Abstract
ABSTRACT
Objective
To translate, cross‐culturally adapt, and validate the short version of the Patterns of Activity Measure‐Pain (POAM‐P) in Brazilian individuals with chronic pain.
Methods
This was a questionnaire validation study. Brazilian participants reporting pain for more than 3 months were included. The measurement properties assessed were structural validity, construct validity, reliability, internal consistency, and agreement.
Results
The study included 125 individuals with chronic pain for the analysis of structural and construct validity. The internal structure of the reduced version of the POAM‐P, comprising 3 domains and 15 items, was considered adequate (comparative fit index = 0.98; Tucker‐Lewis index = 0.97; root mean square error of approximation = 0.07; standardized root mean square residual = 0.06; and chi‐square/degrees of freedom = 1.55). Correlations ranging from 0.30 to 0.50 were observed between the avoidance and pacing domains and kinesiophobia, catastrophizing, and disability. The overdoing domain showed no significant correlations with external measures, associating negatively only with the avoidance and pacing domains. Internal consistency was assessed in the total sample (
n
= 125) using composite reliability (CR), with values ranging from 0.76 to 0.93 across the three POAM‐P domains. Test–retest reliability was assessed in a subsample of 33 individuals and proved satisfactory, with intraclass correlation coefficients ≥ 0.79. The test–retest interval was planned to range from 7 to 14 days, with an observed mean of 8 days (7–13). Agreement, assessed via Bland‐Altman plots in the same subsample (
n
= 33), was considered acceptable. No ceiling or floor effects were identified.
Conclusion
The Brazilian short version of the POAM‐P demonstrated adequate structural validity, internal consistency, and test–retest reliability. Nevertheless, further studies with larger retest samples are warranted to confirm the stability of these findings. Regarding construct validity, the avoidance and pacing domains showed correlations with external measures as expected, whereas the overdoing domain was correlated only with the other domains of the instrument.