home

Methylphenidate was linked to a higher relative risk of cardiovascular problems, including starting blood pressure medicine.
The absolute risk was small. About 99% of children starting it had no recorded event over four years.
The finding supports monitoring. The study did not weigh the benefits of treatment, so it cannot settle the decision to treat on its own.
Check pulse and blood pressure at baseline, around each dose change and every six months. Use the right cuff and age-related centiles.
Parents can ask at every review: "When were the pulse and blood pressure last checked?"

Dr Odet Aszkenasy is a Consultant Community Paediatrician and the author of The Genetics of Autism: A Guide for Parents and Professionals.

Reference

Berthod C, Loppinet T, Ndiaye M, Boussaha I, Jacquet-Lagreze M, Kassaï-Koupai B. Methylphenidate and long-term risk of cardiovascular diseases: a nationwide cohort study. Pharmacoepidemiology and Drug Safety. 2026;35(10):e70493. https://doi.org/10.1002/pds.70493