Measuring Pain in Aphasia: Validity and Reliability of the PACSLAC-D

Post-stroke pain in patients with an inability to communicate is not systematically assessed and therefore not sufficiently treated. This stresses the need to study pain assessment instruments that do not require good communication skills.To examine the validity and reliability of the Pain Assessment Checklist for Seniors with Limited Ability to Communicate - Dutch version (PACSLAC-D) in stroke patients with aphasia.Sixty stroke patients (mean age 79.3 years, standard deviation [SD] 8.0), of whom 27 had aphasia were observed during rest, activities of daily living (ADL), and physiotherapy using the Pain Assessment Checklist for Seniors with Limited Ability to Communicate - Dutch version (PACSLAC-D). The observations were repeated after two weeks. To examine convergent validity, correlations between the PACSLAC-D, self-report pain scales, and the clinical judgment of a health care professional (pain present yes/no) were used. To examine discriminative validity, differences in pain were investigated between rest and ADL, in patients who use pain medication and those who do not, and in patients with and without aphasia. Internal consistency and test-retest reliability were assessed to determine reliability.Convergent validity failed to meet the acceptable threshold during rest but was adequate during ADL and physiotherapy. Discriminative validity was only adequate during ADL. The internal consistency was 0.33 during rest, 0.71 during ADL, and 0.65 during physiotherapy. Test-retest reliability varied from poor during rest (intraclass correlation coefficient [ICC] = 0.07; 95% confidence interval [CI]: -0.40-0.51) to excellent during physiotherapy (ICC = 0.95; 95% CI: 0.83-0.98).The PACSLAC-D captures pain in patients with aphasia who are unable to self-report, during ADL and physiotherapy, but may be less accurate during rest.