Association of Descriptors of Breathlessness With Diagnosis and Self-Reported Severity of Breathlessness in Patients With Advanced Chronic Obstructive Pulmonary Disease or Cancer

Sarah Chowienczyk, Shagayegh Javadzadeh, Sara Booth, Morag Farquhar

Verbal descriptors are important in understanding patients' experience of breathlessness. To examine the association between selection of breathlessness descriptors, diagnosis, self-reported severity of breathlessness and self-reported distress due to breathlessness. We studied 132 patients grouped according to their diagnosis advanced COPD (n=69) or advanced cancer (n=63), self-reported severity of breathlessness: mild breathlessness (Numerical Rating Scale (NRS) ≤3, n=53), moderate breathlessness (4≤NRS≥6, n=59) or severe breathlessness (NRS ≥7, n=20), and distress due to breathlessness: mild distress (NRS ≤3, n=31), moderate distress (4≤NRS≥6, n= 44) or severe distress (NRS ≥7, n=57). Patients selected three breathlessness descriptors. The relationship between descriptors selected and patient groups was evaluated by cluster analysis. Different combinations of clusters were associated with each diagnostic group; the cluster chest tightness was associated with cancer patients. The association of clusters with patient groups differed depending on their severity of breathlessness and their distress due to breathlessness. The air hunger cluster was associated with patients with moderate or severe breathlessness, the chest tightness cluster was associated with patients with mild breathlessness. The air hunger cluster was associated with patients with severe distress due to breathlessness. The relationship between clusters and diagnosis is not robust enough to use the descriptors to identify the primary cause of breathlessness. Further work exploring how use of breathlessness descriptors reflects the severity of breathlessness and distress due to breathlessness could enable the descriptors to evaluate patient status and target interventions.