Epidemiological studies have reported the association between hypertension and exposure to road traffic noise, but the association between noise frequency characteristics is not clear. This study investigated the association between exposure to different frequency components of road traffic noise and the prevalence of hypertension in central Taiwan. We recruited 820 residents living near main roads for more than 3 years. Frequency components of traffic noise and traffic flow rates during 0900-1700 on weekdays were measured simultaneously in 2008. Multiple logistic regressions were conducted to estimate odds ratios (ORs) for diagnosed hypertension, adjusting for potential confounders and the total traffic flow rate. The high-exposure group (≥ the median of noise levels decibels, dB) at 63 Hz, 125 Hz and 1000 Hz had ORs for hypertension of 2.77 (95% confidence interval CI: 1.17-6.52), 4.08 (95% CI: 1.57-10.63) and 1.98 (1.00-3.92) (95% CI: 1.00-3.92), respectively, compared to the low-exposure group (< the median of noise levels dB). There was an increasing trend in the prevalence of hypertension by exposure to road traffic noise at 63, 125 and 1000 Hz in all subjects and in men. Total subjects exposed to ≥ 51 dB at 125 Hz had an OR of 4.65 (95% CI = 1.46-14.83) compared to those exposed to < 47 dB. With the possible bias of exposure misclassification and a bias from using diagnosed hypertension, these results suggest that exposure to road traffic noise at low and hearing-sensitive frequencies may be associated with hypertension and exposure to noise at 125 Hz may have the greatest risk for hypertension.