The main purpose of this study was to describe how patients treated in an intensive care unit (ICU) perceive their sleep and to compare patients' and nurses' perceptions of the patients' sleep. The study also determined the percentage of patients in the ICU who were able to fill in the Richard Campell Sleep Questionnaire (RCSQ). This instrument consists of six items and utilises a visual analogue scale (VAS). The results of five of the RCSQ questions are used to calculate a total sleep score, ranging between 0 and 100 (0=the worst possible sleep, 100=the best sleep).Approximately half of the patients were able to answer the RCSQ (n=31). The patients' rating of their sleep varied widely (total sleep score: range 0-97, mean 45.5). Patients who had received hypnotics or sedatives during the night (n=12) had a significantly lower total sleep score (mean=31.6) than the rest of the patients (mean 54.3; P=0.037). On comparing the patients' and the nurses' perceptions of the patients' sleep, no significant difference between the groups was seen. This indicates that nurses can use the RCSQ to assess the sleep of patients who are unable to report their sleep themselves.