Improved diagnostic tests that can accurately identify subjects at high risk for aspiration pneumonia (AP) are needed. One measure of this accuracy is the false-negative rate (FNR), which determines the failure of a test to identify a group at high risk. This study compares FNRs for AP among dysphagic stroke patients for two prognostic techniques: modified barium swallow (MBS) alone and MBS combined with laryngopharyngeal sensory discrimination testing (MBS + LPSDT). MBS and LPSDT were performed within 4 weeks of stroke in 20 subjects who were prospectively followed for at least 2 years to identify the frequency of AP. MBS identified 10 patients as not at risk based on the finding of no aspiration on initial MBS; four of these patients developed AP (FNR = 40%). MBS + LPSDT identified five patients as not at risk based on the findings of neither aspiration nor bilateral sensory deficits; none of these patients developed AP (FNR = 0%). The combination of MBS criterion (aspiration) with the LPSDT criterion (bilateral sensory deficits) improves prognostication of outcome in dysphagic stroke patients by identifying a subgroup at high risk for developing AP (nonaspirators with bilateral deficits).