Differences in Staff-Assessed Pain Behaviors among Newly Admitted Nursing Home Residents by Level of Cognitive Impairment.

OBJECTIVE Pain is common among nursing home residents with cognitive impairment and dementia. Pain is often underdiagnosed and undertreated, which may lead to adverse health outcomes. Nonverbal behaviors are valid indicators of pain, but the extent to which these behavioral expressions vary across levels of cognitive impairment is unknown. This study sought to examine differences in the prevalence of pain behaviors among nursing home residents with varying levels of cognitive impairment. METHODS The Minimum Data Set, version 3.0, was used to identify newly admitted nursing home residents with staff-assessed pain (2010-2016, n = 1,036,806). Staff-assessed pain behaviors included nonverbal sounds, vocal complaints, facial expressions, and protective body movements or postures over a 5-day look-back period for residents unable or unwilling to self-report pain. The Cognitive Function Scale was used to categorize residents as having no/mild, moderate, or severe cognitive impairment. Modified Poisson models provided adjusted prevalence ratios (aPR) and 95% CIs. RESULTS Compared to residents with no/mild cognitive impairments (any pain: 48.1%), residents with moderate cognitive impairment (any pain: 42.4%; aPR: 0.94 [95% CI 0.93-0.95]) and severe cognitive impairment (any pain: 38.4%; aPR: 0.86 [95% CI 0.85-0.88]) were less likely to have any pain behavior documented. Vocal pain behaviors were common (43.5% in residents with no/mild cognitive impairment), but less so in those with severe cognitive impairment (20.1%). Documentation of facial expressions and nonverbal pain behaviors was more frequent for residents with moderate and severe cognitive impairment than those with no/mild cognitive impairment. CONCLUSIONS The prevalence of behaviors indicative of pain differs by level of cognitive impairment. Pain evaluation and management plays an important role in treatment and care outcomes. Future work should examine how practitioners' perceptions of pain behaviors influence their ratings of pain intensity and treatment choices.