Speech Breathing in Patients with Lung Disease

Abstract

Lung volumes and breathing patterns used during speech differ from those of quiet respiration and may be expected to vary with different types of lung disease. To test this possibility, 41 patients with asthma, emphysema, or sarcoidosis and 16 healthy subjects completed a speech protocol. Volumes, times, and flow rates were recorded during conversation and during a counting task. A total of 16 measured variables were derived for each breath and analyzed statistically. Alterations in speech breathing were disease and task specific. Discriminant function analysis applied to data from either speech task could correctly classify subjects with more than 50% accuracy, showing that different patterns were significantly disease specific. Compared with healthy subjects during conversation, all patients averaged a more rapid respiratory rate and increased the proportion of time spent on inspiration (Ti/Ttot). During counting, patient groups showed a variety of patterns, most commonly subordinating metabolic need to communication drive and sounding more breathless to observers. Regression analysis was used to determine how strongly changes in measured speech variables related to degree of physiologic impairment. The effect of severity of disease on speech production is distinguishable from the effect of the diagnostic category.