The Study That Found 9 Causes of 90% of Heart Attacks — and Why Your Doctor Probably Hasn’t Mentioned Two of Them
What the findings mean for your heart, which factors are still being underestimated, and what to actually do about each one.
In 2004, a team led by cardiologist Salim Yusuf published a study that should have changed how medicine thinks about cardiovascular prevention forever. Twenty years later, its most striking findings are still largely absent from standard clinical consultations.
The study was called INTERHEART.1
It enrolled 30,000 participants across 52 countries, all six inhabited continents, every major ethnic group, and both sexes. It asked a simple question:
Which modifiable risk factors account for the global burden of first heart attacks?
The answer was startling.
9 factors. 90% of the risk. In men, women, young, old, rich, poor, East and West, the same nine factors, with the same relative weights, consistently predicting the overwhelming majority of first myocardial infarctions worldwide.
“This is important,” said Yusuf at the time, “because most people believe that only half the risk of heart attacks can be predicted. This study convincingly shows that 90% of the global risk is preventable.”
That was in 2004. The science has held up, but the clinical implementation has not.
What “Population Attributable Risk” (PAR) Actually Means
Before we go through the nine factors, one concept deserves plain explanation, because it’s the number attached to each factor, and it’s easily misread.
Population attributable risk (PAR) is the proportion of all heart attacks in the population that would theoretically disappear if a given risk factor were eliminated entirely. It combines both the strength of the association (how much does this factor increase your individual risk?) and the prevalence of the factor (how common is it in the population?).
In plain terms: PAR tells you not just whether a factor is dangerous, but how much of the total problem it accounts for. It’s the most useful metric for understanding which of your personal risk factors deserves the most urgent attention.
With that in mind, here is what INTERHEART found.
The Nine Factors: What They Are, What They Account For, and What to Do About Each
Factor 1: Abnormal Lipids (ApoB/ApoA1 Ratio) — PAR 49.2%
This is the single most important modifiable risk factor for heart attack globally, and it’s not the one being measured in most standard blood panels.
INTERHEART measured lipids not as LDL cholesterol but as the ApoB/ApoA1 ratio — the ratio of apolipoprotein B (the protein carried by all atherogenic particles: LDL, VLDL, and IDL) to apolipoprotein A1 (the main protein of HDL, the protective lipoprotein).
This ratio carries a population attributable risk of 49.2% — meaning nearly half of all first heart attacks globally are attributable to abnormal lipid profiles measured this way.1
A person with normal LDL-C but an unfavorable ApoB/ApoA1 ratio — common in people with metabolic syndrome, insulin resistance, or elevated triglycerides — is carrying a risk that the standard cholesterol test does not capture. Twenty years after INTERHEART, the 2026 ACC/AHA Dyslipidemia Guidelines finally endorsed ApoB testing as a formal clinical target.
The evidence has been here since 2004.
What to do: Ask for ApoB and ApoA1 at your next blood panel — not just the standard lipid panel. Your standard LDL result may be missing the number that matters most for your cardiovascular risk. This is covered in full in the ApoB/ApoA1 post.
Factor 2: Smoking — PAR 35.7%
Smoking is responsible for over a third of all first heart attacks globally. Current smokers have an odds ratio of 2.87 for first MI compared to never-smokers.1
The mechanism operates through multiple simultaneous pathways: nicotine elevates heart rate and blood pressure; carbon monoxide reduces the blood’s oxygen-carrying capacity; cigarette smoke directly damages the arterial lining, accelerating atherosclerosis; smoking raises fibrinogen levels and platelet aggregation, increasing clot formation; and it reduces HDL while oxidizing LDL.
Even one cigarette per day increases MI risk by approximately 5%.
The relationship is not linear, the risk difference between zero and one cigarettes is disproportionately large relative to the increment from one to ten. Former smokers still carry elevated risk that takes approximately 15 years to approach (but not fully reach) never-smoker levels.
What to do:
Smoking cessation is the single highest-return cardiovascular intervention available to any current smoker. Nothing else comes close. Varenicline has the strongest evidence for cessation of any pharmacological intervention, producing roughly three times the long-term quit rate of placebo. Nicotine replacement significantly improves quit rates over unaided attempts. Combination approaches (medication plus behavioral support) produce the best outcomes. Vaping is not a cardiovascular-safe alternative, short-term studies show endothelial dysfunction, increased oxidative stress, and elevated inflammatory markers comparable to conventional cigarettes.
Factor 3: Psychosocial Factors — PAR 32.5%
This is the finding that still does not receive the clinical attention it deserves.
