Purpose: This study used prospective swallowing data to establish the following: (1) whether doses to the pharyngeal constrictor muscles (PCMs) were significantly associated with swallowing outcomes; and (2) a mean dose constraint to aim for in intensity modulated radiation therapy planning. Methods and Materials: The PCMs were contoured and radiation dose data obtained for 55 patients with head and neck cancer. Associations between radiation dose and percentage of pharyngeal residue, penetration-aspiration and activity limitation measured at 6 months posttreatment were analyzed. Pretreatment swallowing function, tumor site, T classification, and chemotherapy were accounted for in multivariate analyses. Results: On multivariate analysis, the percentage of pharyngeal residue was statistically significantly associated with the mean dose to the superior PCM (95% confidence interval [CI], 0.15-1.66; P = .02). Penetration-aspiration was associated with the mean dose to the superior, middle, and inferior PCMs (95% CI, 1.02-1.27; P = .003; 95% CI, 1.02-1.23; P = .003; 95% CI, 1.04-1.21; P = .003, respectively) and the mean dose to the total PCM (95% CI, 1.05-1.31; P = .001). Activity limitation was also associated with the mean dose to the superior, middle, and inferior PCMs (95% CI, 1.01-1.20; P = .02; 95% CI, 1.00-1.15; P = .04; 95% CI, 1.01-1.15; P = .02, respectively) and the mean dose to the total PCM (95% CI, 1.02-1.23; P = .01). On univariate analysis, all 3 swallowing measures were statistically significantly worse for patients who received a biologically equivalent mean dose of >. 60 Gy to the PCMs. This remained significant on multivariate analysis for both penetration-aspiration and activity limitation (95% CI, 2.05-58.2, P = .004 and 95% CI, 1.14-27.7, P = .03, respectively). Conclusions: The radiation dose to the PCMs is significantly associated with swallowing dysfunction. Limiting the mean PCM dose to less than 60 Gy results in better swallowing outcomes.