did each year of adhd medication use go with a 4% rise in cardiovascular risk in a swedish study
“each year of ADHD medication use bumped up cardiovascular risk by 4%.”
a host, a psychiatrist said on The Carlat Psychiatry Podcast, 2026-10-26, at 01:49. hear it in context → the speaker cited: a Swedish study.
the episode: Balancing Focus and Cardiovascular Risk with ADHD Treatment (at 01:49). we link the show's own page; we do not re-host audio.
first, the part that holds
the number is right. a swedish nested case-control study in jama psychiatry reported that each additional year of cumulative adhd medication use was associated with a 4% higher risk of cardiovascular disease (adjusted odds ratio 1.04, 95% ci 1.03-1.05).
what we found when we went and looked
we read the paper in jama psychiatry (published online 2023-11-22; pmc10851097). it used swedish national registers covering 278,027 people aged 6 to 64 with a new adhd diagnosis or a first adhd medication dispensation between 2007 and 2020, and compared 10,388 people who developed cardiovascular disease with 51,672 matched controls. its abstract says each 1-year increase in adhd medication use was associated with a 4% increased risk of cardiovascular disease (aor 1.04, 95% ci 1.03-1.05), so the speaker's figure, study design and country all match. two details the quote leaves out: the per-year rise was 8% (aor 1.08) over the first 3 years and then levelled off, and the extra risk showed up mainly as hypertension and arterial disease. the authors say themselves that the design cannot prove causation.
assertion by assertion
| assertion | verdict | what the record says |
|---|---|---|
| the study was a swedish nested case-control study | confirmed | nested case-control design using swedish nationwide registers: 278,027 people with adhd aged 6-64, enrolled 2007-2020, with 10,388 cardiovascular disease cases and 51,672 matched controls, median follow-up 4.1 years. [1] |
| each year of adhd medication use was associated with a 4% increase in cardiovascular disease risk | confirmed | "each 1-year increase of ADHD medication use was associated with a 4% increased risk of CVD (AOR, 1.04 [95% CI, 1.03-1.05])". this is an adjusted odds ratio from a case-control design, which here stands in for relative risk. it is a relative figure, not an absolute one. [1] |
| the 4% per-year rise holds steadily across all years of use | overstated | the rise was not even across time. over the first 3 years the per-year figure was 8% (aor 1.08, 95% ci 1.04-1.11). by duration band: 0-1 year aor 0.99, 1-2 years 1.09, 2-3 years 1.15, 3-5 years 1.27, more than 5 years 1.23. so the risk plateaued after about 3 years instead of climbing 4% every year. the speaker did not say the rise was steady; this row checks a common way the figure gets heard. [1] |
| the extra risk applies to cardiovascular disease across the board | misattributed | the increase was concentrated in hypertension (aor 1.72 at 3-5 years, 1.80 at more than 5 years) and arterial disease (1.65 at 3-5 years). the study found no significant association for arrhythmias, heart failure, ischemic heart disease, thromboembolic disease or cerebrovascular disease. the composite 4% figure is real, but most of it comes from these two outcomes. [1] |
| medication use 'bumped up' the risk, read as the medication causing it | misattributed | the study reports an association. the authors write that "we could not prove causality" and name time-varying confounding, residual confounding (including adhd severity and lifestyle), exposure misclassification from non-adherence, and under-ascertainment of cardiovascular diagnoses as limits. the claim as published says 'associated with', so this row covers the causal reading the spoken wording invites. [1] |
both directions
for the speaker: the study's abstract and results state the speaker's number word for word: 4% per year, aor 1.04 (95% ci 1.03-1.05). the design (nested case-control), the country (sweden) and the outcome (cardiovascular disease) all match what the speaker said. the paper also found a dose-response pattern: risk rose with longer cumulative use and with higher average doses, reaching significance at 1.5 or more defined daily doses. that kind of gradient makes the association harder to dismiss as noise. the risk was higher for stimulants than for nonstimulants: methylphenidate aor 1.19 at more than 5 years, lisdexamfetamine elevated, atomoxetine significant only in the first year. on the early years the speaker's figure is conservative, since the first-3-years rate was 8% per year.
against: the 4% is an average across all durations and does not describe a steady climb. the per-year rise was 8% early on, and risk levelled off after about 3 years (aor 1.27 at 3-5 years, 1.23 at more than 5 years). the figure is relative, and on its own it says nothing about how many people are affected in absolute terms. the composite mostly reflects hypertension and arterial disease: there was no significant signal for arrhythmia, heart failure, ischemic heart disease, stroke or thromboembolism. 'bumped up' sounds causal, but the authors say the observational design cannot prove causality and could be affected by confounding, including adhd severity and lifestyle.
what this cannot say
this is one observational registry study, and an odds ratio from a case-control design cannot establish cause. it also cannot give a person's absolute risk. people with preexisting cardiovascular disease were excluded, and both the cardiovascular diagnoses and actual medication-taking (dispensing is not the same as adherence) may be misclassified. the cohort is swedish and aged 6-64, so results may not carry over to other populations. our own limits: we read the abstract and the pmc full text through an automated reader rather than the supplementary tables. we did not listen to the episode beyond the quoted line, so we cannot say whether the speaker added caveats elsewhere. we did not review later studies that may agree or disagree.
the medication pages behind this check
sources, each one fetched and read
- Zhang L et al. Attention-Deficit/Hyperactivity Disorder Medications and Long-Term Risk of Cardiovascular Diseases. JAMA Psychiatry, published online 2023-11-22; 2024;81(2). doi:10.1001/jamapsychiatry.2023.4294. PMC10851097 What we read there: 10,388 cases and 51,672 matched controls; each 1-year increase of adhd medication use associated with a 4% increased risk of cvd (aor 1.04, 95% ci 1.03-1.05), 8% in the first 3 years (aor 1.08, 1.04-1.11); duration bands from aor 0.99 (0-1 y) to 1.27 (3-5 y) and 1.23 (>5 y); signal in hypertension and arterial disease, none for arrhythmias, heart failure, ischemic heart disease, thromboembolic or cerebrovascular disease; "we could not prove causality." Fetched 2026-10-26 (HTTP 200). https://pmc.ncbi.nlm.nih.gov/articles/PMC10851097/
speakers are described by role, not named. one sentence is quoted under fair use, with attribution and a link to the original at the timestamp; we do not re-host audio or video. nothing on this page is a reason to start, stop or change a medication. if you have the file behind anything marked unverifiable, the research board is where people can answer. adam