Chronic psychological stress, depression, and perceived loss of control over life account for 32.5% of the global risk of first heart attack — more than hypertension (17.9%), more than diabetes (9.9%), and more than abdominal obesity (20.1%).1,2
Psychosocial factors are the third most important contributor to global heart attack risk. The odds ratio was 2.67 — almost as strong as smoking.1
A 2025 review in the American Journal of Preventive Cardiology documented the mechanism: chronic stress activates the HPA axis, producing sustained cortisol elevation that promotes atherosclerotic plaque development.3
The catecholamines (adrenaline and noradrenaline) increase heart rate, peripheral vascular resistance, and platelet aggregation. Sustained sympathetic activation drives the release of pro-inflammatory mediators from the bone marrow and spleen, producing a generalized state of chronic vascular inflammation.2
A 2024 review in Heart and Mind added that chronic stress increases susceptibility to type 2 diabetes through cortisol-mediated insulin resistance, compounding the cardiovascular harm.4
Twenty years after INTERHEART, standard cardiovascular risk calculators do not include psychosocial stress. No validated screening question for psychological risk is embedded in routine primary care cardiovascular assessment.
What to do: Start by honestly assessing your psychosocial load across the four domains INTERHEART measured: work stress, home stress, financial stress, and major life events. The biology connecting chronic stress to your arteries is specific, documented, and addressable. This is covered in full in the psychosocial factors post.
Factor 4: Abdominal Obesity — PAR 20.1%
INTERHEART measured obesity not as BMI but as waist-to-hip ratio — the ratio of waist circumference to hip circumference, reflecting fat distribution rather than total quantity.1 The PAR of 20.1% is attributable to the top two tertiles of waist-to-hip ratio, not to weight alone.
Visceral fat — stored around the abdominal organs and reflected in waist circumference — is metabolically active in a way that subcutaneous fat is not. It releases inflammatory cytokines, free fatty acids, and adipokines that directly promote insulin resistance, dyslipidemia, endothelial dysfunction, and hypertension.
A person can have a normal BMI with a high waist-to-hip ratio — the “metabolically obese normal weight” phenotype — and carry substantially elevated cardiovascular risk that weight measurement alone would miss.
What to do: Start with a measuring tape, not a scale. Your waist-to-hip ratio tells you what your weight doesn’t — where the cardiovascular risk actually sits in your body. This is covered in full in the abdominal obesity post.
Factor 5: Hypertension — PAR 17.9%
High blood pressure has an odds ratio of 1.91 in INTERHEART — near-doubling of MI risk — with a PAR of 17.9%.1
The 2025 AHA statistics confirm that hypertension remains the most common and most modifiable risk factor for cardiovascular disease, responsible for an estimated 13.8% of all cardiovascular deaths globally.5
The mechanism is direct: sustained elevated arterial pressure causes endothelial stress and injury, accelerates arterial stiffening, promotes left ventricular hypertrophy, and contributes to atherosclerotic plaque development through mechanical and inflammatory pathways.
What to do:
The 2025 ACC/AHA Hypertension Guidelines target systolic blood pressure below 130 mmHg for most adults with hypertension, with active encouragement to achieve below 120 mmHg where tolerated. Home blood pressure monitoring (5 minutes of quiet sitting, both arms, twice morning and twice evening for 7 days) provides a more representative picture than clinic measurements. Lifestyle interventions with documented reduction are in full covered in the hypertension post.
Factor 6: Diabetes — PAR 9.9%
Diabetes doubled the individual risk of MI in INTERHEART (odds ratio 2.37), with a PAR of 9.9%.1
The cardiovascular harm operates through multiple mechanisms: advanced glycation end-products damage endothelial function; hyperglycemia promotes oxidative stress and inflammation; diabetic dyslipidemia compounds atherogenic risk; and the insulin resistance that precedes formal diabetes diagnosis already drives cardiovascular pathology years before the diagnostic threshold is crossed.
What to do:
The most important clinical window is prediabetes — fasting glucose between 5.6 and 6.9 mmol/L (100-125 mg/dL) or HbA1c between 5.7 and 6.4%. The Diabetes Prevention Program showed that intensive lifestyle modification reduced progression from prediabetes to type 2 diabetes by 58% — more than twice the effect of metformin alone. For people with established type 2 diabetes and cardiovascular risk, GLP-1 receptor agonists and SGLT2 inhibitors have the strongest cardiovascular outcome evidence of any glucose-lowering medications. For a complete overview of lifestyle interventions with documented benefits for preventing and reversing metabolic decline, see the diabetes post.
