Purpose: To validate an adaptation of the Behavioral Pain Scale (BPS) for its use in non-intubated intensive care unit (ICU) patients unable to self-report their pain because of the occurrence of delirium. The "vocalization" domain was inserted to construct the BPS-non intubated (BPS-NI) scale, ranging from 3 (no pain) to 12 (most pain).
Design: Prospective psychometric study in a medical-surgical ICU.
Methods: The same physician and one bedside nurse rated pain in non-intubated patients unable to self-report their pain during four conditions: before and after a catheter dressing change (non-nociceptive procedure) and before and after turning the patient (nociceptive procedure). Delirium was assessed by the Confusion Assessment Method for the ICU (CAM-ICU).
Results: A total of 120 paired evaluations were performed in 30 consecutive adult patients, 84% with delirium (CAM-ICU positive). BPS-NI scores were higher during painful procedures than at rest [6.0 (5.0-8.0) vs. 3.0 (3.0-3.8); P < 0.001], while no changes in BPS-NI scores were found during non-nociceptive procedures (discriminative validity). The BPS-NI had good internal consistency (standardized Cronbach alpha = 0.79), and each domain reflected the pain expression factor in a balanced way (coefficients between 0.57 and 0.59). The BPS-NI had a good inter-rater reliability (weighted kappa coefficient = 0.89 for the four conditions and 0.82 during nociceptive procedures) and a good responsiveness, with an effect size ranging from 1.5 to 3.6.
Conclusions: Pain during procedures is perceived even in non-intubated ICU patients with delirium. In those patients, pain level can be assessed with the BPS-NI scale since this instrument exhibited good psychometric properties. Electronic supplementary material The online version of this article (doi:10.1007/s00134-009-1590-5) contains supplementary material, which is available to authorized users.