The most prevalent cardiac dysrhythmia is atrial fibrillation (AF), which is a significant cause of morbidity and mortality. It is estimated that about 2.3 million people currently have AF in the United States.¹ AF often occurs silently, however, and many individuals with chronic or recurrent/paroxysmal AF are never diagnosed. Therefore, our ability to estimate the true number of individuals affected with AF has been significantly limited.

Recently, we were able to take advantage of extensive data from the long-standing Framingham Heart Study cohorts to estimate the remaining lifetime risk of AF in adults at different ages. The Framingham Heart Study afforded us a unique opportunity to study AF because of the careful surveillance of participants with repeated examinations (including recording of 12-lead electrocardiograms [ECGs]) approximately every 2 to 4 years, as well as a comprehensive review of all inpatient and outpatient medical records and ECGs of participants. Lifetime risk methods are a particularly useful means for estimating the burden of disease in the population, and they allow for comparisons of absolute risks between different diseases, as discussed later. In addition, knowledge of lifetime risks may help to motivate changes in health behaviors, as has been observed with mammographic screening for breast cancer.

Methods
The Framingham Heart Study was established in 1948, when 5,209 male and female residents of Framingham, Massachusetts, aged 28 to 62 years, were enrolled in a prospective epidemiologic cohort study. Participants were evaluated every 2 years, with medical histories, physical examinations, ECGs, and selected laboratory tests.² In 1971, 5,124 offspring of the original participants and offspring spouses were enrolled in the Framingham Offspring Study, with follow-up evaluations every 4 years.³ In the present study, we included all Framingham participants who were free of AF before their earliest examination between 1968 and 1999, provided they were examined at least once between the ages of 40 and 94 years and they had been followed up after their earliest eligible examination.

For participants who attended their routine clinic examinations, physical examinations, medical histories, ECGs, and review of interim medical records were used to determine the occurrence of AF and cardiovascular disease (CVD) events. For those who did not attend their regular examinations, outside physician records and hospital charts were evaluated. All ECG tracings were reviewed by one of two cardiologists affiliated with the Fra­m­in­g­ham Heart Study, and AF was di­agnosed only if paroxysmal or persistent AF or atrial flutter was noted. Standardized criteria for diagnosing CVD events4 were applied by a panel of three trained hospital physicians who reviewed all suspected CVD events.

Lifetime risk estimation is a modified form of Kaplan-Meier survival analysis. In traditional Kaplan-Meier methods, when someone dies free of the disease of interest, their risk for that disease does not become zero. In effect, their risk is spread among the remaining subjects who are still alive and at risk for the disease. The as­sumption is that everyone would eventually get the disease of interest. This is useful when examining the etiology of a disease or the effects of treatment, but it does not accurately reflect the “real-world” situation, in which a person who dies can have no further risk of the disease of interest. In lifetime risk analyses, the occurrence of death free of the disease of interest is considered not as a censoring event, but as a competing event alongside the occurrence of disease. Thus, the differences between cumulative incidence of a disease estimated by standard Kaplan-Meier methods and by lifetime risk methods may be small if the risk of competing causes of death is low, but the difference increases substantially when competing risk of death is high (as typically happens at older ages).

Another way to think of the comparison is that in a standard Kaplan-Meier analysis, everyone in the group under observation is as­sumed to live to the end of the observation period, whereas in a lifetime risk analysis, the last individual is assumed to live to the end of the observation period.

We applied lifetime risk methods5,6 to estimate the remaining lifetime risk of AF for men and women who were free of AF at ages 40, 50, 60, 70, and 80 years. Participants were followed until the occurrence of AF, death from another cause free of AF, attainment of age 95, or the last Framingham or medical contact at which they were known to be free of AF. Because congestive heart failure (CHF) and an­tecedent myocardial infarction (MI) are known to be risk factors for AF, we also performed analyses to estimate the lifetime risks for AF in the absence of these conditions. For these analyses, we included only those who were free of CHF and MI at their initial examination, and only counted AF events that occurred in the absence of antecedent or concurrent CHF or MI.

