Premium
P3‐039: Performance of the long and short Brazilian versions of the IQCODE as screening for dementia in low‐educated elderly
Author(s) -
Perroco Tı́bor R.,
Bustamante Sonia E.Z.,
Alegria Renne P.,
Moreno Maria P.Q.,
Hototian Sérgio R.,
Lopes Marcos A.,
Azevedo Dionı́sio,
Litvoc Júlio,
Jacob Wilson,
Bottino Cássio M.C.
Publication year - 2008
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1016/j.jalz.2008.05.1602
Subject(s) - dementia , cronbach's alpha , medicine , portuguese , logistic regression , receiver operating characteristic , gerontology , socioeconomic status , mini–mental state examination , psychometrics , clinical psychology , disease , population , environmental health , linguistics , philosophy
this study to determine the clinimetrical properties (construct validity and reliability) of the NPI Q and NPI D in a group of Mexican demented patients. Methods: We conducted a transversal, analytical and validation essay. We selected 100 dyads, with dementia diagnosis (of any type). The dyads were interviewed with neuropsychological tools (Mini Mental State Examination, Geriatric Depression Scale, Screening Caregiver Burden, Sleep Disorders Inventory NPI Q, NPI D), sociodemographic questionnaire (age, sex, caregiver relationship, work status, scholarship, etc), a thorough record of comorbidities, medications, and an exploration to complete the Mini Nutritional Assessment. NPI Q s total score was correlated with MMSE, GDS, NPI D and SCB; and individual domains were also evaluated (depression, psychotic symptoms, nutrition and sleep). The NPI D s total score was correlated with SCB and NPI Q. Six patients were selected for a test-retest, to evaluate reliability. Spearman s one tailed correlation, intraclass correlation coefficient (ICC) and Cronbach s alpha, were performed for statistical analysis. Results: Eighty-seven dyads were interviewed, general characteristics are shown in Table 1. The NPI-Q had a significant correlation with MMSE of 0,190 (p 0.001), NPI-D of r 0,887 (p 0,001), with GDS of r 0,385 (p 0.0001). The NPI-Q with a testretest of r 0,829 (p 0.021), ICC 0.824 and internal consistency of 0.778, the individual domain results are shown in Table 2. The NPI D had a significant correlation of r 0.555 with the SCB; test-retest r 0.986, ICC 0.857 and 0.779. Conclusions: Our NPI-Q and NPI-D versions are valid in the evaluation of Mexican patients with dementia. Our results are similar to those reported previously, but add the significant evaluation in the individual domains of sleep and oral intake.