Global and regional effects of potentially modifiable risk factors associated with acute stroke in 32 countries (INTERSTROKE): a case-control study.
Level 4 - case-series / case-control
International multicenter case-control study
PubMed 27431356 · doi:10.1016/S0140-6736(16)30506-2
What was done
An international standardized case-control study was conducted across 32 countries in Asia, America, Europe, Australia, the Middle East, and Africa between 2007 and 2015. Cases were patients with acute first stroke (13,447 total: 10,388 ischemic stroke and 3,059 intracerebral hemorrhage) enrolled within 5 days of symptom onset and 72 hours of admission. Controls were 13,472 hospital- or community-based individuals without prior stroke, matched 1:1 by age and sex. Participants completed clinical assessments, provided blood and urine samples, and odds ratios (OR) alongside population attributable risks (PAR) with 99% confidence intervals were calculated for 10 potentially modifiable risk factors.
What was found
The ten risk factors collectively accounted for 90.7% of the PAR for all stroke worldwide (91.5% for ischemic stroke, 87.1% for intracerebral hemorrhage), ranging regionally from 82.7% in Africa to 97.4% in Southeast Asia. Individual risk factor associations were: - Hypertension (history or blood pressure >=140/90 mm Hg): OR 2.98 (99% CI 2.72-3.28); PAR 47.9% (99% CI 45.1-50.6) - Regular physical activity: OR 0.60 (99% CI 0.52-0.70); PAR 35.8% (99% CI 27.7-44.7) - ApoB/ApoA1 ratio (highest vs lowest tertile): OR 1.84 (99% CI 1.65-2.06); PAR 26.8% (99% CI 22.2-31.9) - Diet quality (highest vs lowest tertile of mAHEI): OR 0.60 (99% CI 0.53-0.67); PAR 23.2% (99% CI 18.2-28.9) - Waist-to-hip ratio (highest vs lowest tertile): OR 1.44 (99% CI 1.27-1.64); PAR 18.6% (99% CI 13.3-25.3) - Psychosocial factors: OR 2.20 (99% CI 1.78-2.72); PAR 17.4% (99% CI 13.1-22.6) - Current smoking: OR 1.67 (99% CI 1.49-1.87); PAR 12.4% (99% CI 10.2-14.9) - Cardiac causes: OR 3.17 (99% CI 2.68-3.75); PAR 9.1% (99% CI 8.0-10.2) - High or heavy episodic alcohol intake: OR 2.09 (99% CI 1.64-2.67); PAR 5.8% (99% CI 3.4-9.7) - Diabetes mellitus: OR 1.16 (99% CI 1.05-1.30); PAR 3.9% (99% CI 1.9-7.6) Hypertension was more strongly associated with intracerebral hemorrhage, whereas smoking, diabetes, apolipoproteins, and cardiac causes were more strongly associated with ischemic stroke (p < 0.0001).
Why it matters
Ten modifiable risk factors account for approximately 90% of stroke burden across diverse populations, providing a strong rationale for both global and regionally tailored prevention strategies.
Limits
The case-control design cannot establish causality and is susceptible to recall bias and survivor bias (fatal out-of-hospital strokes were not captured). Controls came from a mix of hospital and community settings, introducing potential selection bias.
Cited by
- supports Metabolic disease and high blood pressure together cause the majority of strokes.
- context Strokes and heart attacks are vascular events primarily caused by vascular inflammation stemming from unrecognized metabolic disease.