Results
During the study period from 1968 through 1999, we followed 3,999 men and 4,726 women, for a total of 176,166 person-years of observation. As expected, the prevalence of pre-disposing conditions increased substantially according to the age of the participants. By age 70 and older, two thirds had hypertension, 5% had a prior history of MI, and 2.5% had been diagnosed as having CHF. Dur­ing follow-up, a total of 936 participants developed AF, and 2,621 died without prior AF.

The remaining lifetime risks of AF are shown in Figure 1, according to age and sex. At age 40, the lifetime risk of developing AF was about 25%, or 1 in 4 for both men and women. In other words, looking forward from age 40, 1 in 4 men and women could expect to develop AF before they die. With increasing index age, the lifetime risk of AF remained about the same, as a result of two competing issues. Older individuals have a shorter life span (ie, higher competing risk of death) during which they may develop AF, but the risk of AF also increases dramatically with advancing age, keeping pace with the increasing risk of death from other causes. The effect of these competing forces essentially balances out. Figure 2 shows a graphic representation of this phenomenon. Note that with increasing index age (40 through 80 years), the slope of the adjusted cumulative risk becomes substantially steeper, even in the face of a shorter remaining life span. The overall remaining lifetime risk of AF is represented by the point at which the cumulative risk curve intersects the right side of the graph, at age 95.

When we examined the lifetime risk of AF occurring in the absence of antecedent or concurrent CHF or MI, we found that lifetime risks were somewhat lower (Figure 1). At all ages, the lifetime risk of AF in the absence of CHF or MI was still high, at approximately 16%. When we ex­amined the lifetime risks of AF in the context of blood pressure levels measured at Framingham Heart Study clinic examinations, increasing blood pressure was a better predictor of AF in older than in younger individuals.

Discussion
Using intensive epidemiologic ob­ser­vation methods, including review of all available medical records, we observed that the lifetime risk of AF for adults aged 40 years and over was 1 in 4. Importantly, the remaining lifetime risks of AF were similar in men and women at all ages. The remaining lifetime risk of AF did not diminish with advancing age, despite the shorter remaining life span in which to develop AF. Instead, the increasing risk of AF with advancing age kept pace with the increasing competing risk of death from other causes. Even in the absence of an­tecedent or concurrent CHF or MI, the lifetime risk of AF was high, at 1 in 6. Thus, other predisposing conditions clearly also play an important role.

AF is a substantial cause of morbidity and mortality, increasing the risk of stroke, other embolic complications, CHF, and death.^7-9 Ap­prox­i­mately 15% of all strokes are caused by AF, and a person with AF has five times the risk of developing a stroke.⁸ A number of epidemiologic studies have defined risk factors for development of AF. Chief among the risk factors is advancing age, with steady increases in incidence and prevalence, beginning at about 65 years of age and continuing to the end of life. Structural heart disease also increases the risk of AF, whether it is a result of prior MI, CHF, valvular heart disease, or other causes. Cardiac surgery, hypertension, and thyroid disease are also important risk factors for AF. The public health problem of AF will most likely increase in the future because of the aging of the population and its associated in­creased prevalence of age-related risk factors, such as hypertension, im­proved survival after occurrence of MI, development of CHF, and increasing use of cardiac surgery procedures.

The recent Rate Control Versus Electrical Cardioversion for Persistent Atrial Fibrillation10 and Atrial Fibril­lation Follow-Up Investigation of Rhythm Management11 studies indicate that attempts to maintain sinus rhythm do not improve survival compared with anticoagulation treatment and rate control. With both these ap­proaches, however, event rates are higher than optimal, given the unpredictability of treatment with warfarin (Coumadin) and the high rates of recurrent AF. Prevention of AF warrants additional research because of the suboptimal results of treatment and the high lifetime risk.

Our results highlight the significant public health problem resulting from AF and put it in a context that may be more readily understood by clinicians and patients. The 1-in-4 lifetime risk of developing AF is extremely high compared with noncardiovascular diseases. The lifetime risk of hip fracture, which is a significant cause of morbidity in elderly individuals, is 1 in 6 for white women and 1 in 20 for white men.¹² Women have a re­maining lifetime risk of breast cancer of 1 in 8 at 40 years of age.¹³ The lifetime risk of breast cancer is 1 in 14 at 70 years of age,¹³ whereas the lifetime risk of AF remains about 1 in 4.

