On this pageJump to a section7
What the Study Looked At
Migraine and smoking are each known to raise the risk of ischemic stroke. A new Danish study in Neurology asked what happens when a woman has both, and whether the combined risk is larger than you would expect from adding the two together.
The researchers used nationwide Danish health registries to build a cohort study of every woman aged 15 to 60 who gave birth in Denmark between 1996 and 2022. That came to 1,677,707 women. Each was placed in one of four groups: migraine only, smoking only, both, or neither. They were then followed until a stroke, death, emigration, their 61st birthday, or the end of 2022.
The focus was on premature stroke, meaning a stroke at age 60 or younger. The outcomes were ischemic stroke (caused by a blocked artery), hemorrhagic stroke (caused by bleeding), and death from any cause. The median age at the start was 29 to 32 years across the groups, so this is a study about young and middle-aged women.
Stroke Risk by Group Over 20 Years
The clearest way to read this study is through absolute risk: the share of women in each group who had a premature stroke or died within 20 years.
| Group | Number of women | Ischemic stroke | Hemorrhagic stroke | Death from any cause |
|---|---|---|---|---|
| Neither migraine nor smoking | 1,389,361 | 0.8% | 0.3% | 1.0% |
| Migraine only | 65,422 | 1.3% | 0.5% | 1.2% |
| Smoking only | 214,277 | 1.5% | 0.6% | 2.3% |
| Migraine and smoking | 8,647 | 2.8% | 0.9% | 3.4% |
All figures are 20-year absolute risks of the event at age 60 or younger.
Most women in every group did not have a stroke. Still, the ischemic stroke risk in women with both migraine and smoking was about 3.5 times the risk in women with neither: 2.8 per 100 compared with 0.8 per 100. Put another way, about 2 extra ischemic strokes per 100 women over 20 years.
Relative Risk After Adjustment
The researchers also calculated adjusted hazard ratios, which compare the rate of each outcome in a group with the rate in women who had neither migraine nor smoking. A hazard ratio of 2.0 means roughly twice the rate. The ranges in parentheses are 95% confidence intervals.
| Group (vs neither) | Ischemic stroke | Hemorrhagic stroke | Death from any cause |
|---|---|---|---|
| Migraine only | 1.85 (1.66 to 2.05) | 1.46 (1.23 to 1.74) | 1.08 (0.96 to 1.22) |
| Smoking only | 2.10 (1.99 to 2.20) | 2.02 (1.87 to 2.18) | 2.44 (2.34 to 2.54) |
| Migraine and smoking | 3.61 (3.03 to 4.31) | 2.60 (1.91 to 3.54) | 3.30 (2.80 to 3.88) |
Migraine on its own was linked to higher rates of both types of premature stroke, but not to a clear rise in overall deaths, since that confidence interval crosses 1. Smoking on its own was linked to more strokes and more than double the death rate. Women with both had the highest rates on every measure.
When Two Risks Add Up to More
The main question was whether migraine and smoking together carry more risk than the sum of their separate effects. Statisticians call this an additive interaction, and they measure it with the relative excess risk due to interaction, or RERI. A RERI above 0 means the combination is worse than adding the two risks.
If the extra risks from migraine (1.85) and smoking (2.10) were purely additive, women with both would have an expected hazard ratio of about 2.95. The observed figure was 3.61. The gap of roughly 0.67 is the RERI for ischemic stroke.
- Ischemic stroke: RERI 0.67 (95% CI 0.22 to 1.13), so a measurable excess risk from the combination.
- Death from any cause: RERI 0.78 (95% CI 0.34 to 1.22), also a measurable excess.
- Hemorrhagic stroke: RERI 0.11 (95% CI -0.59 to 0.82), which does not show a clear excess.
So the extra risk from combining migraine and smoking showed up for strokes caused by blocked arteries and for deaths, but not for bleeding strokes.
What This Study Cannot Tell Us
This is an observational registry study, so it shows an association, not proof of cause. The authors adjusted for other factors, but some confounding from things the registries do not capture well, such as diet, exercise, alcohol, or how much a person smoked, is always possible.
- Only women who gave birth were included. The results may not apply in the same way to women who never gave birth or to men.
- Registry labels are blunt. Migraine and smoking came from health records, so some women with migraine or who smoked were likely counted in the wrong group, which usually blurs differences rather than creating them.
- No detail on aura or hormones. The main results are not broken down by migraine with aura, attack frequency, or estrogen-containing birth control, all of which matter for stroke risk in migraine.
- The combined group was small. With 8,647 women, the estimates for migraine plus smoking are less precise than for the other groups, which shows in the wider confidence intervals.
The study was also not designed to test whether quitting smoking lowers stroke risk for women with migraine.
What This Means for People With Migraine Who Smoke
The findings line up with other large studies of migraine and vascular risk in women. A large Korean study linked migraine to more heart attacks and ischemic strokes before and after menopause, and a 130,000-woman cohort tied a migraine history to modestly higher stroke risk after menopause. The Danish data add that, in younger women, smoking was linked to the largest extra risk when combined with migraine.
Migraine itself cannot be switched off, but smoking is a risk factor that can change. The authors concluded that their findings support targeted smoking cessation for women with migraine. That fits with long-standing advice in our guide to stroke prevention in people with migraine, which puts quitting smoking near the top of the list.
Hormonal birth control is the other piece many women ask about. This study did not test it, but US contraception guidelines already advise against estrogen-containing methods for people who have migraine with aura, because of stroke risk. If you have aura, our explainer on aura characteristics can help you describe it accurately to your clinician.
A practical next step is to bring your smoking status, aura history, and birth control method to your next visit. The migraine appointment summary sheet has space for these, and our overview of the migraine and cardiovascular connection covers blood pressure, cholesterol, and other risks to review at the same time.
- In 1.68 million Danish women who gave birth, migraine and smoking were each linked to more strokes at age 60 or younger.
- Women with both had a 20-year ischemic stroke risk of 2.8%, compared with 0.8% for women with neither.
- The combination carried more ischemic stroke and death risk than migraine and smoking added separately, but not more bleeding-stroke risk.
- Migraine alone was not clearly tied to higher overall death rates; smoking was.
- The study is observational, and its main results do not separate out aura, attack frequency, or birth control.
This article is for educational purposes only and does not replace medical advice, diagnosis, or treatment. Talk with your physician before changing any medication, including birth control, and for help with quitting smoking. If you have sudden weakness, numbness, trouble speaking, vision loss, or the worst headache of your life, call emergency services right away.
References
- Pedersen AL, Fuglsang CH, Al-Mashhadi SK, et al. The combined effect of migraine and smoking on premature stroke and all-cause mortality in parous women: a population-based cohort study. Neurology. 2026;107(12):e218752. doi:10.1212/WNL.0000000000218752
- Nguyen AT, Curtis KM, Tepper NK, et al. U.S. Medical Eligibility Criteria for Contraceptive Use, 2024. MMWR Recommendations and Reports. 2024;73(4):1-126. doi:10.15585/mmwr.rr7304a1


This article is for education and is not medical advice. Talk with your own clinician before starting, stopping or changing any treatment.
