IN THIS ISSUE OF JAMA, THE REPORT BY CONEN AND COLleagues provides further evidence to confirm an increased mortality risk among middle-aged women with new-onset atrial fibrillation (AF). The study cohort consisted of 34 722 health care professionals in the Women’s Health Study (WHS) who agreed to prospective follow-up after the end of the randomized treatment trial. These women were aged 49 to 59 years and free of cardiovascular disease at baseline. During a median follow-up of 15.4 years, 1011 women (2.9%) developed new-onset AF, and 63 deaths occurred among these women. In multivariable models, incident AF was associated with an increased adjusted risk of all-cause mortality, as well as cardiovascular and noncardiovascular death. Analyses of mortality outcomes in the Framingham Heart Study demonstrated that the development of AF was associated with attenuation of the female survival advantage. Other studies that have examined mortality have also observed an association between AF and death after adjusting for potential confounders, with the exception of studies of lone AF. The study by Conen et al adds to the existing literature that AF is not a benign condition but in fact is associated with premature death. Newly identified AF in seemingly healthy women should be taken seriously and treated aggressively, recognizing that anticoagulation, which is an integral part of AF management, reduces stroke and mortality risk. In the WHS cohort, nearly half of the women who subsequently developed AF had hypertension, a third had hypercholesterolemia, and 9% were current smokers at baseline. Compared with women who remained free of AF, those who developed AF had higher prevalence of hypertension, diabetes, hypercholesterolemia, smoking, and body mass index higher than 25 at baseline, representing a higher-risk subgroup. Therefore, while the cohort was event-free at baseline, whether these women can be considered “healthy” is questionable. Why was the mortality risk greater once AF was identified? Conen et al suggest that the increased risk was at least partly mediated through nonfatal cardiovascular events such as congestive heart failure and stroke, as has been reported by others. Conen et al also observed an association between incident AF and myocardial infarction, which was thought to also contribute to an increased mortality risk among these women. This observation is consistent with results of the Olmsted County AF study, in which a strong relationship between incident AF and subsequent coronary events was observed, and extends the evidence of this relationship to younger women. In the Olmsted County study, incident AF was associated with substantially higher risk of coronary events in women than that predicted by the Framingham risk score, with the excess mortality risk significantly greater among women than men. While standardized echocardiographic studies were not performed in all women with incident AF in the WHS cohort, structural abnormalities were common in the subgroup for whom echocardiographic data were available. Using linear left atrial dimension, a relatively crude way to measure left atrial size and with a cutoff of 40 mm for enlargement, 41% of women who developed AF had an enlarged left atrium at the time of diagnosis. Furthermore, 32% had left ventricular hypertrophy. These data suggest that women who developed AF had preexisting structural substrates for this arrhythmia. Left atrial enlargement and left ventricular hypertrophy are typical features of hypertensive heart disease. In a study of lone AF, patients with normal-sized atria were found to have a benign clinical course throughout a 30-year follow-up. In contrast, patients with increased left atrial volume at diagnosis or during follow-up experienced adverse events. Although left atrial enlargement is not the direct cause of AF, stroke, heart failure, or death, it has consistently been shown to be the common denominator for the pathophysiologic cascade that leads to AF, stroke independent of AF, heart failure, and death. While it is