We hope that our data on AF will help to raise awareness of this important disease and its consequences. The well-known information about the lifetime risk of breast cancer¹³ has been used to alter the public’s behavior. The increased rate of breast cancer screening in the 1990s¹⁴ and the increase in research funding for treatment and prevention appears to have been, in part, the result of publicizing the high lifetime risk of breast cancer. The significant lifetime risk of AF highlights the importance of additional studies on causes, prevention, and treatment of AF.

Presently, the American Heart As­sociation recommends that patients have their pulse checked to screen for AF at all physician visits, regardless of the reason, and at least every 2 years. Our data certainly support that recommendation, particularly for older individuals. Furthermore, patients can be taught to check their own pulse for irregularity, which may lead to earlier diagnosis and treatment, thus potentially avoiding some of the devastating consequences of AF.

Conclusion
Men and women aged 40 years and over have a 1-in-4 lifetime risk for developing AF. With the aging of the population and increased prevalence of predisposing conditions, the public health burden of AF is almost certain to increase. We must expand efforts to prevent, screen for, and effectively treat this important disease.

References
1. American Heart Association. Heart Disease and Stroke Statistics—2004 Update. Dallas, Tex: American Heart Association; 2003.

2. Dawber TR, Kannel WB, Lyell LP. An approach to longitudinal studies in a community: the Framingham study. Ann NY Acad Sci. 1963;107: 539-556.

3. Kannel WB, Feinleib M, McNamara PM, et al. An investigation of coronary heart disease in families. The Framingham offspring study. Am J Epidemiol. 1979;110(3):281-290.

4. Abbott RD, McGee DL. The Framingham Study: an epidemiological investigation of cardiovascular disease, Section 37: The Probability of Developing Certain Car­diovascular Diseases in Eight Years at Specified Values of Some Characteristics. Bethesda, Md: National Heart, Lung, and Blood Institute; 1987.

5. Gaynor JJ, Feuer EJ, Tan CC, et al. On the use of cause-specific failure and conditional failure probabilities. J Am Stat Assoc. 1993;88(421):402-409.

6. Beiser A, D’Agostino RB, Seshadri S, et al. Computing estimates of incidence, including lifetime risk. Alzheimer’s disease in the Framingham study. The practical incidence estimators (PIE) macro. Stat Med. 2000;19(11-12):1495-1522.

7. Benjamin EJ, Wolf PA, D’Agostino RB, et al. Impact of atrial fibrillation on the risk of death: the Framingham Heart Study. Circulation. 1998;98(10):946-952.

8. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke. 1991;22(8):983-988.

9. Stewart S, Hart CL, Hole DJ, et al. A population-based study of the long-term risks associated with atrial fibrillation: 20-year follow-up of the Renfrew/Paisley study. Am J Med. 2002;113(5):359-364.

10. Van Gelder IC, Hagens VE, Bosker HA, et al. A comparison of rate control and rhythm control in patients with recurrent persistent atrial fibrillation. N Engl J Med. 2002;347(23):1834-1840.

11. Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347(23):1825-1833.

12. Cummings SR, Black DM, Rubin SM. Lifetime risks of hip, Colles’, or vertebral fracture and coronary heart disease among white postmenopausal women. Arch Intern Med. 1989;149(11):2445-2448.

13. Feuer EJ, Wung L, Boring CC, et al. The lifetime risk of developing breast cancer.J Natl Cancer Inst. 1993;85(11): 892-897.

14. Blackman DK, Bennett EM, Miller DS. Trends in self-reported use of mammograms (1989-1997) and Papanicolaou tests (1991-1997)—Behavioral Risk Factor Sur­veillance System. MMWR CDC Sur­veill Summ. 1999;48(6): 1-22.

A more detailed discussion of this topic may be found in Lloyd-Jones DM, Wang TJ, Leip EP, et al. Lifetime risk for development of atrial fibrillation: the Framingham Heart Study. Cir­culation. 2004;110(9): 1042-1046.