'Painting my pain': the use of pain drawings to assess multisite pain in women with primary dysmenorrhea.

Background: To verify the use of pain drawing to assess multisite pain in with primary dysmenorrhea (PD) and to assess its divergent validity, test-retest reliability, intra- and inter-rater reliability and measurement errors.

Methods: Cross-sectional study. Adult women with self-reported PD three months prior to the study. Women answered the Numerical Rating Scale (NRS) and the pain drawing during two consecutive menstruations. The pain drawings were digitalized and assessed for the calculation of total pain area (%). Intra- and inter-rater reliability and the test-retest reliability between the first and the second menstruations were assessed with the intraclass correlation coefficient (ICC). Measurement errors were calculated with the standard error of measurement (SEM), smallest detectable change (SDC) and the Bland-Altman plot. Spearman correlation (rho) was used to check the correlation between the total pain area and pain intensity of the two menstruations.

Results: Fifty-six women (24.1 ± 3.1 years old) participated of the study. Their average pain was 6.2 points and they presented pain in the abdomen (100%), low back (78.6%), head (55.4%) and lower limbs (50%). All reliability measures were considered excellent (ICC > 0.75) for the total pain area; test-retest SEM and SDC were 5.7% and 15.7%, respectively. Inter-rater SEM and SDC were 8% and 22.1%, respectively. Correlation between total pain area and pain intensity was moderate in the first (rho = 0.30; p = 0.021) and in the second menstruations (rho = 0.40; p = 0.002).

Conclusion: Women with PD presented multisite pain, which could be assessed with the pain drawing, considered a reliable measurement.