Background: Physicians' ratings about their patients' pain prove to be invalid compared to patients' ratings. This is especially true if pain rating acts as an indicator for therapy outcome. The aim of this study was to compare physicians' and patients' ratings of pain relief following acupuncture and to identify correlations between patient characteristics and potential miscalibrations.
Patients and methods: In a cross-sectional study 291 pain patients with gonarthrosis or chronic low back pain and their attending physicians were asked to give their rating of patients' pain relief following acupuncture. Patients were interviewed by telephone, and doctors responded to questionnaires.
Results: The proportion of false-positive physicians' ratings was 81% referring to patients without self-reported benefit from acupuncture. Just every fifth patient without pain relief was correctly classified by his physician. There was no correlation between patients' characteristics and false-positive ratings of physicians.
Conclusions: Evaluation of treatment in daily medical routine should be primarily based on information provided by patients.