Factor 7: Daily Fruits and Vegetables — PAR 13.7%
The absence of daily fruit and vegetable consumption was associated with a PAR of 13.7% in INTERHEART.1 Daily consumers had 30% lower MI risk (odds ratio 0.70).
This likely reflects polyphenol and flavonoid antioxidant and anti-inflammatory effects, fiber’s impact on lipid profiles and gut microbiome composition, potassium’s role in blood pressure regulation, and displacement of higher-calorie, lower-nutrient foods.
What to do:
The Mediterranean dietary pattern — daily fruit and vegetables, olive oil, legumes, whole grains, fish — is the most extensively studied dietary pattern in cardiovascular medicine and is endorsed by the 2026 ACC/AHA guidelines. The PREDIMED trial found that a Mediterranean diet supplemented with extra-virgin olive oil reduced major cardiovascular events by 30% in high-risk adults — one of the largest dietary intervention effects on a hard cardiovascular endpoint published. Aim for at least 5 portions of fruit and vegetables daily. Make olive oil your primary fat. Eat legumes at least 3 times per week. Eat oily fish at least twice per week.
Factor 8: Regular Physical Activity — PAR 12.2%
Regular physical activity was associated with a PAR of 12.2% in INTERHEART, with an odds ratio of 0.86 — a 14% individual risk reduction.1
The protective effects are mediated through improved lipid profiles, blood pressure reduction, insulin sensitivity, anti-inflammatory myokine production (covered in the muscle post), improved cardiac function and VO2 max, and weight management.
What to do:
The minimum effective dose for cardiovascular prevention is 150 minutes of moderate aerobic activity per week (brisk walking, cycling, swimming) or 75 minutes of vigorous activity. For cardiac remodeling and VO2 max improvement, the Norwegian 4×4 Protocol (4 minutes at 85-95% max heart rate, 4 rounds, 3 times per week) produces eccentric cardiac hypertrophy not achievable with moderate continuous exercise alone. Resistance training independently reduces cardiovascular mortality by approximately 15-17% in meta-analyses through effects on glucose metabolism, blood pressure, and body composition that are partially non-overlapping with aerobic exercise. Both types matter.
Factor 9: Regular Alcohol Consumption — PAR 6.7%
In 2004, moderate regular alcohol consumption appeared modestly protective.1 The 2023-2025 evidence has meaningfully revised this picture.
More rigorous Mendelian randomization studies have substantially weakened the evidence for a cardioprotective effect of alcohol and identified that many apparent “protections” reflected healthy user bias. WHO’s 2023 position is that there is no safe level of alcohol consumption from a health perspective, and the 2025 AHA statistics no longer endorse alcohol as a protective cardiovascular factor.
The honest position in 2026: if you already consume light-to-moderate amounts of alcohol, the incremental cardiovascular harm is likely small. If you don’t drink, the INTERHEART finding is not a reason to start. The broader health evidence — cancer risk, liver risk, neurological effects, sleep disruption — does not support recommending alcohol as a cardiovascular intervention.
The Three Numbers That Stop Most People Cold
Here’s what INTERHEART tells us when we rank the factors by their contribution:
Abnormal lipids: 49.2%
Smoking: 35.7%
Psychosocial factors: 32.5%
These three factors alone account for the majority of global heart attack risk. Two are the subject of established clinical interventions. One is almost never formally assessed or addressed in a cardiology appointment.
The combined odds ratio of having the three major traditional risk factors (smoking, hypertension, diabetes) versus having none was 13.01 — a 13-fold increase in risk. Add ApoB/ApoA1, and the combined odds ratio rises to 42.3.1
These are large, consistent, globally validated findings from the largest cardiovascular case-control study ever conducted.
What INTERHEART Found About Family History
One of the most clinically important findings in INTERHEART is what happened when family history of coronary heart disease was added to the model containing the nine modifiable risk factors.
Family history independently increased MI risk with an odds ratio of 1.55 and a PAR of 12%. But when the nine modifiable factors were already accounted for, adding family history only raised the total PAR from 90.4% to 91.4%.1
Family history adds only about 1 percentage point of explanatory power once the nine modifiable factors are controlled. The overwhelming proportion of what family history predicts is already captured by the nine factors, because what families share is largely shared environments, dietary patterns, stress levels, and metabolic tendencies, not simply shared DNA with an immutable trajectory.
Having a strong family history of heart disease is primarily a signal that you need to manage the nine modifiable factors more aggressively, not that your fate is sealed.
