Comparing the Psychometric Properties of the Checklist of Nonverbal Pain Behaviors (CNPI) and the Pain Assessment in Advanced Dementia (PAIN-AD) Instruments

Document Type


Publication Date



pain, pain Measurement, dementia, assessment, cognitive impairment, nursing homes

Digital Object Identifier (DOI)



Objective. To examine and compare the psychometric properties of two common observational pain assessment tools used in persons with dementia.

Design. In a cross-sectional descriptive study nursing home (NH) residents were videotaped at rest and during a structured movement procedure. Following one training session and one practice session, two trained graduate nursing research assistants independently scored the tapes using the two pain observation tools.

Setting. Fourteen NHs in Western Washington State participating in a randomized controlled trial of an intervention to enhance pain assessment and management.

Participants. Sixty participants with moderate to severe pain were identified by nursing staff or chosen based on the pain items from the most recent Minimum Data Set assessment.

Measures. Checklist of Nonverbal Pain Indicators (CNPI) and the Pain Assessment in Advanced Dementia (PAINAD), demographic and pain-related data (Minimum Data Set), nursing assistant reports of participants' usual pain intensity, and Pittsburgh Agitation Scale.

Results. Internal consistency for both tools was good except for the CNPI at rest for one rater. Inter-rater reliability for pain presence was fair (K = 0.25 for CNPI with movement; K = 0.31 for PAINAD at rest) to moderate (K = 0.43 for CNPI at rest; K = 0.54 for PAINAD with movement). There were significant differences in mean CNPI and PAINAD scores at rest and during movement, providing support for construct validity. However, both tools demonstrated marked floor effects, particularly when participants were at rest.

Conclusions. Despite earlier studies supporting the reliability and validity of the CNPI and the PAINAD, findings from the current study indicate that these measures warrant further study with clinical users, should be used cautiously both in research and clinical settings and only as part of a comprehensive approach to pain assessment.

Was this content written or created while at USF?


Citation / Publisher Attribution

Pain Medicine, v. 11, issue 3, p. 395-404