The aim of the present prospective study was to evaluate pain treatment during the first postoperative 24 h for 40 patients (age over 18) undergoing tonsillectomy. Patients were divided into two groups: group A (n = 20) received analgesics on demand and group B (n = 20) on a regular basis. Basic pain treatment consisted of paracetamol 750 mg x 6 and diclofenac 50 mg x 3. Pain measurement was performed using a visual analogue scale (VAS): a 10 cm line with 0 cm equalling no pain and 10 cm equalling the worst pain ever felt. The following parameters were studied: VAS values, the need for rescue analgesics, intra- and postoperative bleeding, nausea and vomiting, postoperative food intake and hospital time. Only 4 of 20 (20%) patients in group B needed rescue analgesics in the postoperative ward compared with 15 of 20 (75%) in group A (p < 0.01, chi2 test). In group B, 13 of 20 (65%) patients could eat solid food before they were discharged from the ward, compared with 7 of 19 (37%) monitored patients in group A (p < 0.01, chi2 test). The observed VAS values were generally rather low in both groups. The mean value for all observed VAS values was less than 4 in both study groups. However, no significant difference in VAS values was observed between the two study groups. Our results suggest that regularly given postoperative pain treatment after tonsillectomy, starting intraoperatively with paracetamol and diclofenac, has significant advantages compared with a regimen in which patients receive analgesics only on demand.