A study of longitudinal data examining concomitance of pain and cognition in an elderly long-term care population

PURPOSE: To examine if a concomitant relationship exists between cognition and pain in an elderly population residing in long-term care. BACKGROUND/SIGNIFICANCE: Prior research has found that cognitive load mediates interpretation of a stimulus. In the presence of decreased cognitive capacity as with dementia, the relationship between cognition and increasing pain is unknown in the elderly. PATIENTS AND METHODS: Longitudinal cohort design. Data collected from the Minimum Data Set-Resident Assessment Instrument (MDS-RAI) from the 2001–2003 annual assessments of nursing home residents. A covariance model was used to evaluate the relationship between cognition and pain at three intervals. RESULTS: The sample included 56,494 subjects from nursing homes across the United States, with an average age of 83 ± 8.2 years. Analysis of variance scores (ANOVAs) indicated a significant effect (P < 0.01) for pain and cognition, with protected t test revealing scores decreasing significantly with these two measures. Relative stability was found for pain and cognition over time. Greater stability was found in the cognitive measure than the pain measure. Cross-legged effects observed between cognition and pain measures were inconsistent. A concomitant relationship was not found between cognition and pain. Even though the relationship was significant at the 0.01 level, the correlations were low (r ≤ 0.08), indicating a weak association between cognition and pain. CONCLUSION: Understanding the concomitance of pain and cognition aids in defining additional frameworks to extend models to include secondary needs, contextual factors, and resident outcomes. Cognitive decline, as with organic brain diseases, is progressive. Pain is a symptom that can be treated and reduced to improve resident quality of life. However, cognition can be used to determine the most appropriate method to assess pain in the elderly, thereby improving accuracy of pain detection in this population.