Pain Assessment in Nursing Home Residents With Dementia: Psychometric Properties and Clinical Utility of the CNA Pain Assessment Tool (CPAT)

Abstract

To examine the psychometric properties and clinical utility of the CPAT, an instrument to assess pain in nursing home residents with dementia. Instrument development and testing. Three regional skilled nursing facilities. One hundred forty-five residents of 3 skilled nursing facilities. The inter-rater reliability, test-retest reliability, construct validity, and criterion validity of the CPAT was measured after initial CNA training. Measurement of inter-rater reliability and test-retest reliability was repeated after modified CNA training. Internal consistency (Cronbach's alpha) was calculated for all reliability measures. The clinical utility/feasibility of the CPAT was measured by means of a practicality survey. The CPAT was found to have acceptable levels of both interrater reliability (ICC=0.71) and test-retest reliability (ICC=0.67). Construct validity as measured by a paired t test was statistically significant (P=.043). Criterion validity as measured by Spearman's rank correlation coefficient was also statistically significant (P=.048). Internal consistency was acceptable for all measures as calculated by Cronbach's alpha, which ranged from 0.72 to 0.84. As determined by a practicality survey, the CPAT was shown to be a clinically useful and feasible instrument. This study provides evidence that the CPAT is a reliable and valid pain assessment instrument when used in nursing home residents with dementia. It has also been shown to be a tool with suitable clinical utility and feasibility. Further study is warranted to ascertain if its use will lead to improved resident function and quality of life and whether the results of this study are reproducible in other dementia populations. The CPAT's ability to quantify pain and measure treatment response has not been determined.