We appreciate Professors Widdicombe and Addington's comments regarding our study of aspiration risk in stroke patients1 but disagree with their assertion that “subjective measurement of voluntary cough has not been shown to give less acceptable results than the objective method.” McCullough et al2 found that although the specificity of subjective bedside assessments of voluntary cough ranged from 79 to 89%, sensitivity was poor, ranging from 26 to 42%. This indicates that, if used alone, subjective assessment of voluntary cough would miscategorize at least half of the patients at increased aspiration risk. Our results show that objective assessment of voluntary cough has sensitivities and specificities in excess of 90% in predicting patients at risk of aspiration, and this represents an improved accuracy over standard subjective measures. These findings indicate that objective measures of voluntary cough may be useful to screen stroke patients for the need for further evaluation of swallow function by a speech language pathologist.1 Therefore, the higher sensitivity and specificity of objective analysis of voluntary cough when compared to a traditional bedside assessment (including evaluation of reflexive cough) for the identification of aspiration risk is clinically important. The necessary equipment needed for voluntary cough testing is readily available and easy to use. The test can be completed quickly and is safe (ie, it does not require patients to risk aspiration by attempting to swallow any material).