Unrelieved postoperative pain is still reported to be a rather common clinical problem which may be related to inadequate routines for pain assessment. Therefore, the aim of the study was to describe strategies used by experienced and less experienced nurses and physicians in their assessment of postoperative pain and to relate different approaches, clinical experience, and professional role to the accuracy of the pain ratings. Data collection was based on repeated interviews with nurses (n = 30) and physicians (n = 30) in connection with clinical pain assessments (n = 180) including VAS-scoring. Commonly used strategies in the pain assessment were: -how the patient looks, -what the patient says, -the manner of talking, and -past experience of similar circumstances. The mean VAS-score given by the patients (6.1 +/- 1.1) was significantly (P < 0.001) higher than that rated by the staff members (4.9 +/- 1.2). Nurses as well as physicians overestimated low and underestimated high levels of pain indicated by the patients. The accuracy of the ratings by nurses, especially by more experienced ones (> or = 10 years in nursing), was found to be less precise than that of physicians. The pain assessment of these very experienced nurses was characterized by a systematic underestimation. The present study emphasizes a need for definition of more precise strategies for clinical postoperative pain assessment which better take into consideration the pain experiences and needs of individual patients.