Your Personal Priority List
INTERHEART gives you a framework for prioritizing your own cardiovascular prevention efforts based on your personal status on each factor.
Assess all nine.
ApoB/ApoA1 (blood test)
Smoking status (self-evident)
Psychosocial stress (honest self-assessment using the questions in factor 3)
Waist-to-hip ratio (measuring tape)
Blood pressure (home monitor for 7 days)
Fasting glucose and HbA1c (blood test)
Daily fruit and vegetable intake (food diary for one week)
Weekly physical activity minutes
Alcohol intake (units per week)
The sequence that makes sense for most people:
If you smoke, stopping is the single highest-return intervention available. Nothing else comes close.
If you don’t smoke, optimizing your lipid profile (specifically ApoB) and addressing abdominal obesity and blood pressure are the next highest-return interventions.
If your traditional risk factors are already managed, psychosocial factors and physical activity are the most likely underaddressed variables in your cardiovascular picture.
The conversation with your doctor:
“I’d like to go through the nine INTERHEART risk factors and make sure I have a clear action plan for each one. Specifically, I’d like ApoB and ApoA1 added to my next blood panel, and I’d like to discuss whether my psychosocial risk is being adequately addressed in my cardiovascular prevention plan.”
What INTERHEART Didn’t Include
The study was conducted in 2004. Several risk factors now carry strong independent evidence but were not part of the original design.
Sleep duration and quality: Short sleep (below 7 hours) and sleep apnea are independently associated with elevated cardiovascular risk, operating through many of the same pathways as the psychosocial factors INTERHEART did capture. The 2025 AHA Life’s Essential 8 now includes sleep as a formal cardiovascular health component.
Lipoprotein(a): Lp(a) is an independent genetic cardiovascular risk factor affecting 20% of the population and is not captured in ApoB/ApoA1 analysis. The 2026 ACC/AHA guidelines require that all adults have Lp(a) measured once in their lifetime.
Air pollution: Chronic exposure to fine particulate matter (PM2.5) is now a recognized cardiovascular risk factor with an estimated PAR comparable to several INTERHEART factors in highly polluted urban environments.
Social isolation: Its cardiovascular risk is mediated through the same psychosocial pathways INTERHEART identified. Its inclusion would likely modify the PAR estimate for psychosocial factors upward.
The INTERHEART nine are the foundation. Addressing all nine systematically and accurately puts you in a position where the remaining modifiable cardiovascular risk is genuinely small, and where the remaining fixed risk from genetics and age is the unavoidable fraction, not the dominant one.
Nine Factors, Ninety Percent, One Priority
The most important finding from INTERHEART is not any individual risk factor. It’s the cumulative message.
The overwhelming majority of heart attacks are preventable. They result from a small number of well-characterized, measurable, modifiable inputs.
For most people, the gap between their current cardiovascular risk and an optimized one is not a pharmaceutical gap. It’s a measurement gap (nobody has assessed all nine), a knowledge gap (most people don’t know their ApoB or their waist-to-hip ratio), and an action gap (even known risk factors are incompletely addressed).
The tools are available. The evidence is two decades old and continuously validated. What’s required is the systematic attention — to all nine factors, in sequence, with appropriate targets — that most cardiovascular consultations don’t currently provide.
To your zenith within,
Sara Redondo, MD, MS
References:
Yusuf S, Hawken S, Ôunpuu S, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet. 2004;364(9438):937-952. doi:10.1016/S0140-6736(04)17018-9
Rosengren A, Hawken S, Ôunpuu S, et al. Association of psychosocial risk factors with risk of acute myocardial infarction in 11,119 cases and 13,648 controls from 52 countries (the INTERHEART study): case-control study. Lancet. 2004;364(9438):953-962. doi:10.1016/S0140-6736(04)17019-0
Raggi P, Fatemi O, Quyyumi AA, Henein MY, Vaccarino V. Psychosocial stress and cardiovascular disease. Am J Prev Cardiol. 2025;22:100968. doi:10.1016/j.ajpc.2025.100968
Lee JH, Garg M, Sundaram V. Impact of chronic psychological stress on cardiovascular disease risk: a narrative review. Heart Mind. 2024;8(4):268-278. doi:10.4103/hm.HM-D-24-00040
Writing Group Members; Fryar CD, Heard DG, et al. 2025 Heart Disease and Stroke Statistics: a report of US and global data from the American Heart Association. Circulation. 2025. doi:10.1161/CIR.0000000000001303