Thirty-six patients underwent assessment of behavioural breathlessness which included monitoring of breathing patterns and end tidal CO2 concentration and completion of questionnaires relating to hyperventilation (HV), anxiety and depression. Twenty-two patients had a positive assessment and underwent breathing retraining. Assessments were repeated immediately after re-training and 2 months later. Ten of the patients (Group A) had behavioural breathlessness either as the primary problem or secondary to an established clinical condition, and twelve (Group B) in association with chronic fatigue. Before re-training, resting end-tidal PCO2 was significantly lower in Group A than Group B (p < 0.05), but there was no significant difference in mean scores for HV-related symptoms, anxiety or depression. Following breathing retraining, both groups showed improvements in breathing patterns, end tidal CO2 levels and scores for HV-related symptoms which were sustained. In Group A the mean score for anxiety decreased (p < 0.01) and the score for depression was significantly lower than in Group B (p < 0.05). Although mean scores for anxiety and depression in Group B did not change significantly, some individuals in the group did show sustained improvement. There was no improvement in symptoms associated with chronic fatigue in Group B. In behavioural breathlessness, breathing retraining is of benefit, not only in restoring more normal patterns of breathing but also in reducing anxiety, particularly in patients without the complication of chronic fatigue.