Comparisons of patient and physician assessment of pain-related domains in cancer pain classification: results from a large international multicenter study

Abstract

The aim of the present study is to compare physician clinical assessment with patient-rated evaluations in the classification of cancer pain patients into groups with different pain levels, according to the presence of incident/breakthrough pain, neuropathic pain, and psychological distress. Average pain in the previous 24 hours was used as the dependent variable in multivariate linear regression models, and incident/breakthrough pain, neuropathic pain, and psychological distress were tested as regressors; in the assessment of regressors, physicians used the Edmonton Classification System for Cancer Pain, whereas patients used structured self-assessment questionnaires. The amount of variability in pain intensity scores explained by the 2 sets of regressors, physician and patient rated, was compared using R(2) values. When tested in 2 separate models, patient ratings explained 20.3% of variability (95% confidence interval [CI] = 15.2-25.3%), whereas physician ratings explained 6.1% (95% CI = 2.2-9.8%). The higher discriminative capability of patient ratings was still maintained when both regressor sets were introduced in the same model, with R(2) indices of 17.6% (95% CI = 13.0-22.2%) for patient ratings vs 3.4% (95% CI = .9-5.9%) for physician ratings. Patients' self-assessment of subjective symptoms should be integrated in future cancer pain classification systems.

Perspective: Our results indicate that patient-structured assessment of incident/breakthrough pain, neuropathic pain, and psychological distress significantly contributes to the discrimination of cancer patients with different pain levels. The integration of patient self-assessment tools with more objective clinician assessments can improve the classification of cancer pain.

Keywords: Cancer pain; pain assessment; pain classification; patient-reported outcomes, palliative care.