Objectives: To compare the risk of congestive heart failure (CHF) in elderly individuals treated with any peripheral alpha‐1 antagonist for hypertension with any thiazide, test whether the risk persists in subjects without cardiovascular disease (CVD) at baseline, and examine CHF risk in normotensive men with prostatism treated with alpha antagonists.

Design: Prospective cohort study.

Setting: Four U.S. sites: Washington County, Maryland; Allegheny County, Pennsylvania; Sacramento County, California; and Forsyth County, North Carolina.

Participants: A total of 5,888 community‐dwelling subjects aged 65 and older.

Measurements: Adjudicated incident CHF.

Results: The 3,105 participants with treated hypertension were at risk for CHF; 22% of men and 8% of women took alpha antagonists during follow‐up. The age‐adjusted risk of CHF in those receiving monotherapy treated with alpha antagonists was 1.90 (95% confidence interval=1.03–3.50) compared with thiazides. In subjects without CVD at baseline receiving monotherapy, women taking an alpha antagonist had a 3.6 times greater age‐adjusted risk of CHF, whereas men had no difference in risk. Adjustment for systolic blood pressure attenuated statistical differences in risk. There were 930 men without hypertension at risk for CHF; 5% used alpha antagonists during follow‐up, with no observed increase in CHF risk.

Conclusion: Subjects receiving alpha antagonist monotherapy for hypertension had a two to three times greater risk of incident CHF, also seen in lower‐risk subjects, but differences in blood pressure control partly explained this.