Capacity of the Edmonton Symptom Assessment System and the Canadian Problem Checklist to screen clinical insomnia in cancer patients.

Abstract

Insomnia is highly prevalent in cancer patients. Efficacious treatments exist for this condition but, for these interventions to be offered, an effective screening needs to be carried out beforehand. The pan-Canadian practice guidelines on sleep disturbances provide recommendations on how to use the Edmonton Symptom Assessment System (ESAS) and the Canadian Problem Checklist (CPC) to screen for sleep difficulties in cancer. However, empirical evidence to support these recommendations is lacking. The goal of this study was to assess the capacity of the ESAS (drowsiness and "other" items) and the CPC (sleep item) to screen for clinical insomnia in cancer patients. As part of routine care, 615 patients with various cancer types completed the ESAS, the CPC, and the Insomnia Severity Index, used as the standard for establishing the presence of clinical insomnia. None of the criteria provided an effective screening when used alone. No patient used the ESAS-other item to report sleep difficulties. The sensitivity and specificity rates of the CPC-sleep item alone were 60.4 and 89.6%, respectively. A score ≥2 on the ESAS-drowsiness item had a sensitivity of 61.5% and a specificity of 75.4%. When used in combination, the best option was scoring positively on the CPC-sleep item OR a score ≥2 on the ESAS-drowsiness item (sensitivity 84.2%; specificity 69.7%). When used alone, the CPC-sleep and the ESAS-drowsiness items yielded insufficient sensitivity rates for a first screening, but when used in combination, they provided a good balance between sensitivity and specificity.