Brief Pain Inventory: A proposal to extend its clinical application.

Background: This study presents an adaptation of the Brief Pain Inventory (BPI) extending its use in clinical/epidemiological contexts and the evaluation of the properties of BPI (short form) in a sample of Brazilian adults.

Methods: Part of item 1 of this instrument was removed because it prevented the participation of individuals with usual pain. In addition to the reference period of original response "last 24 hr," a new period "last pain experience" was proposed. Individuals responded about the presence/lack and onset of pain. Individuals who reported pain in the last 24 hr before the interview answered the BPI considering both reference periods. Confirmatory factor analysis was performed to check the fit of five theoretical BPI models.

Results: A total of 1,176 adults participated (79.0% women; 38.7 (SD = 10.8) years), 29.2% did not report pain in the last 24 hr, 33.6% reported pain <3 months and 37.2% pain ≥3 months. All theoretical BPI models presented adequate fit indices (GFI ≥ 0.9; RMSEA < 0.1; α ≥ 0.7) when both reference periods were used. In conclusion, the adaptations proposed can contribute to extend the use of BPI.

Conclusions: The reference period of responses and the theoretical model used must be chosen according to the needs of the researcher and/or physician.

Significance: This study presents evidence related to the validity of applying the Brief Pain Inventory (BPI) in adults with and without pain considering the present pain or memory of pain, enabling the clinician to collect additional information that may be relevant to the clinical management of pain.