Objectives: This study examined the reliability and validity of geriatricians' assessments of pain in cognitively impaired nursing home residents.
Design: Cross-sectional analysis.
Setting: A large suburban nursing home.
Participants: Seventy-nine nursing home residents participated in the study. Of these, 31 had mild/moderate cognitive impairment (average Mini-Mental State Examination (MMSE) = 16.04) and 48 were severely cognitively impaired (average MMSE = 1.91). More than 80% of the participants were female, and the average age was 87.
Measurements: Two geriatricians from outside the nursing home examined laboratory results, performed a physical examination, and completed a detailed assessment of pain. The personal geriatricians of 42 of the participants also completed the same assessment.
Results: Intergeriatrician agreement rates were statistically significant and moderate in magnitude. When examined by subgroup, the correlations were significant only for those with mild/moderate impairment. Some of the geriatricians' ratings of pain correlated significantly with residents' self-reports. All relationships were weaker in the severely cognitively impaired group. Ratings of greater pain were significantly correlated with higher cognitive functioning.
Conclusions: The results validate geriatricians' evaluations of pain during a medical examination for moderately impaired persons and question their ability to evaluate pain in the severely cognitively impaired. There is a need for increased awareness of pain in this population and a need for improved methodologies to identify it.