Research On Pain Assessment Instuments For Cognitively Impaired Elderly

Objectives:â‘ To develop a pain assessment instrument for elderly patients with cognitive impairment which is clinically applicable and translate pain assessment in advances dementia (PAINAD)and the checklist of nonverbal pain indicators (CNPI) , test the reliability and validity of the three scales.â‘¡To compare the difference of pain intensity between elderly people with cognitive impairment and cognitive intact elderly people.Methods: Based on the literature review, related assessment tools and consultation with expert clinicians, the pain assessment scale for cognitively impaired elderly (PACIE) was developed. The pain assessment in advances dementia (PAINAD) and checklist of nonverbal pain indicators (CNPI) were translated. Five experts specialized in pain, anaesthesia and nursing were invited to evaluate the content validity of the three pain assessment instruments and made modify. The trained raters applied the three instruments and NRS to assess three groups that cognitively impaired elderly patients with pain, cognitively intact elderly patients with pain and cognitively impaired elderly patients without pain. Psychometric properties of the three pain assessment instruments were evaluated. And the pain intensity between elderly people with cognitive impairment and intact elderly people was compared.Results:1. The validity and reliability of PACIE Index of content validity (CVI) of PACIE was 0.978, and all experts consider the items of the scales were concise and easy to be understood. The principal components analysis and factor analysis demonstrated that two factors soluted in PACIE, that cumulative explained 69.008% of the total variability, the two factors were named pain objective factor and pain subjective factor. PACIE and NRS were significantly correlated( r=0.879, P<0.01), and it was also significantly correlated with PAINAD and CNPI( r=0.638, 0.712, P<0.01), which indicate that it has satisfactory criterion-related validity. It can detect pain and without pain patients that demonstrate the discriminate validity was fine. Adequate levels of inter-rater reliability of PACIE was achieved between two raters with the intraclass correlation coefficient as 0.954 (95% CI = 0.812~0.989). Satisfactory test-retest reliability was achieved with the correlation coefficient as 0.693(P<0.001). Correlation coefficients between items of PACIE were 0.189~0.687, items to total correlation coefficients were 0.438 ~ 0.807, Cronbach alpha was 0.851.2. The validity and reliability of PAINAD and CNPI CVI of PAINAD and CNPI were 0.930, 0.900. PAINAD with two factors soluted explained 75.012% of the total variability. CNPI with two factors soluted each both at movement and at rest, explained 65.578% and 64.219% of the total variability. The Correlation coefficients of PAINAD and CNPI with NRS were 0.791, 0.771(P<0.01). Inter-rater reliability of PAINAD and CNPI were 0.892, 0.927 (95% CI = 0.598~0.974, 0.715~0.983), test-retest reliability were0.550, 0.607(P<0.001). Correlation coefficients between items of the PAINAD were 0.097~0.646, items to total correlation coefficients were 0.576~0.820, Cronbach alpha was 0.815. Correlation coefficients between items of the CNPI were 0.065~0.652, items to total correlation coefficients were 0.480~0.757, Cronbach alpha was 0.626 at movement, 0.535 at rest and 0.759 with total.3. There was no significant difference in the pain scores between cognitive impairment and cognitively intact elderly patients (P>0.05), and the pain scores of cognitively impaired elderly patients without pain was significant lower than the other two groups (P<0.001) using any one scale. There were significant differences between cognitively impaired and cognitively intact elderly patiens with pain in pain behaviors scores in the items of language/voice and body gesture (P<0.05), there was no significant difference in other items (P>0.05).Conclusions:1. The pain assessment scale for cognitively impaired elderly (PACIE) has satisfactory validity and reliability, is suitable for assessing the pain indensity of cognitively impaired elderly.2. PAINAD and CNPI also have satisfactory validity and reliability for assessing the pain indensity of cognitively impaired elderly.3. There was no significant difference between cognitively impaired and cognitively intact elderly patients in pain intensity, but the pain behaviors between the two groups have difference, the result sould be testified in futuer.